MDDM Master’s Degree Program in

Data-Driven Marketing

Creativity and Problem-Solving in Advertising:

The Impact of ChatGPT within the Design Thinking Framework

Ana Catarina Benvindo Casquinha

Master Thesis
presented as a partial requirement for obtaining a Master’s Degree in Data-Driven Marketing

NOVA Information Management School

Instituto Superior de Estatística e Gestão de Informação
Universidade Nova de Lisboa
 NOVA Information Management School
Instituto Superior de Estatística e Gestão de Informação
Universidade Nova de Lisboa

Creativity and Problem-Solving in Advertising:

The Impact of ChatGPT within the Design Thinking Framework

by
Ana Catarina Benvindo Casquinha

Master Thesis presented as partial requirement for obtaining the Master’s degree in Data-
Driven Marketing, with a specialization in Marketing Intelligence.

Supervised by
Prof. Nadine Côrte-Real, PhD in Information Management, NOVA Information Management
School

February, 2024
 STATEMENT OF INTEGRITY
I hereby declare having conducted this academic work with integrity. I confirm that I have not
used plagiarism or any form of undue use of information or falsification of results along the
process leading to its elaboration. I further declare that I have fully acknowledged the Rules
of Conduct and Code of Honor from the NOVA Information Management School.

Lisbon, 22/02/2024

i
 DEDICATION
One of my main goals was to achieve a Master's degree, a task I initially believed would be
straightforward for me. Historically, I've excelled academically, never faced significant issues
with my study habits, and have always been characterized by my strong focus and dedication.
However, the past two and a half years have introduced a series of challenges, placing
obstacles in my path that I've managed to navigate with considerable effort. Striking a balance
between a demanding job, maintaining a social life, pursuing physical fitness, engaging in
more reading, and dedicating time to my family—all while returning to my studies—proved
to be a challenging endeavor. There were moments when my motivation waned, but I
remained steadfast in my commitment to this goal.

This journey would not have been possible without the unwavering support of the
extraordinary individuals in my life. I am deeply grateful to my parents, grandparents,
boyfriend, his family, and my friends for their constant encouragement and positive outlook.
Their belief in my potential was a source of strength, helping me to persevere through the
most challenging periods.

I must also express my gratitude towards my dog, whose companionship and emotional
support were invaluable to me during this time.

Thank you all so much.

ii
 ACKNOWLEDGMENTS
In the journey toward completing this project, I have been fortunate to receive guidance,
support, and assistance from a number of esteemed individuals and groups, to whom I owe a
debt of gratitude.
I extend my sincerest thanks to my advisor, Prof. Nadine Côrte-Real, whose expertise,
patience, and insight have been essential. I am also profoundly grateful to my workplace for
fostering an environment that values learning and development. The support and flexibility
offered by the organization have been critical in allowing me to balance my professional
responsibilities with my academic pursuits. My colleagues also deserve special mention for
their assistance and encouragement. Lastly, I wish to acknowledge all the professionals who
generously offered their time and expertise to assist me with this project. Their prompt and
willing support played a significant role.

To each of you, I express my deepest appreciation for your invaluable contribution to my

journey.

iii
 ABSTRACT
In the context of the impact of creativity in advertising, this thesis focuses on the incorporation
of Generative Artificial Intelligence, in particular OpenAI's ChatGPT tool, within the Design
Thinking framework. Through a qualitative analysis, including interviews with advertising
professionals, the study explores the synergies between generative AI technologies and
creative processes and problem-solving. It identifies key benefits, such as the tool's support
for idea generation, workflow efficiency and time optimization, while acknowledging
significant challenges, such as ensuring creative integrity and managing the limitations of AI
technology. The research highlights the key role of human supervision in harnessing ChatGPT's
potential to complement human creativity, proposing an updated Design Thinking framework
that integrates GenAI tools to promote innovation and efficiency in creative practices. This
comprehensive analysis not only sheds light on the operational and creative advantages of
ChatGPT but also addresses the concerns and challenges associated with integrating AI into
the creative industry.
The findings contribute to the ongoing update of traditional processes on the intersection of
technology and creativity, offering valuable insights for advertising professionals aiming to
harness AI's potential while maintaining the human essence.

KEYWORDS

ChatGPT; Design Thinking; Advertising; Creative Problem-Solving; Generative AI; Workflow

Optimization; Human-AI Collaboration; Artificial Intelligence.

Sustainable Development Goals (SDG):

iv
 TABLE OF CONTENTS

Statement of Integrity ........................................................................................................ i

Dedication ......................................................................................................................... ii
Acknowledgments ............................................................................................................ iii
Abstract ............................................................................................................................ iv
Table of Contents .............................................................................................................. v
List of Figures................................................................................................................... vii
List of Tables ................................................................................................................... viii
List of Abbreviations and Acronyms................................................................................. ix
1. Introduction .................................................................................................................. 1
2. Literature review .......................................................................................................... 3
2.1. Design Thinking Overview ..................................................................................... 3
2.1.1. A Human-Centric Approach............................................................................ 5
2.1.2. The Design Thinking Stages ............................................................................ 5
2.1.3. Design Thinking applications in different areas ............................................. 8
2.2. Generative Artificial Intelligence ........................................................................... 9
2.2.1. ChatGPT ........................................................................................................ 10
2.2.2. ChatGPT’s Limitations .................................................................................. 11
2.2.3. ChatGPT’s applications in different areas .................................................... 12
2.2.4. ChatGPT integration in Design Thinking for the advertising industry.......... 12
3. Methodology .............................................................................................................. 14
3.1. Research Motivation ........................................................................................... 14
3.2. Research Methodology ....................................................................................... 15
3.3. Participant Selection............................................................................................ 15
3.4. Company Selection .............................................................................................. 16
3.5. Investigation’s Output ......................................................................................... 16
4. Empirical Study ........................................................................................................... 17
4.1. Study Design ........................................................................................................ 17
4.1.1. Research Objectives ..................................................................................... 17
4.1.2. Premises ....................................................................................................... 17
4.2. Sampling Method ................................................................................................ 18
4.2.1. Intentional and Stratified Sampling ............................................................. 18
4.2.2. Participant Sample ....................................................................................... 19
4.3. Data Collection .................................................................................................... 20

v
 4.3.1. Initial Contact and Invitation Process........................................................... 20
4.3.2. Benefits and Limitations of Email-Based Data Collection ............................ 20
4.3.3. Interview Questions ..................................................................................... 21
4.4. Data Analysis ....................................................................................................... 21
4.4.1. Qualitative Content Analysis with TQF......................................................... 22
4.4.2. Coding process ............................................................................................. 23
4.4.3. CATEGORIZATION PROCESS .................................................................................. 25
4.4.4. LIMITATIONS OF THE DATA ANALYSIS PROCESS ....................................................... 26
5. Results and Discussion................................................................................................ 28
5.1. Overview.............................................................................................................. 28
5.2. Results ................................................................................................................. 29
5.2.1. Enhancement of Creative and Problem-Solving .......................................... 29
5.2.2. Workflow Optimization ................................................................................ 30
5.2.3. Challenges and Limitations of AI Integration ............................................... 30
5.2.4. Integrating ChatGPT into Design Thinking Stages ........................................ 32
5.3. Discussion ............................................................................................................ 33
5.3.1. Overview of Key Findings ............................................................................. 33
5.3.2. The integration of ChatGPT in the advertising industry’s workflows .......... 34
5.3.3. The impact of ChatGPT specifically on problem-solving creativity .............. 35
5.3.4. The Integration of ChatGPT into the Design Thinking Framework .............. 35
5.3.5. The Challenges of ChatGPT in the Design Thinking Process ........................ 37
5.4. AI-Driven Design Thinking Framework ................................................................ 37
5.4.1. Introduction to AI-Driven Design Thinking Framework ............................... 37
5.4.2. The foundation of the AI-Driven Design Thinking Framework .................... 39
5.4.2.1. Framework Adaptability .................................................................... 40
5.4.2.2. Responsible AI Use ............................................................................. 40
5.4.3. Framework Limitations................................................................................. 41
6. Conclusions and Future Works ................................................................................... 43
Bibliographical References .............................................................................................. 45
Appendix A: Questionnaire Invitation ............................................................................. 50
Appendix B: Questionnaires ............................................................................................ 51
Appendix C: Codebook .................................................................................................... 61

vi
 LIST OF FIGURES
Figure 1- Interest in “Design Thinking” over time (last 10 years) by Google Trends ................. 3
Figure 2 - The Five Phases of the IDEO/d.school Design Thinking Process ................................ 6
Figure 3- Interest in “ChatGPT” over time (last 5 years) by Google Trends ............................ 10
Figure 4 - Stratified Sampling by Scribbr (Thomas, 2023) ........................................................ 18
Figure 5 – Phases in qualitative content analysis in TQF (Roller & Lavrakas, 2015) ................ 22
Figure 6 - Categorization of codes............................................................................................ 26
Figure 7 - Research Key Findings .............................................................................................. 34
Figure 8 - AI-Driven Design Thinking Framework ..................................................................... 41

vii
 LIST OF TABLES
Table 1- Interview’s Sample ..................................................................................................... 19
Table 2 - Interview Questions .................................................................................................. 21
Table 3 - AI-Driven Design Thinking Framework Stages .......................................................... 39

viii
 LIST OF ABBREVIATIONS AND ACRONYMS

AI Artificial Intelligence

DT Design Thinking

GAI Generative Artificial Intelligence

GAN Generative Adversarial Network

GEN-AI Generative Artificial Intelligence

GPT Generative Pre-training Transformer

LLM Large Language Models

LR Literature Review

NLP Natural Language Processing

P1 Premise one

P2 Premise two

TQF Total Quality Framework

ix
 1. INTRODUCTION

In this ever-changing world, teeming with cutting-edge technologies, initial research

suggests that Artificial Intelligence (AI) has the potential to bring immense value and
serve as a formidable ally for revolutionizing business (Kuczmarski, 2003). Beyond its
widely recognized applications in our daily routines – think robots, chatbots, and self-
driving vehicles – artificial intelligence has taken strides toward aiding us in our day-to-
day tasks (Zhang & Lu, 2021). As we navigate through an era of constant transformation
and advancement, it is crucial to acknowledge the role that AI can play in propelling
businesses forward. With its ability to analyze vast amounts of data at lightning speed
and make informed decisions based on patterns and trends unseen by human eyes alone
- there's no denying the impact it could have on driving innovation (Bughin et al., 2017).

In November of 2022, OpenAI revolutionized the market and launched one of the best-
known AI tools for generating content – ChatGPT-3. This type of AI is capable of
reproducing human language creating written content of astonishing quality. ChatGPT,
a sophisticated and globally recognized linguistic model, has captivated minds around
the world with its ability to generate human-like text (OpenAI, 2022). Its emergence on
the global stage prompts us to reassess the boundaries of AI's capabilities, venturing
into realms that were once thought to be the exclusive province of human intellect
(Chui, 2023).
However, as attention on AI's capabilities intensifies, exploration of its application in
some traditional humanized processes, such as Design Thinking methodology, remains
markedly scarce, creating a research gap that is intriguing and worth investigating
(Verganti et al., 2020). Despite a growing body of literature on the pragmatic
applications of Large Language Models (LLM) in a myriad of sectors, there is a gap in the
application of ChatGPT in problem-solving and creative thinking methods.

Nowadays, businesses are highly dependent on constant evolution, updating of

methods, and innovation in the services or products they offer to remain competitive.

1
For this very reason, various methods have been developed to help create thoughts and
methods for innovation, creativity, and rapid problem-solving.
One of the most studied and well-known academic processes is the Design Thinking
process, which is a methodology that suggests stages that lead to the development of
new, unique, efficient as well effective ideas (T. Brown, 2008). Although there are a few
approaches to the framework of the DT process (Johansson-Sköldberg et al., 2013;
Liedtka, 2014), there is a classic proposal developed by a world-renowned college -
D.stanford School (Johansson-Sköldberg et al., 2013).

This study looks for a research gap, embarking on a mission to elucidate the potential
integration of ChatGPT on the Design Thinking human-centered process. This
investigation sets out to redefine the DT framework in the context of ChatGPT
integration advertising agencies. The main objective of this thesis is to explore and
analyze the application of integrating ChatGPT, a generative AI model, within the Design
Thinking process, aiming to provide a revised framework that potentially enhances
creative problem-solving for advertising agencies.

Thus, the research question is: How does the integration of ChatGPT into the Design
Thinking process influence creative problem-solving in advertising agencies, and what
are the potential benefits and challenges of this integration? The potential hypothesis
of this research is that incorporating ChatGPT into the Design Thinking process can lead
to a more dynamic and efficient problem-solving structure, promoting greater creativity,
deeper knowledge, and more comprehensive solutions.

2
 2. LITERATURE REVIEW

2.1. DESIGN THINKING OVERVIEW

Design Thinking is an iterative approach that looks into understanding users, challenging
the assumptions, and find alternatives to problems. The purpose is to identify special
ways and solutions to problems and questions and understanding the people to whom
products are made is central to design thinking (Dam & Siang, 2022).

The increase in literature on Design Thinking in recent years has been substantial, with
a surge in popular management books and articles in leading business publications such
as The Economist and Harvard Business Review (T. Brown, 2008). This surge reflects the
widespread application of Design Thinking across various fields and its extensive study
in universities as a method for innovation (Liedtka, 2014). The growing interest in Design
Thinking is evident from the increasing number of practitioner writings and academic
publications that mention the term "design thinking" (Mahmoud-Jouini et al., 2016).

Figure 1- Interest in “Design Thinking” over time (last 10 years) by Google Trends

Furthermore, the application of Design Thinking has been reported in major

organizations, including P&G, Samsung, and others (Micheli et al., 2019), and has been
recognized as a way to problem-solving definition. DT has been also recognized as a
method for innovation in various sectors, including healthcare (Altman et al., 2018)and
education (Clark et al., 2020), and empirical evidence supports the relevance of Design
Thinking for innovation activities (Roth et al., 2020). The methodology of Design Thinking
has been recognized as a method of creating innovative and original ideas that respond

3
to the demand that is growing in companies and organizations (Lim et al., 2022). Also, it
emphasizes the need for experimental studies involving Design Thinking to understand
further its impact (Clark et al., 2020).

The concept of Design Thinking, while widely discussed, lacks a universally recognized
definition, and even its label is debated among those who use and support it (Liedtka,
2014). Peter Rowe, a Harvard professor in architecture and urban planning, was one of
the first to use the term in a significant way (Peter G. Rowe, 1987), but his focus was on
architectural design, which differs from the contemporary business application of the
term. It's more accurate to link the current business-focused interpretation of Design
Thinking to the innovation consultancy IDEO, under the influence of its founders David
Kelley, and Tim Brown.

Despite its widespread use, there is a lack of concrete definitions from its early
champions. They often describe design thinking in vague terms, referring to it as what
designers engage in, or applying designers' methodologies to problem-solving (T. Brown,
2008). Others offer a more formal definition, seeing it as the use of design methods by
diverse teams across various innovation challenges (Seidel & Fixson, 2013).

Design Thinking has been recognized as an approach human-centered around people’s

needs, drawing on design principles for fostering innovation. Although it has been
adopted by numerous large companies in recent years, there are still few empirical
studies on its implementation in organizational environments, which leads to a limited
understanding of its effects when integrated into business practices (Seidel & Fixson,
2013).

In discussions among business executives, DT is not often associated with the traditional
domains of design, such as engineering, industrial design, or communication design.
Instead, it is portrayed as a human-centered, widely applicable methodology for
increasing creativity and promoting innovation (T. Brown, 2008). Prestigious institutions

4
such as the Rotman School of Management and Stanford University have launched
Design Thinking programs. These programs are built around more prescriptive models
of DT, incorporating frameworks equipped with specific tools and highlighting the
importance of a human-centered innovation process and the collaboration of
interdisciplinary teams (Camacho, 2016).

2.1.1. A Human-Centric Approach

The design thinking places the user in the middle and it involves understanding the
needs and perspectives of the users together with cross-functional team collaboration,
a culture of experimentation, and iteration.
Another main value of design thinking is empathy which enables the designers to
comprehend how the users feel, relate effectively, and think of a product or service
(Elsbach & Stigliani, 2018).
Empathy allows seeing the user’s problem through other’s eyes thus yielding valuable
input on the design process resulting in innovative user-friendly designs and solutions
(Dam & Siang, 2022). By employing empathy, designers put themselves in the place of
the users, explore the users’ feelings, and offer responses addressing what people need
and want which defines DT as a human-centered approach (Carlgren et al., 2016).

2.1.2. The Design Thinking Stages

In the context of Design Thinking research, it is clear that numerous variations of the
process are currently in use, ranging from three to seven distinct phases and more static
or dynamic modes (Johansson-Sköldberg et al., 2013; Liedtka, 2014).
Stanford's d.school stands as a pioneering institution, especially because of Its unique,
hands-on approach to Design Thinking, deeply rooted in empathy and human-centric
design. The Stanford d.school's framework, distinguished by its innovative blend of 5
steps (empathy, define, ideate, prototype and test), aligns seamlessly with the
challenges and opportunities my research aims to address (Stanford d.school, 2018).

5
 Figure 2 - The Five Phases of the IDEO/d.school Design Thinking Process

1. Empathize
Empathy stands as the bedrock of design methodologies centered around human needs.
To obtain empathy, the approach entails observing - watching users in their day-by-day
environment to grasp their behaviors, interacting - engaging with users, and
experiencing - stepping into the shoes of the person to recognize their studies. In a
human-focused process, it is crucial to have a deep understanding of the people for
whom the designs are meant because the challenges addressed are usually faced by
specific users, and not the designers themselves. Observing individuals in their
environment sheds mild on their actions, emotions, and needs and these insights are
key to crafting revolutionary solutions.
2. Define
During the “define” stage, designers take into account everything that they have noted
down in the preceding “empathize” phase. In this stage, various insights obtained from
users are assembled to form a succinct yet usable problem statement. In other words,
it is about defining very well the problem or opportunity. This process involves the
analysis of collected information to determine what trends and learnings arise from
users’ actions, sentiments, and thoughts. These insights have now been framed in such
a manner to direct the design process leading usually to a point of view statement.
3. Ideate
The Ideate stage is characterized with the aid of its open, brainstorming nature, in which
the aim is to discover an extensive spectrum of opportunities, encouraging divergent
thinking and pushing past obvious answers to uncover innovative approaches. The

6
ideation is not about immediately finding the right answer but generating ideas and
exploring many different directions for the same insight.
During Ideate phase, various techniques and tools can be employed to stimulate creative
thinking such as brainstorming sessions, mind mapping, sketching, role-playing, or using
analogies and metaphors.
The emphasis is on quantity over quality at this point, under the principle that a large
number of ideas increases the chance of developing a truly effective and innovative
solution. The ideas generated are then used as the foundation for the subsequent stages
of prototyping and testing.
4. Prototype
Prototype means giving life to thoughts – turning them into real stuff that can be
explored. In this stage, prototypes may be as basic as drawings or highly interactive
representations of the model based on the complexity of the solution and available
resources.
These prototypes are however made quickly and cheaply to help designers to delve into
several elements within their ideas and identify what works and what does not.
Experts should be able to point out any problems observed during the prototype phase
at its completion point and any issues remaining unresolved.
5. Test
In the Testing stage, users are given prototypes to use, while their interaction and
feedback are closely monitored. The main goals here are to gather user feedback, ideally
explore their experience, and assess how the proposed solution fares in a real-use
situation. Although this stage marks the end of the five-stage design thinking model, it
is important to remember that design thinking is cyclical, and insights gained can often
redefine problems or some pain points of a project. This deeper understanding can shed
light on how users interact with the product and their thoughts, behaviors, and
emotional responses to it and such insights may prompt a return to earlier stages of the
design process. Continuing further iterations, the process involves adjustments and
refinements, possibly considering other solutions.

7
2.1.3. Design Thinking applications in different areas

Design thinking is catching on these days, especially for spurring new ideas and solving
problems in various areas. It has not only gained significant attention and interest from
academics (Micheli et al., 2019), but has also been widely applied in fields such as
business, education, health, sustainable development, and the creativity industry
(Bahrini et al., 2023).

DT has gained attention for its application to drive innovation encourage and address
innovation challenges (Seidel & Fixson, 2013), and its applications in manufacturing,
product, and service design have been praised indicating its importance, for example, in
commercial product design (Pratomo et al., 2021). In addition, DT has been recognized
as useful for project and creativity success, although it has been criticized for deviating
from the standard business processes, making implementation difficult (Roth et al.,
2020).
There is a call in education to pay more attention to the application of design thinking,
especially in business education, as it is effective in enhancing students’ creativity and
entrepreneurial orientation (Pratomo et al., 2021).

Furthermore, in the context of sustainability, it is a potentially flexible approach to

addressing sustainability challenges and is widely applicable to innovation and solving
problems (Clark et al., 2020). The use of design concepts in healthcare has shown
positive results, with users reporting greater satisfaction with the design thinking
interface as well as increased productivity and effectiveness (Altman et al., 2018).

Design thinking methodology has a significant influence on the creativity industry by

fostering creativity, innovation, and problem-solving. A. G. Lafley, the former CEO of
P&G, emphasized that design thinking promotes creativity and innovation in products,
services, and organizational approaches (Micheli et al., 2019). This is achieved through
a divergent phase in the design thinking process, which aims to generate multiple ideas
and reframe problems to envision potential solutions (Dell’Era et al., 2020), where

8
professionals and researchers emphasize the need for theoretical frameworks to
enhance and teach creative thinking strategies.

2.2. GENERATIVE ARTIFICIAL INTELLIGENCE

The business landscape in the 21st century has undergone significant transformations,
primarily influenced by technological advancements, including the rise of artificial
intelligence which has been improving the efficiency, productivity, and quality of
processes and workers’ outputs (Lee et al., 2018).
The evolution and broadening of machine learning have paved the way for advanced
technologies in digital content creation, particularly in generative AI (McKinsey &
Company, 2023). Generative AI models operate on unsupervised or semi-supervised
machine learning techniques, producing human-like artifacts through statistical and
probabilistic methods (Goodfellow et al., 2020).
Among various generative AI models, two prominent types are the Generative
Adversarial Network (GAN) and the Generative Pre-trained Transformer (GPT)
(Abukmeil et al., 2021,Jovanovic & Campbell, 2022); Sun et al., 2017). GAN, commonly
used in generative AI applications, employs dual neural networks—the generator and
the discriminator. The generator creates outputs, like facial images, while the
discriminator evaluates their authenticity. This process continues until the discriminator
cannot distinguish between synthetic and real outputs, ultimately considering the
synthetic as real (Wang et al., 2017; Liu et al., 2023). Therefore, GANs are extensively
utilized in generating data-driven voice, graphics, visuals, and videos.

On the other hand, GPT models harness extensive quantities of publicly available digital
information for natural language processing (NLP), allowing them to produce text that
closely resembles human writing in multiple languages (Hirschberg & Manning, 2015).
These models demonstrate remarkable writing abilities, ranging from crafting
paragraphs to composing entire research articles on diverse topics (Aydın & Karaarslan,
2023). GPT-3, an advanced iteration of these models, boasts 175 billion parameters,
surpassing previous models in task-agnostic capabilities and rivaling the best fine-tuning

9
methods (T. B. Brown et al., 2020). Subsequent sections will delve into an overview of
ChatGPT, its potential to enhance creative industries, its limitations, and how
professionals might leverage it to streamline business processes.

In conclusion, there are several potential benefits of using Generative AI (GAI)

nowadays, that can lead to efficiency improvements, time savings, and cost reduction.
GAI can also raise the bar from innovators’ work, to enhance customer experiences
through personalized marketing campaigns and even improved customer service.
Finally, GAI can reshape the roles, potentially leading to new opportunities for
innovation and creativity (Kanbach et al., 2023).

2.2.1. ChatGPT

OpenAI ChatGPT was released a few months ago, and many top business leaders have
stated that it would transform our lives and the way we work. ChatGPT is an extension
of the generative artificial intelligence (AI) family, which includes tools for producing
text, code, graphics, audio, and video material, among other types of information
(Korzynski et al., 2023). As Google's rival ChatGPT, Bard is also based on transformer
technology, which is a neural network architecture that generates predictions based on
inputs. ended up gaining a greater reputation due to its early open-source tool
condition.

Figure 3- Interest in “ChatGPT” over time (last 5 years) by Google Trends

10
Generative Pre-training Transformer (GPT) stands out as a Large Language Model (LLM)
as a result of its extraordinary dimensions and the massive volume of data that was
applied in its training. As one of the largest language models in the world, GPT relies on
billions of data sources since the algorithm that drives it has access to the whole
Internet. GPT is a versatile software that may be used for a range of language-based
tasks, such as question-answering, translation, and text production. (Lund et al., 2023).
Its use of deep learning methods and advanced algorithms sets it apart from other
language processing tools by allowing it to comprehend the context of a document and
provide replies that seem human.

By training ChatGPT on additional data and enhancing its algorithms, OpenAI aims to
improve its precision and reliability. In addition, it's critical to remember that ChatGPT
is a tool and should not be utilized in place of human intellect or judgment. Instead, it
should be used sparingly and in tandem with human oversight and evaluation (Haleem
et al., 2022).

2.2.2. ChatGPT’s Limitations

While it holds great potential, it also risks reinforcing and magnifying pre-existing
societal biases and inequalities, which could adversely affect educational processes and
outcomes. For instance, if the training data for a model is prejudiced in favor of certain
demographics, the output may exhibit bias against others (Kasneci et al., 2023).
Therefore, it's crucial to guarantee that data used for training or fine-tuning on specific
tasks is inclusive and reflective of diversity. Consequently, human supervision is
essential and vital in reducing bias and ensuring the positive use of large language
models in educational settings. And saying that, several concerns have been raised
about the use of AI and ChatGPT in education, for example. Some educators worry that
these technologies could lead to a decrease in educational activities and foster laziness
among teachers and students due to reduced analytical skills (Grassini, 2023).
Additionally, ChatGPT has significant limitations as a research tool because sometimes

11
it gives incorrect answers, suggests nonexisting sources and can be overly sensitive to
arbitrary differences in prompt phrasing (Gordijn & Have, 2023).

There are also several ethical concerns and potential risks associated with using
ChatGPT. This encompasses concerns related to privacy and security, ethical dilemmas,
the potential for mistakes or misinterpretations, the likelihood of biased suggestions,
and the danger of plagiarism (Bahrini et al., 2023). To address these concerns, it is
important to have clear ethical guidelines and regulations in place to ensure that
ChatGPT is used ethically and responsibly.

2.2.3. ChatGPT’s applications in different areas

Despite its limitations, ChatGPT has a lot of capabilities that have been used in various
fields, such as education, health, and psychology. In the field of education, for example,
the tool has been used for conducting research and as an auxiliary in studying and
teaching methods (Aydın & Karaarslan, 2023). The influence and impact of the
application of this GAI tool on education methods and processes have been approached
in every research article because of its impact on traditional and even scientific
education (Cooper, 2023). ChatGPT has a wide range of potential applications across
various industries, especially in the business world (Shaji George et al., 2023) and
problem-solving industries which is the area this research will focus on. Because of its
capabilities of synthesizing information, data analysis, and content generation, this tool
can be used in many different tasks such as operations management, business analytics,
marketing analysis, strategic thinking, and others (Bahrini et al., 2023).

2.2.4. ChatGPT integration in Design Thinking for the advertising industry

Specifically, in the ChatGPT case, it is an innovative Generative Artificial Intelligence

(GAI) tool that has been used by many advertising agencies’ professionals nowadays,
since its launch in November 2022. Not only for brainstorming and generating content
(Fattepure, 2023) but also for information providers. This tool became viral because of

12
its utility and free access to everyone, however, there is still no research made on its
application in creative advertising and project management jobs.
Now focusing on the main capabilities of DT - observe, frame and reframe, imagine,
design, and experiment (Beckman, 2020) –, it can easily lead to a functional tool for
creatives who want to create novel ideas and as well for marketers or project managers
who need problem-solving methods (Mahmoud-Jouini et al., 2016). Although there are
many reframing proposals for Design Thinking (Johansson-Sköldberg et al., 2013), there
is no research done about the reframing of the DT process including the ChatGPT as a
tool to enhance its utility.
The importance of AI lies in its ability to overcome certain human limitations, enhancing
human performance through data-driven solutions and tools (Samuel et al., 2023).
However, as we can see in Design Thinking’s literature review, empathy, and human
approach are one of its core values, so can a GAI approach be integrated into this
process?

13
3. METHODOLOGY

3.1. RESEARCH MOTIVATION

In recent years, the business landscape has experienced a transformative shift with the
advent of AI technologies, notably in how they augment creative and management
processes within organizations. At the agency where I currently work, a Portuguese
advertising agency, we have embraced a creative and problem-solving methodology
that closely aligns with Design Thinking principles. This approach has not only catalyzed
innovation in our creative team but has also shown potential for application in project
and client management. Building on this insight, this research proposes an innovative
integration of DT across our agency’s creative and project management teams.
This agency, with more than 12 years of experience in the national and international
market, works with the most diverse clients in the most diverse business areas. It is
made up of 40 employees, divided into creativity, project and client management, and
accounting departments. Although they have different tasks, these departments have
one thing in common: the need to solve problems, whether creative or operational. The
cornerstone of this proposed integration is the incorporation of ChatGPT, an AI tool that
has gained significant traction. ChatGPT's versatility and widespread adoption within our
agency make it an ideal candidate for this integration. The rationale behind choosing
ChatGPT hinges on its proven effectiveness in enhancing creative brainstorming,
decision-making processes, and operational efficiency.
However, despite its growing popularity and apparent potential, a notable gap exists in
the literature and practice regarding the application of ChatGPT within the DT process.
This research aims to bridge this gap by exploring how ChatGPT can be effectively
integrated into the DT methodology. By doing so, we seek to not only optimize our
internal processes but also contribute to the broader discourse on the synergy between
AI tools and design thinking in organizational settings.
This study, therefore, embarks on a journey to explore and document the potential
benefits and challenges of integrating the ChatGPT into the DT process. The goal is to

14
provide a comprehensive framework that can be adopted by our agency and others in
the industry seeking to leverage AI tools within their design and management practices.

3.2. RESEARCH METHODOLOGY

This study aimed to understand the integration of ChatGPT in the Design Thinking
process within the advertising industry and to achieve this, it adopted a qualitative
exploratory approach. Also, in line with the Total Quality Framework (TQF) principles,
this study incorporates Credibility, Analyzability, Transparency, and Usefulness during
its research process (Roller & Lavrakas, 2015). This approach allows for an in-depth
exploration of experiences and perceptions, crucial for Credibility by uncovering
insights. While this approach prioritizes depth over statistical generalizability, its
findings will lay the groundwork for future quantitative or mixed-methods research.
Analyzability is achieved through data processing methods ensuring accurate and
complete analysis and also the principle of Transparency is rigorously applied, detailing
every aspect of the research process - from the initial design and selection of
participants to the execution of data collection and analytical procedures. Lastly, the
commitment to Usefulness is seen in widening knowledge about the role of AI in creative
processes and also laying a foundation for future research thus improving the state of
knowledge in the emerging field.

3.3. PARTICIPANT SELECTION

This investigation conducted twelve interviews with a diverse group of advertising

professionals. These professionals were selected based on their varying levels of
experience, job roles within their respective agencies, and age groups. This diverse
selection criteria ensured a comprehensive understanding of the subject matter from
multiple perspectives. It was carefully chosen by a mix of individuals including creative
directors, copywriters, account managers, project managers, and strategists to capture
varied insights based on their specific functions within the advertising agencies.
Additionally, it included participants ranging from industry veterans to mid-career
professionals and newcomers to ensure a comprehensive perspective across different
career stages. This diverse selection aimed to gather a broad spectrum of experiences

15
and viewpoints, offering a nuanced understanding of how different professionals
perceive and utilize ChatGPT in the Design Thinking process.

3.4. COMPANY SELECTION

The three advertising agencies invited and involved in this study were chosen based on
their popularity and well-known status within the industry. This selection guarantees
that the findings will be relevant and applicable across the advertising spectrum,
encompassing agencies that are innovating at the forefront of AI integration and those
that are just beginning to explore such technologies.

3.5. INVESTIGATION’S OUTPUT

The primary focus of the investigation was to identify emerging patterns and differences
in how the interviewees described the integration of ChatGPT in the DT process. This
objective aligns with the exploratory nature of our study, allowing us to gain in-depth
insights into a relatively new area of application within the advertising industry for
future research.

16
4. EMPIRICAL STUDY

4.1. STUDY DESIGN

4.1.1. Research Objectives

The primary objective of this thesis is to conduct an exploratory and analytical

investigation into the integration of ChatGPT within the Design Thinking framework
utilized by advertising agencies. This exploration aims to understand the influence of
ChatGPT on the creative problem-solving processes within these agencies, as well as
identify potential benefits and challenges arising from the integration of ChatGPT into
the Design Thinking process. And also proposes a revised framework for the Design
Thinking process that incorporates ChatGPT as an output to implement in advertising
agencies.

4.1.2. Premises

Based on the initial understanding of the subject and the defined research question, the
study posits the following premises:
• P1 - ChatGPT acts as an allied tool in the creative and problem-solving process,
aiding in the generation of ideas and solutions.
• P2 - The incorporation of ChatGPT into the creative process enhances the
workflow’s dynamism and efficiency within advertising agencies.

These premises will be explored through qualitative analysis of experiences and

perceptions of advertising professionals regarding their use of ChatGPT in the Design
Thinking process.

17
4.2. SAMPLING METHOD

4.2.1. Intentional and Stratified Sampling

The study employs an intentional sampling strategy (Thomas, 2023; Turner, 2003),
specifically targeting individuals within the advertising industry who are acquainted with
both ChatGPT and creative or management processes. This focused approach ensures
that participants are well-positioned to provide informed and relevant insights into the
practical integration of AI tools like ChatGPT in their workflows.
To mitigate potential biases that often accompany intentional sampling, the research
incorporates stratified sampling techniques. This methodological choice aims to ensure
a diverse representation across different agency departments and levels of experience.
By doing so, the study captures a breadth of perspectives, from strategic management
to hands-on creative execution, offering a comprehensive understanding of ChatGPT's
role in various facets of the advertising industry.

Figure 4 - Stratified Sampling by Scribbr (Thomas, 2023)

After defining the target population of this study - people working in the advertising
industry, they were divided and organized by agencies operating in Portugal, namely in
Lisbon. After defining the population, it was necessary to find out which jobs in an
agency potentially use tools such as ChatGPT. As such, the population was divided into
account managers, designers/art-directors and copywriters. Other positions of great
interest encourage diversity of response: creative directors and strategists.
Another important point in stratifying the sample was to consider agencies operating in
Portugal, taking into account factors that always enhance diversity and rich insights. In
this way, three agencies with different profiles were considered. In this way, two
arguments were taken into account:

18
 1. Representative diversity: The agencies selected - two national and one
multinational, based in Lisbon - collectively represent a diverse cross-section of
the advertising scene in Portugal. This diversity is reflected in their size, scope
and main focus, which ranges from traditional advertising to digital performance.
2. Feasibility and Opportunity for depth insights: Working with a smaller, more
specific group of agencies allows for a more in-depth analysis of each
organization's unique experiences and practices. This depth is crucial in a
qualitative study, where the richness of the data often exceeds the need for a
large sample size.

4.2.2. Participant Sample

This selection ensures a varied and comprehensive exploration of viewpoints, capturing

how different roles within each agency perceive and utilize ChatGPT in the Design
Thinking process. The number of participants from each agency was determined based
on the size and complexity of the agency, ensuring a representative sampling from each
organizational structure.

Table 1- Interview’s Sample

Agency ID Dimension Business Core Job Roles

2 Account managers
1 Designer
AGEN1 48 employees Advertising 1 Copywriter
1 Creative Director
1 Account manager
Advertising and digital 1 Designer
AGEN2 41 employees 1 Copywriter
performance
1 Creative Director
1 Account Manager
Advertising and events 1 Designer
AGEN3 27 employees
production
1 Copywriter

19
4.3. DATA COLLECTION

4.3.1. Initial Contact and Invitation Process

The data collection process commenced with initial contact with potential participants
via mobile phone between November 30 and December 8, 2023. This step involved
inviting advertising professionals from the selected agencies and obtaining their email
addresses for further communication. The purpose of this initial contact was not only to
extend the invitation but also to introduce the research and establish a preliminary
rapport.
Following the phone calls, formal email invitations were sent out from December 8 2023
to January 8 2024. These emails included a direct link to a questionnaire hosted on the
NOVA IMS Qualtrics platform. This approach was chosen to provide participants with
the flexibility to respond at their convenience, ensuring thoughtful and in-depth
answers. The timeframe allowed for ample response time, ensuring participants could
engage with the questions without feeling rushed.

4.3.2. Benefits and Limitations of Email-Based Data Collection

The data collection method of email interviews used in this study has inherent
limitations and advantages. Although Internet access and the need for digital literacy
are mandatory, this factor did not influence the study given the target chosen for it.
However, the absence of non-verbal clues and the human conduct of the interview
presented challenges in building relationships, interpreting accurately, and maintaining
a dynamic dialog, and also increased the risk of bias due to the potential exclusion of
participants. On the other hand, the method facilitated the disclosure of sensitive
information due to the privacy it provided, allowed for a more in-depth exploration of
topics with the opportunity for participants to give nuanced and thoughtful responses
when it suited them and proved effective in engaging diverse populations, particularly
those that may be more difficult to reach through traditional interview methods.

20
4.3.3. Interview Questions

The focal point of this thesis is a set of unstructured interview questions (Qu & Dumay,
2011), conceived to uncover the intricacies of integrating ChatGPT into the Design
Thinking framework at advertising agencies. These questions are the conduits through
which we seek to understand, analyze, and ultimately refine the creative problem-
solving process that is important to the advertising industry. The questions are
structured around key themes that align with the study's primary objectives to assess
the influence of ChatGPT on creative problem-solving, to identify the potential benefits
and challenges of its integration, and to propose enhancements to the existing Design
Thinking framework.

Table 2 - Interview Questions

Objectives Questions

Can you describe how you have integrated

ChatGPT integration on the professional’s ChatGPT into your professional workflow?
workflow.
When do you find it most beneficial?
In what stages of Design Thinking do you
Potential ChatGPT integration in the Design envision integrating ChatGPT specifically?
Thinking Process. What potential benefits do you foresee from
incorporating ChatGPT into Design Thinking?
What challenges or limitations do you anticipate
Identify challenges and limitations of ChatGPT might arise from integrating ChatGPT into Design
Thinking?
In what ways do you think ChatGPT can enhance
Impact of ChatGPT on Creativity and Problem
creativity and problem-solving within Design
Solving
Thinking?

4.4. DATA ANALYSIS

The Total Quality Framework (Roller & Lavrakas, 2015) provides a comprehensive
approach to qualitative research that emphasizes quality and integrity throughout the
research process and this framework is particularly pertinent to this thesis, which seeks
to explore the nuanced dynamics of ChatGPT's role in the Design Thinking process within
the advertising industry. As mentioned before, TQF's emphasis on Credibility,

21
Analyzability, Transparency, and Usefulness also serves as cornerstones for conducting
a rigorous qualitative analysis.

4.4.1. Qualitative Content Analysis with TQF

Qualitative content analysis under the TQF involves a systematic coding and categorizing
approach to make sense of large amounts of textual data and it enables one to sift
through the data with a lens focused on discerning meaningful patterns, themes, and
insights.
The empirical data, collected from interviews with professionals across Portuguese
advertising agencies, will be subjected to this systematic method. Using a tailored code
form developed following TQF principles (Roller & Lavrakas, 2015) and the data content,
the interviews were coded, categorizing the content into themes that reflect the
experiences and perceptions regarding ChatGPT's integration. This method will enable
us to capture a broad spectrum of impacts, from enhancing creativity to identifying
potential barriers to effective implementation and leading to a new and reframed DT
framework.

Figure 5 – Phases in qualitative content analysis in TQF (Roller & Lavrakas, 2015)

The empirical data, derived from interviews was systematically analyzed using the 8
basic steps of qualitative content analysis as outlined by the TQF. These steps are
organized into two main phases: Data Generation and Data Analysis, as described in the
following topics.
Also aligned with TQF’s suggestion (Roller & Lavrakas, 2015) ATLAS.ti, a robust software
tool for qualitative data analysis, was utilized in the coding of the interview, and
organizing for a more efficient analysis. This coding facilitated the categorization of

22
content into themes reflecting the experiences and perceptions regarding ChatGPT's
integration. The use of ATLAS.ti bolstered the methodological rigor, providing a means
for reliably managing the data in line with TQF principles of Credibility, Analyzability,
Transparency, and Usefulness.

4.4.2. Coding process

The methodology employed for data analysis is grounded in a qualitative framework,
focusing on the systematic interpretation of textual data. Initially, deductive coding was
made where initial codes were formulated based on pre-existing insights from the
Literature Review chapter, combining that method with an inductive coding approach
to sift through the interview transcripts, allowing patterns and categories to emerge
naturally from the data. This approach facilitated the identification of recurring concepts
and insights related to the integration of ChatGPT in advertising agencies and the
different job rules. Following the initial coding, a rigorous process of review was
undertaken to refine and validate the codes, ensuring they accurately reflected the
interviewees' thoughts and the research objectives. To ensure reliability, codes were
constantly compared against the data, enabling the consolidation of related codes into
coherent categories. This structured organization of codes into groups facilitated the
identification of overarching themes, painting a comprehensive picture of ChatGPT's
impact on the creative workflows and the Design Thinking process framework.
• Initial Coding
The coding process began with an immersive review of the interview transcripts to
familiarize with the data and review the research question as well as the literature
review that was made (Step 1 – Absorb content). Subsequently, the unit of analysis
was determined (Step 2 – Determine unit of analysis), which in this study, were the
individual responses and quotes from the interview transcripts. This established a
clear focal point for the analysis and set the parameters for what would be
considered a meaningful unit for coding. Once the unit of analysis was established,
initial codes were created deductively, in a first approach, based on LR and
inductively, rooted in the respondents' language and experiences (Step 3 – Develop
unique codes). The first set of unique deductive codes provided an initial scaffold for

23
the analysis, ensuring alignment with the research objectives and the literature. In
the initial set, a set of codes was considered, aligned with some information taken
from the LR, related to some benefits and implications of using ChatGPT, such as
"Biased information" (Kasneci et al., 2023), "Data insecurity" and "Power of
synthesizing information" (Bahrini et al., 2023), "Brainstorming tool" (Fattepure,
2023) and "Efficient writing" (Aydın & Karaarslan, 2023). However, to encapsulate
the richness of the data and the nuanced perspectives of interviewees, an inductive
approach was adopted to refine and expand these codes. This iterative process,
rooted in the respondents' language and experiences, enriched the analysis,
ensuring the emergence of patterns and categories directly from the data itself. It
facilitated the identification of novel insights and interpretations, enhancing the
depth and breadth of the analytical findings, in alignment with TQF principles (Roller
& Lavrakas, 2015). This dynamic interplay between deductive reasoning and
inductive discovery upheld the credibility of the study by fostering a robust,
evidence-based understanding of ChatGPT's integration into advertising agencies
and its influence on various job roles, as displayed in Appendix B - Codebook.
• Code Review and Refinement
A preliminary coding and systematic review of these initial codes was conducted to
refine them, ensuring their alignment with the data's nuances and complexities
(Step 4 - Conduct preliminary coding). This iterative process involved constant
comparison between codes, merging similar codes, and splitting broader codes into
more precise categories. The reliability of the coding process was enhanced through
constant evaluation of codes where they were contested and revised.
• Codes
The culmination of Phase 1 was the systematic coding of content (Step 5 – Code
Content), which involved meticulously labeling each relevant excerpt from the
interviews. This step was essential for organizing the data into manageable
segments for further analysis. The coding at this stage remained close to the data,
with a focus on staying true to the participants' voices and the contextual richness
of their responses. This coding and categorization work culminated in the final

24
 Codebook presented in Appendix B - Codebook, with a total of 5 categories and 40
codes.

4.4.3. CATEGORIZATION PROCESS

Moving into Phase 2, data analysis, the next crucial step was to identify categories across
codes. This step marked the transition from coding to categorization, where similar
codes were grouped to form coherent categories (Step 6 – Identify categories across
codes). The categories were conceptualized to reflect broader patterns in the data,
going beyond mere aggregation to interpret the significance of the codes within the
context of the research questions. This stage was marked by an iterative process, where
categories were constantly refined and redefined as the analysis progressed. The goal
was not simply to organize the data but to begin synthesizing it into meaningful clusters
that would facilitate a deeper understanding of the themes and patterns emerging from
the interviews, analyzed in the next chapter Results and Discuss.
This organization was guided by the patterns that emerged from the data, reflecting
broader themes or concepts relevant to the research questions. The grouping process
was iterative, with categories being refined, merged, or divided as deeper analysis
revealed more about the relationships between codes. This phase was critical for
moving beyond descriptive coding to analytical coding, where the data began to inform
the research findings more substantively. The use of visual mapping tools facilitated this
process, enabling a holistic view of how codes interconnected and formed broader
thematic categories.

25
 Figure 6 - Categorization of codes

4.4.4. LIMITATIONS OF THE DATA ANALYSIS PROCESS

When analyzing the interview data through a qualitative content analysis, this research
recognized the limitations inherent in this methodology, as the Total Quality Framework
(Roller & Lavrakas, 2015) mentioned. Firstly, the process of interpreting textual material
presents significant challenges, namely avoiding misinterpreting causal relationships
and constructing narratives that are not fully substantiated by the data. This analysis
was conducted with a critical eye toward these pitfalls, emphasizing thematic
identification without presuming causality to mitigate the risk of over-interpretation.
Furthermore, the TQF highlights the potential distortions introduced by the inherent
biases and perceptions of interview participants.
This study has been mindful of these factors, striving to present a balanced view that
encapsulates both individual experiences and broader insights into the creative industry.
The dynamic nature of participants' responses and the variability in their perspectives,
informed by both personal and collective experiences, were considered to discern

26
patterns that reflect a collective experience while acknowledging the unique context
each participant contributes.
.

27
5. RESULTS AND DISCUSSION

5.1. OVERVIEW

This chapter aims to elucidate the findings of the empirical study conducted on the
integration of ChatGPT within the Design Thinking process in advertising agencies. Given
the exploratory and qualitative nature of this research, the chapter is structured to
provide a comprehensive analysis of the data collected from interviews with
professionals in the advertising sector. This investigation was driven by two primary
objectives: firstly, to understand how ChatGPT is being integrated into creative
workflows and problem-solving processes within these agencies; and secondly, to assess
the impact of such integration on the efficiency and dynamism of the Design Thinking
framework.
Building on the theoretical foundation laid in the literature review, this chapter is
organized to systematically address the premises highlighted before:
• P1 - ChatGPT acts as an allied tool in the creative and problem-solving process,
aiding in the generation of ideas and solutions.
• P2 - The incorporation of ChatGPT into the creative process enhances the
workflow’s dynamism and efficiency within advertising agencies.
The subsequent topics are dedicated to presenting the findings related to key themes
that emerged from the data analysis (TQF Step 7 – Identify themes/patterns across
categories). These themes include the integration of ChatGPT into professional
workflows, its perceived benefits and challenges, and its overall impact on problem-
solving creativity.
This chapter not only aims to present these findings but also to discuss them to
contribute to the ongoing discourse on the intersection of technology and creativity,
offering new perspectives on the integration of generative AI tools like ChatGPT in
creative processes (TQF Step 8 - Draw interpretations and implications).

28
5.2. RESULTS

5.2.1. Enhancement of Creative and Problem-Solving

The ChatGPT enhances both creative processes and strategic planning capabilities and
it includes leveraging AI for efficient copywriting, generating ideas, and providing
strategic insights.
Keywords: Efficient copywriting, creative insights, strategic planning, innovative paths,
brainstorming, conceptualization of ideas.
Examples: “…important assistant when it comes to building ideas that are creative but
strategically functional”; “... é um meio rápido e eficaz de produzir mensagens...”;
“...processo de brainstorming ele é muitas vezes mais prático para encontrar insights.”
• Sub-theme: Creativity increase
This sub-theme focuses on enhancing the creative process through efficient copywriting
for creativity and explanation, conceptualizing ideas, and providing diverse insights. It
illustrates how ChatGPT serves as a tool for creative enhancement, offering diverse
ideas and inspiration.
Examples: “... ferramenta de apoio à pesquisa de insights...”;“...processo de
brainstorming ele é muitas vezes mais prático para encontrar insights…”; “… ferramenta
pode dinamizar o processo criativo e ajudar a explorar ideias…”.
• Sub-theme: Problem-solving driven
This sub-theme describes how ChatGPT aids in strategic planning and problem-solving
and it shows how ChatGPT optimizes problem-solving processes, unlocks alternative
viewpoints, and suggests practical execution strategies, thereby enhancing creative
problem-solving capabilities.
Examples: “... it also might be useful for seeking feedback and alternative ideas.”;
“...quando estou presa num raciocínio e preciso de novas hipóteses ou pontos de vista.”;
“… desbloqueio da criatividade”.

29
5.2.2. Workflow Optimization

This theme explores the role of ChatGPT in streamlining operations and enhancing
workflow efficiency. It focuses on how AI tools can improve planning, organizing, and
executing work processes but does not dive into the creative aspects of AI applications.
Keywords or Phrases: Workflow efficiency, process optimization, time management,
quick adoption, collaborative workflow.
Examples: “I've integrated ChatGPT as a collaborative tool in my professional workflow,
shaping my thinking approach.”; “ChatGPT is the most useful tool for stages like
planning, structuring, and organizing processes and information.”; “A discreet
integration ensures a more streamlined process, offering valuable input without
overshadowing the creative brief writing…”
• Sub-theme: Time-saving
Highlights the benefits of enhancing time management and it fits into Theme 2 by
specifically demonstrating how ChatGPT facilitates operations by effectively reducing
time in some processes.
Examples: “... pode ajudar a encurtar certas partes do processo...”; “... economiza o
nosso tempo...”; “... consigo filtrar rapidamente ganhando tempo...”; “… offers
considerable time-saving benefits.”
• Sub-theme: Information search and feedback
This sub-theme emphasizes the importance of using ChatGPT for instant feedback, and
information retrieval, and how it boosts productivity.
Examples: “... ferramenta de apoio à pesquisa de insights e informação...”; “…
sistematização e pesquisa de informações, referências e publicações sobre determinado
assunto...”; “… feedback imediato sobre as ideias apresentadas…”; “... providing
feedback on our ideas and finding some solutions to our problems.”

5.2.3. Challenges and Limitations of AI Integration

This theme focuses on the limitations and challenges associated with integrating GenAI
tools like ChatGPT, focusing on creative, operational, and ethical concerns.

30
Keywords or Phrases: Creative boundaries and limitations, the need for human
oversight, data security, operational risks, ethical concerns.
Examples: “A nível de motor de busca, para responder a determinadas questões, acaba
por ser limitado...”; “… biased information…”; “Não acho que seja melhor na sugestão
de ideias criativas…”; “... a criatividade e a inovação vão precisar sempre do cunho
humano.”.
• Sub-theme: Creative and Critical Thinking Limitations
This sub-theme deals with the limitations of AI in fostering creative and critical thinking,
highlighting the need for human oversight and recognizing the autonomy limitations of
AI. It aligns with Theme 3 by outlining concerns over this AI tool’s impact on creative
differentiation and critical thinking abilities.
Examples: “Falando diretamente em criatividade, é contudo uma ferramenta limitada,
uma vez que ainda se baseia apenas em texto.”; “A versão gratuita do ChatGPT, de certo
modo, não é uma ferramenta de criatividade não acho que tenha o potencial para
cumprir, autonomamente, o processo completo.”; “... penso que não é igualmente útil
para todas as tarefas criativas, depende da complexidade do pedido ou do tema...”; “…
great need for human thought to accompany the information so that it doesn't become
strange or random.”; “...uso frequente pode ter um impacto negativo no
desenvolvimento da capacidade criativa pessoal...”
• Sub-theme: Bias and Ethical Concerns
Data security and bias information issues, the risk of overreliance on AI, and the
potential for process robotization are the main keywords of this sub-theme and it fits
within theme 3 by discussing the operational efficiencies and ethical dilemmas set by
integrating AI tools into workflows.
Examples: “… information breaches (related with data security and rgpd)”; “Robotização
de processos…”; “Se o assumirmos como um primeiro passo em qualquer fase criativa
estamos a privar o nosso cérebro de fazer esse processo, o que pode levar à habituação
e a um consequente "atraso" no desenvolvimento da habilidade criativa.”; “... o facto
de o modelo de "machine learning" se basear em datasets predefinidos acaba por criar
um "bias" que irá viciar sempre os resultados…”.

31
5.2.4. Integrating ChatGPT into Design Thinking Stages

The integration of ChatGPT across the stages of the Design Thinking process, highlights
how AI can contribute to a deeper understanding of users’ environment, a clearer
definition of project goals, an enhanced ideation process, and support in prototype
development. It focuses specifically on the application of AI within the framework of
Design Thinking. In this data set, none of the participants considered the Test phase as
a potential for integrating the ChatGPT tool, which is why it was not considered in these
results.
Keywords or Phrases: Empathize stage, Define stage, Ideate stage, Prototype stage, user
understanding, project goals.
Examples: “… start and go into more in-depth research…”; “Empathize and Define
stages.”; “In the empathize and define stages.”; “Mas a integração do ChatGPT no
processo pode revelar-se incrivelmente útil das fases 2 – 4...”.
• Sub-theme: Design Thinking Process Enhancement
ChatGPT can be utilized across the Empathize, Define, Ideate, and Prototype stages of
the Design Thinking process. It showcases the tool's potential to enhance 4 out of 5 steps
of the Design Thinking process, from gaining empathetic insights into users' needs to
assisting in the rapid development and evaluation of prototypes. ChatGPT's role in this
process is to facilitate deeper insights, foster creative ideation, and streamline the
prototyping process, thereby enhancing the overall effectiveness and efficiency of
Design Thinking.
Examples: “In the "Prototype" stage as well - in strategy planning. A discreet integration
ensures a more streamlined process, offering valuable input without overshadowing the
creative brief writing and strategic planning aspects of my role.”; “... útil na fase de criar
ideias...”; “Na fase inicial de pesquisa, recolha de informação”; “o ChatGPT é bastante
eficiente para a materialização de textos explicativos em determinados projectos”; “…
start and go into more in-depth research…”.

32
5.3. DISCUSSION

By addressing the dynamics between artificial intelligence and creative processes in

advertising agencies, this research aimed to clarify the perception of ChatGPT as a tool
in the creative agency industry and thus propose a natural integration of ChatGPT into
the Design Thinking framework. Through a meticulously designed qualitative
methodology that also encompasses qualitative analysis, this study attempted to
highlight real insights and conclusions about the perception of this topic among
professionals in the respective field. The overall aim was not only to explore how
ChatGPT could improve creativity and problem-solving, but also to identify the current
limitations and challenges in its use. This chapter seeks to dissect the implications of the
results, culminating in a final product that will be the updated DT framework with the
possible integration of ChatGPT. In this chapter, we will navigate through the
synthesized ideas, critically analyzing their alignment and restructuring what is the
conceptual framework of Design Thinking.

5.3.1. Overview of Key Findings

The research conducted into the use of ChatGPT in the workflows of advertising
professionals, as well as its integration into the Design Thinking framework, revealed
valuable insights into how this tool is making its mark on the landscape of advertising
agencies. The data suggests that ChatGPT is being incorporated significantly in the
ideation and conceptual development phases, offering a vast, robust and rapid
availability of information that allows for an improvement in the creative and problem-
solving process. The analysis also identifies a sense of remarkable improvement in the
efficiency of workflow processes, attributed to ChatGPT's ability to quickly generate
diverse ideas, possible solutions and instant feedback. Key themes emerging from the
data include the transformative potential of ChatGPT in fostering a culture of
collaboration and experimentation between AI and human thinking, along with the
critical need to address ethical considerations and the development of guidelines for its
effective integration.

33
 Figure 7 - Research Key Findings

5.3.2. The integration of ChatGPT in the advertising industry’s workflows

The integration of ChatGPT into professional workflows has been shown to streamline
processes, improve the information search, and save time on filtering information. This
efficiency allows creative professionals to devote more time to creative activities that
depend almost entirely on the human mind. This reinforces the value of ChatGPT in
facilitating a more dynamic and efficient workflow, which can be critical in fast-paced
advertising environments. This conclusion supports Premise 1 (P1) and Premise 2 (P2),
indicating that ChatGPT is indeed a valuable ally in the creative and problem-solving
processes that enhance the advertising agency’s workflows. These various applications
of ChatGPT, in this particular study, underline and highlight its versatility and its

34
potential to improve traditional workflows, aligning with findings from “Is ChatGPT
Leading Generative AI? What is Beyond Expectations?” (Aydın & Karaarslan, 2023) and
“ChatGPT: Applications, Opportunities, and Threats” (Bahrini et al., 2023) articles, who
highlight ChatGPT's role in driving innovation and efficiency.

5.3.3. The impact of ChatGPT specifically on problem-solving creativity

ChatGPT's role in the creative problem-solving process emerges as a key theme, with
the tool enhancing accessibility to vast information and offering alternative solutions
and points of view in one or more situations. The study also suggests that ChatGPT can
be a valuable asset supporting the ideation phase, contributing to more informed
decision-making and problem-solving approaches. This capability of providing insights,
potential solutions, and a broad perspective on creative challenges helps to make the
brainstorming and ideation process more organized and structured. Also, ChatGPT
significantly aids, as a complementary tool, in the conceptualization of ideas and
efficient copywriting, serving as a dynamic tool for generating creative content. This
aligns with previous studies highlighting AI's potential to augment human productivity
by providing diverse perspectives and rapid ideation capabilities (Lee et al., 2018).
However, while ChatGPT enhances certain aspects of creative work, such as expediting
research and generating initial ideas, it also introduces challenges in ensuring originality
and depth in creative outputs. This dichotomy reflects the broader debate on the role
of AI in creative industries, suggesting that while AI can augment human creativity, it
cannot replace the nuanced understanding and emotional depth that human creatives
bring to their work (Haleem et al., 2022).

5.3.4. The Integration of ChatGPT into the Design Thinking Framework

The integration of ChatGPT into the Design Thinking process, as highlighted by the
findings, opened a rich vein of discussion in this thesis, particularly in how AI can
revolutionize creative processes across several stages—Empathize, Define, Ideate, and
Prototype—except for the Test phase, where its integration was notably absent

35
according to participant feedback. This aligns with works that advocate for the
integration of digital tools to enhance the creativity and efficiency of the Design Thinking
process (Beckman, 2020).

This absence offers a poignant starting point for a discussion about the perceived
limitations or challenges of this type of GenAI for the Test phase in Design Thinking.
Although it was mentioned by some participants that ChatGPT provides quick feedback
on content inserted into the platform, there was no consideration of integrating
ChatGPT into this phase of the process. This may be because the process of evaluating
and providing feedback on creative processes requires a lot of human empathy and
understanding. This gap in the application of AI highlights a critical area for future
exploration, to bridge the gap between the capabilities of AI and the nuanced demands
of user testing.

In addition, the findings highlight the role of ChatGPT in improving the DT process,
marking a change in the way creative problem-solving can be approached. By facilitating
a deeper view of the user, helping to define the project's objectives and structure,
promoting a more dynamic ideation process and simplifying the development of
prototypes, ChatGPT emerges as a transformative and facilitating tool.
However, the selective application in the Design Thinking phases invites a broader
reflection on the role of AI in creative workflows. It suggests that while AI can
significantly leverage human creative capabilities, its integration should be strategically
aligned with the phases where its impact is most beneficial and where human oversight
can effectively complement AI's data-driven insights.
Some of the practical implications of integrating ChatGPT - evidenced by participants'
experiences in improving brainstorming sessions, accelerating initial research, and
refining prototypes - point to a nuanced understanding of AI's potential not only to
automate but also to profoundly enrich creative processes. This practical integration of
ChatGPT into advertising agencies not only increases efficiency, but also offers new
perspectives and solutions that may not be quickly and immediately visible through

36
traditional methodologies. These findings resonate with the perspectives shared by
participants, who highlighted ChatGPT's role in unlocking creative pathways, increasing
ideational diversity, and streamlining the explication of complex concepts.
This discussion naturally leads to the strategic integration of ChatGPT within the Design
Thinking framework, highlighting the importance of a balanced approach where AI
supports human creativity without overshadowing the essential intuitive and
empathetic aspects of design. It also points toward the evolving landscape of AI in
creative industries, where tools like ChatGPT are not just facilitators of efficiency but
also partners in creative exploration, where the pressure to innovate is constant.

5.3.5. The Challenges of ChatGPT in the Design Thinking Process

Despite the benefits observed, this study also highlights challenges and limitations
associated with integrating ChatGPT into Design Thinking.
Despite the benefits, it also sheds light on the challenges and limitations associated with
integrating ChatGPT into Design Thinking. Concerns about data security, the potential
to reduce autonomy and critical thinking, and the risk of homogenizing creative
outcomes highlight the need for human oversight and the importance of balancing AI
contributions with human creativity and intuition. These conclusions reflect the
concerns in the literature to ensure that AI tools, such as ChatGPT, complement and do
not compromise the human element in creative processes.

5.4. AI-DRIVEN DESIGN THINKING FRAMEWORK

5.4.1. Introduction to AI-Driven Design Thinking Framework

In the rapidly evolving landscape of the advertising industry, the integration of artificial
intelligence tools such as ChatGPT represents a significant leap forward in the way
creative processes are approached, managed and optimized. Traditional Design Thinking
frameworks have provided a robust structure for understanding user needs, designing
solutions, creating prototypes and testing the results. However, as digital technologies

37
advance, a marked gap emerges in these traditional methodologies - a gap that does not
include the powerful capabilities of AI to transform and elevate the creative process.
This research identifies and aims to fill this gap by proposing a revised Design Thinking
framework that incorporates ChatGPT, marking a breakthrough in the field of design and
creative problem-solving.

The creation of this AI-driven design thinking framework is not just an academic
exercise, but a strategic response to the needs of the contemporary advertising industry,
specifically in the agency where this need arose. It is driven by the personal ambition
and professional need to harness the potential of AI to significantly improve creativity,
increase efficiency and simplify everyday problem-solving processes at the agency.
Integrating ChatGPT into the Design Thinking process opens up new avenues for
advertising professionals to access a deeper understanding of user environments,
articulate clearer project objectives, participate in more enriched ideation sessions, and
support the dynamic development of prototypes. However, the notable absence of
ChatGPT in the testing phase of existing models highlights an area ripe for innovation,
suggesting that even as AI transforms creative workflows, there remains uncharted
territory in fully harnessing AI for comprehensive user testing and feedback analysis.

This new framework is designed to empower advertising agencies not only to keep pace
with the digital transformation, but also to lead it by adopting a more up-to-date and
inclusive AI approach to creativity and project management. It promises to equip
professionals with the tools and methodologies needed to more effectively navigate the
complexities of modern advertising and project management challenges, promoting
solutions that are not only innovative but also deeply aligned with user needs and
expectations. Through this AI-driven approach, the framework aspires to set a new
standard for how design thinking can be expanded and enriched in the digital age.

38
5.4.2. The foundation of the AI-Driven Design Thinking Framework

The integration of ChatGPT into the Design Thinking phases of Empathy, Definition,
Ideation and Prototyping - while notably excluding the Testing phase - represents a
differentiated understanding of the role of AI in creative problem-solving. This structure
aims to strategically harness ChatGPT to increase creativity and efficiency, reflecting on
insights from participant feedback and the wider implications for AI in design. At the
same time, integration of key themes such as Human-AI Collaboration, Flexibility and
Adaptability, and Responsible Use of AI into an AI Inclusive Design Thinking Framework
involves embedding these principles deep into each phase of the Design Thinking
process. This integration creates a comprehensive framework that not only takes
advantage of ChatGPT's capabilities but also does so in a way that aligns with human
creativity, specific project needs, and more ethical considerations.

Table 3 - AI-Driven Design Thinking Framework Stages

Stages Objectives Applications

Reveal underlying user sentiments, needs, and pain points. AI-

driven analytics tools can also track trends and patterns,
Gain a deeper
providing a dynamic view of user behaviors and preferences.
understanding of user
Search The Search phase can be enhanced by AI's ability to sift through
needs and experiences
and make sense of extensive datasets that would be
through AI.
unmanageable for humans alone, leading to a more insightful
and data-driven understanding of the target audience.

Synthesize insights to Synthesis of complex data into actionable insights by identifying

articulate a clear, common themes, contradictions, and gaps in the information
Synthesize concise problem collected during the Search phase. Tools like text summarization
statement with AI's can help condense large volumes of qualitative data into
help. manageable summaries that highlight key insights.

Enhance creative Generate a wide range of ideas and solutions based on the data
ideation with AI- inputs the AI tool has received. In this stage can create new
Ideate generated ideas, scenarios, or alternative perspectives that challenge
ensuring a diverse conventional thinking and creativity. By feeding AI systems with
range of solutions. the problem statement and relevant data, teams can use AI as a
brainstorming partner that offers suggestions beyond human

39
 cognitive biases and experience limitations, ensuring a broader
exploration of potential solutions.

AI-driven tools can quickly generate prototype variations based

Streamline the on specified parameters, allowing users to explore multiple
prototyping process iterations accelerating the prototyping stage. Generative design
with AI-generated software can propose optimal shapes, materials, and
Tuning
materials and tune the configurations based on functional requirements and
prototypes of the constraints. Additionally, AI can simulate user interactions with
generated ideas. prototypes, providing preliminary feedback, thus tuning the
prototypes more efficiently for user trials.

While AI's capability to fully understand and evaluate human

Explore AI's potential to
emotions and nuanced feedback is limited, it can still play a
structure and analyze
supportive role in structuring and analyzing qualitative data
qualitative feedback,
collected during user tests. AI tools can categorize feedback into
recognizing its current
Test themes, track sentiment over different prototype iterations,
limitations in capturing
and identify patterns in user responses that might indicate areas
the nuanced demands
for improvement. The use of AI in this phase should be
of user testing.
complemented by human analysis to interpret the subtleties
and complexities of user feedback.

5.4.2.1. Framework Adaptability

The flexibility and adaptability of the framework allow it to be customized to fit a wide
array of project scopes, creative challenges, and industry requirements. This adaptability
means that teams can scale the integration of AI tools like ChatGPT ensuring that the
technology serves the project's unique needs without overshadowing the human
elements of creativity.

5.4.2.2. Responsible AI Use

The responsible AI use within the framework addresses the critical need for ethical
considerations in AI integration, including user privacy, bias mitigation, and inclusivity
(Bahrini et al., 2023). This includes employing AI in ways that respect user data and
actively seeking to identify and reduce biases in AI-generated insights. Furthermore, the
data privacy policy is a very important issue in the business sector, as data protection
and confidentiality must be guaranteed whenever required in the interaction processes

40
with the AI tool. Thus, we guarantee that a conscious and responsible approach to the
use and integration of this tool is taken.

Figure 8 - AI-Driven Design Thinking Framework

5.4.3. Framework Limitations

While the AI-Driven Design Thinking Framework represents a significant stride toward
integrating advanced AI capabilities into creative processes, it's important to
acknowledge and critically examine its limitations. And one of the primary limitations is
that the framework has not been empirically tested. Its effectiveness and impact are,
therefore, speculative at this stage, based on theoretical integration and anticipated
outcomes. The absence of real-world testing means that the practical challenges of
implementing the framework, especially in diverse and dynamic creative environments,
remain unexplored. Future iterations of the framework will benefit greatly from pilot
studies and case analyses that provide empirical data. Another significant concern is the
inherent limitations and biases within ChatGPT itself considering that is susceptible to
biases present in its training data. These biases can lead to inaccuracies in information,
potentially skewing user insights, problem definitions, and ideation processes.

41
Recognizing and mitigating these biases is crucial, especially in the Empathize and Define
stages, where understanding user needs and challenges is paramount.
Furthermore, the most advanced and complete versions of ChatGPT are not freely
accessible, presenting cost barriers that may limit the widespread adoption of the
framework, especially among smaller agencies or individual practitioners. This limitation
restricts the democratization of advanced ChatGPT tools and updates within the creative
industry, potentially creating disparities in who can leverage these technologies to
enhance their design thinking processes.

42
6. CONCLUSIONS AND FUTURE WORKS

The integration of ChatGPT into the Design Thinking process symbolizes a digital
evolution in advertising agencies, presenting a mix of opportunities and challenges in
the process of creativity and problem-solving. It provokes a re-evaluation of traditional
creative methodologies, promoting a collaborative synergy between human ingenuity
and artificial intelligence. This partnership promises to simplify workflows and catalyze
innovative creative results, depending on whether agencies skillfully manage the
balance between creativity and the influence of AI. However, a gap was identified in the
study in the use of ChatGPT during the Design Thinking test phase. This gap underlines
a broader discussion about the limitations and challenges of applying generative AI in
areas that require deep human empathy and differentiated typical human
understanding. The absence of insights and consideration from the study participants
about the integration of ChatGPT here signals a crucial avenue for future research, to
develop AI-assisted testing methodologies. These methodologies could potentially
bridge the current divide by simulating user interactions, analyzing feedback at scale and
discovering patterns in user responses.

The journey to integrate GenAI into Design Thinking doesn't end with recognizing its
benefits and limitations. Instead, it marks the beginning of an ongoing iterative process
that encourages collaboration, experimentation and a multidisciplinary approach to AI
development. This approach must take into account the constant evolution and
updating of the technology, and the need to continuously review the process and
framework. In addition, it is important to pay attention to ethical issues and the
commitment to mitigating bias and inaccuracies in information. This framework,
although promising, has room for improvement and updates that should be considered
after its first tests and practical applications. The suggestion for future work is to test
this framework to validate the increase in work efficiency, improvement in creativity and
problem management.

43
Considering that this tool has a free version with more limited functionalities, there may
be challenges to the accessibility of this tool, derived from the associated cost, which
ultimately does not advocate equitable access to AI technologies. So, while the AI-Driven
Design Thinking Framework offers a promising path forward, it is crucial to navigate its
implementation with a critical perspective on these challenges, setting the stage for a
future in which AI and human creativity merge to redefine the boundaries of design
thinking.

44
BIBLIOGRAPHICAL REFERENCES

Abukmeil, M., Ferrari, S., Genovese, A., Piuri, V., & Scotti, F. (2021). A Survey of
Unsupervised Generative Models for Exploratory Data Analysis and Representation
Learning. In ACM Computing Surveys (Vol. 54, Issue 5). Association for Computing
Machinery. https://doi.org/10.1145/3450963

Altman, M., Huang, T. T. K., & Breland, J. Y. (2018). Design thinking in health care.
Preventing Chronic Disease, 15(9). https://doi.org/10.5888/pcd15.180128

Aydın, Ö., & Karaarslan, E. (2023). Is ChatGPT Leading Generative AI? What is Beyond
Expectations? SSRN Electronic Journal. https://doi.org/10.2139/ssrn.4341500

Bahrini, A., Khamoshifar, M., Abbasimehr, H., Riggs, R. J., Esmaeili, M., Majdabadkohne,
R. M., & Pasehvar, M. (2023). ChatGPT: Applications, Opportunities, and Threats.
IEEE Systems and Information Engineering Design Symposium.
http://arxiv.org/abs/2304.09103

Beckman, S. L. (2020). To Frame or Reframe: Where Might Design Thinking Research Go

Next? California Management Review, 62(2), 144–162.
https://doi.org/10.1177/0008125620906620

Brown, T. (2008). Design Thinking. Harvard Business Review. www.hbr.org

Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A.,
Shyam, P., Sastry, G., Askell, A., Agarwal, S., Herbert-Voss, A., Krueger, G.,
Henighan, T., Child, R., Ramesh, A., Ziegler, D. M., Wu, J., Winter, C., … Amodei, D.
(2020). Language Models are Few-Shot Learners. http://arxiv.org/abs/2005.14165

Bughin, J., Brussels, |, Hazan, E., Paris, |, Ramaswamy, S., Washington, |, Chui, M.,
Francisco, S., Allas, T., & London, |. (2017). ARTIFICIAL INTELLIGENCE THE NEXT
DIGITAL FRONTIER? In McKinsey Global Institute. McKinsey Global Institute.
www.mckinsey.com/mgi

Camacho, M. (2016). David Kelley: From Design to Design Thinking at Stanford and IDEO.
In She Ji (Vol. 2, Issue 1, pp. 88–101). Tongji University Press.
https://doi.org/10.1016/j.sheji.2016.01.009

Carlgren, L., Rauth, I., & Elmquist, M. (2016). Framing Design Thinking: The Concept in
Idea and Enactment. Creativity and Innovation Management, 25(1), 38–57.
https://doi.org/10.1111/caim.12153

Chui, M. (2023, September). The state of AI in 2023: Generative AI’s breakout year.
McKinsey & Company.

45
Clark, R. M., Stabryla, L. M., & Gilbertson, L. M. (2020). Sustainability coursework:
student perspectives and reflections on design thinking. International Journal of
Sustainability in Higher Education, 21(3), 593–611. https://doi.org/10.1108/IJSHE-
09-2019-0275

Cooper, G. (2023). Examining Science Education in ChatGPT: An Exploratory Study of

Generative Artificial Intelligence. Journal of Science Education and Technology,
32(3), 444–452. https://doi.org/10.1007/s10956-023-10039-y

Dam, R., & Siang, T. (2022). What is Design Thinking and Why Is It So Popular? Interaction
Design Foundation - IxDF. https://www.interaction-
design.org/literature/article/what-is-design-thinking-and-why-is-it-so-popular

Dell’Era, C., Magistretti, S., Cautela, C., Verganti, R., & Zurlo, F. (2020). Four kinds of
design thinking: From ideating to making, engaging, and criticizing. Creativity and
Innovation Management, 29(2), 324–344. https://doi.org/10.1111/caim.12353

Elsbach, K. D., & Stigliani, I. (2018). Design Thinking and Organizational Culture: A Review
and Framework for Future Research. Journal of Management, 44(6), 2274–2306.
https://doi.org/10.1177/0149206317744252

Fattepure, S. (2023, August 5). Beyond Conversations: Using ChatGPT to Generate 80%
More Creative Ideas. Medium. https://medium.com/@Future-AI/beyond-
conversations-using-chatgpt-to-generate-80-more-creative-ideas-d98c46181b35

Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville,
A., & Bengio, Y. (2020). Generative adversarial networks. Communications of the
ACM, 63(11), 139–144. https://doi.org/10.1145/3422622

Gordijn, B., & Have, H. ten. (2023). ChatGPT: evolution or revolution? In Medicine,
Health Care and Philosophy (Vol. 26, Issue 1, pp. 1–2). Springer Science and
Business Media B.V. https://doi.org/10.1007/s11019-023-10136-0

Grassini, S. (2023). Shaping the Future of Education: Exploring the Potential and
Consequences of AI and ChatGPT in Educational Settings. In Education Sciences
(Vol. 13, Issue 7). Multidisciplinary Digital Publishing Institute (MDPI).
https://doi.org/10.3390/educsci13070692

Haleem, A., Javaid, M., & Singh, R. P. (2022). An era of ChatGPT as a significant futuristic
support tool: A study on features, abilities, and challenges. BenchCouncil
Transactions on Benchmarks, Standards and Evaluations, 2(4), 100089.
https://doi.org/10.1016/j.tbench.2023.100089

Hirschberg, J., & Manning, C. D. (2015). Advances in natural language processing.

https://doi.org/DOI: 10.1126/science.aaa868

46
Johansson-Sköldberg, U., Woodilla, J., & Çetinkaya, M. (2013). Design thinking: Past,
present and possible futures. Creativity and Innovation Management, 22(2), 121–
146. https://doi.org/10.1111/caim.12023

Jovanovic, M., & Campbell, M. (2022). Generative Artificial Intelligence: Trends and
Prospects. EEE Computer Society, 55(10), 107–112.
https://doi.org/10.1109/MC.2022.3192720

Kanbach, D. K., Heiduk, L., Blueher, G., Schreiter, M., & Lahmann, A. (2023). The GenAI
is out of the bottle: generative artificial intelligence from a business model
innovation perspective. In Review of Managerial Science. Springer Science and
Business Media Deutschland GmbH. https://doi.org/10.1007/s11846-023-00696-z

Kasneci, E., Sessler, K., Küchemann, S., Bannert, M., Dementieva, D., Fischer, F., Gasser,
U., Groh, G., Günnemann, S., Hüllermeier, E., Krusche, S., Kutyniok, G., Michaeli, T.,
Nerdel, C., Pfeffer, J., Poquet, O., Sailer, M., Schmidt, A., Seidel, T., … Kasneci, G.
(2023). ChatGPT for good? On opportunities and challenges of large language
models for education. In Learning and Individual Differences (Vol. 103). Elsevier Ltd.
https://doi.org/10.1016/j.lindif.2023.102274

Korzynski, P., Mazurek, G., Altmann, A., Ejdys, J., Kazlauskaite, R., Paliszkiewicz, J., Wach,
K., & Ziemba, E. (2023). Generative artificial intelligence as a new context for
management theories: analysis of ChatGPT. Central European Management
Journal, 31(1), 3–13. https://doi.org/10.1108/CEMJ-02-2023-0091

Kuczmarski, T. D. (2003). What is innovation? And why aren’t companies doing more of
it? Journal of Consumer Marketing, 20(6), 536–541.
https://doi.org/10.1108/07363760310499110

Lee, J., Davari, H., Singh, J., & Pandhare, V. (2018). Industrial Artificial Intelligence for
industry 4.0-based manufacturing systems. Manufacturing Letters, 18, 20–23.
https://doi.org/10.1016/j.mfglet.2018.09.002

Liedtka, J. (2014). Perspective: Linking Design Thinking with Innovation Outcomes

through Cognitive Bias Reduction. In Journal of Product Innovation Management
(Vol. 32, Issue 6, pp. 925–938). https://doi.org/10.1111/jpim.12163

Lim, S., Kim, M., & Sawng, Y. W. (2022). Design Thinking for Public R&D: Focus on R&D
Performance at Public Research Institutes. Sustainability (Switzerland), 14(13).
https://doi.org/10.3390/su14137765

Liu, Y., Yang, Z., Yu, Z., Liu, Z., Liu, D., Lin, H., Li, M., Ma, S., Avdeev, M., & Shi, S. (2023).
Generative artificial intelligence and its applications in materials science: Current
situation and future perspectives. In Journal of Materiomics (Vol. 9, Issue 4, pp.
798–816). Chinese Ceramic Society. https://doi.org/10.1016/j.jmat.2023.05.001

47
Lund, B. D., Wang, T., Reddy Mannuru, N., Nie, B., Shimray, S., & Wang, Z. (2023).
ChatGPT and a New Academic Reality: AI-Written Research Papers and the Ethics
of the Large Language Models in Scholarly Publishing. In Journal of the Association
for Information Science and Technology.

Mahmoud-Jouini, S. B., Midler, C., & Silberzahn, P. (2016). Contributions of Design

Thinking to Project Management in an Innovation Context. Project Management
Journal, 47, 144–156. https://doi.org/https://doi.org/10.1002/pmj.21577

McKinsey & Company. (2023). As organizations rapidly deploy generative AI tools,

survey respondents expect significant effects on their industries and workforces
The state of AI in 2023: Generative AI’s breakout year. McKinsey & Company.
https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-
ai-in-2023-generative-AIs-breakout-year

Micheli, P., Wilner, S. J. S., Bhatti, S. H., Mura, M., & Beverland, M. B. (2019). Doing
Design Thinking: Conceptual Review, Synthesis, and Research Agenda. Journal of
Product Innovation Management, 36(2), 124–148.
https://doi.org/10.1111/jpim.12466

OpenAI. (2022, November 30). Introducing ChatGPT. OpenAI.

https://openai.com/blog/chatgpt

Peter G. Rowe. (1987). Design Thinking . MIT Press, Cambridge, Mass.

https://books.google.pt/books?id=ZjZ3mflzJtUC&printsec=frontcover&source=gb
s_ViewAPI&redir_esc=y#v=onepage&q&f=false

Pratomo, L. C., Siswandari, & Wardani, D. K. (2021). The effectiveness of design thinking
in improving student creativity skills and entrepreneurial alertness. International
Journal of Instruction, 14(4), 695–712. https://doi.org/10.29333/iji.2021.14440a

Qu, S. Q., & Dumay, J. (2011). The qualitative research interview. In Qualitative Research
in Accounting and Management (Vol. 8, Issue 3, pp. 238–264). Emerald Group
Publishing Ltd. https://doi.org/10.1108/11766091111162070

Roller, M. R., & Lavrakas, P. J. (2015). Applied qualitative research design: A total quality
framework approach. In Applied qualitative research design: A total quality
framework approach. The Guilford Press.

Roth, K., Globocnik, D., Rau, C., & Neyer, A. K. (2020). Living up to the expectations: The
effect of design thinking on project success. Creativity and Innovation
Management, 29(4), 667–684. https://doi.org/10.1111/caim.12408
Samuel, J., Rostami, M., Bagci, U., & Sigfrids, A. (2023). Human-centricity in AI
governance: A systemic approach. Frontiers in Artificial Intelligence.
https://doi.org/10.3389/frai.2023.976887

48
Seidel, V. P., & Fixson, S. K. (2013). Adopting design thinking in novice multidisciplinary
teams: The application and limits of design methods and reflexive practices. Journal
of Product Innovation Management, 30(SUPPL 1), 19–33.
https://doi.org/10.1111/jpim.12061

Shaji George, A., Hovan George, A., & Martin, Asg. (2023). A Review of ChatGPT AI’s
Impact on Several Business Sectors. Partners Universal International Innovation
Journal (PUIIJ) . https://doi.org/10.5281/zenodo.7644359

Stanford d.school. (2018). The Design Thinking Bootleg. Stanford University.

https://dschool.stanford.edu/resources/design-thinking-bootleg

Sun, S., Luo, C., & Chen, J. (2017). A review of natural language processing techniques
for opinion mining systems. Information Fusion, 36, 10–25.
https://doi.org/10.1016/j.inffus.2016.10.004

Thomas, L. (2023, June 22). Stratified Sampling | Definition, Guide & Examples. Scribbr.
https://www.scribbr.com/methodology/stratified-sampling

Verganti, R., Vendraminelli, L., & Iansiti, M. (2020). Innovation and Design in the Age of
Artificial Intelligence. Journal of Product Innovation Management, 37(3), 212–227.
https://doi.org/10.1111/jpim.12523

Wang, K., Gou, C., Duan, Y., Lin, Y., Zheng, X., & Wang, F. Y. (2017). Generative
adversarial networks: Introduction and outlook. In IEEE/CAA Journal of Automatica
Sinica (Vol. 4, Issue 4, pp. 588–598). Institute of Electrical and Electronics Engineers
Inc. https://doi.org/10.1109/JAS.2017.7510583

Zhang, C., & Lu, Y. (2021). Study on artificial intelligence: The state of the art and future
prospects. Journal of Industrial Information Integration, 23.
https://doi.org/10.1016/j.jii.2021.100224

49
APPENDIX A: QUESTIONNAIRE INVITATION

Espero que este e-mail o encontre bem. O meu nome é Catarina Casquinha e estou
atualmente a desenvolver uma tese como parte integrante do meu Mestrado na NOVA
IMS e, por isso, estou a entrar em contacto para pedir a sua perspetiva para este estudo
que visa explorar o impacto do ChatGPT no processo de Design Thinking,
especificamente na indústria publicitária.

O processo de Design Thinking tem sido muito utilizado nos últimos anos como princípio
base para o desenvolvimento de criatividade, inovação e até de soluções para
problemas, fornecendo um framework estruturado, mas flexível, para o pensamento
criativo e a inovação (em anexo está o framework exemplificativo). O processo baseia-
se em 5 fases principais:
1. Empatia- É uma fase de conhecimento sobre as necessidades e desejos do target,
usando a empatia para ver, sentir e experienciar como se fossemos essas pessoas. Assim
garantimos conhecer da melhor forma possível o contexto do público-alvo.
2. Definição- Esta fase trata-se de identificar o problema do target de forma clara e
objetiva.
3. Idealização- Altura em que se gera o máximo de soluções possíveis para o problema,
sempre com base nas necessidades apresentadas. Esta é a fase da criatividade pura, de
pensar fora da caixa.
4. Protótipo- Esta etapa serve para colocar as ideias em prática. De acordo com a
Stanford d.school, “um protótipo pode ser qualquer coisa com a qual o utilizador pode
interagir – desde uma parede de post-its, até um dispositivo, uma atividade de role-
playing ou uma história em banda desenhada.”
5. Testes- A última fase é a fase de testes. Aqui apresentamos os protótipos criados ao
cliente. Esta etapa do Design Thinking serve para melhorar as ideias e soluções.

Nos últimos anos, houve uma evolução no campo da Inteligência Artificial com
aparecimento de ferramentas de acesso livre, como o ChatGPT, que servem como
recursos auxiliares com potencial de melhorar significativamente o processo criativo. A
minha tese visa aprofundar esta interação dinâmica entre a IA e o Design Thinking e
procura especificamente entender como profissionais estão a integrar o ChatGPT nos
seus workflows assim como os potenciais benefícios e desafios da sua integração no
processo de Design Thinking, e o seu impacto geral na criatividade e resolução de
problemas.
A sua experiência e especialização na área da publicidade vão ser muito úteis para esta
pesquisa e ficaria muito agradecida se pudesse dispensar algum tempo para completar
um breve questionário, que não deverá levar mais de 15 minutos. As suas respostas
serão fundamentais para moldar uma compreensão abrangente do futuro destas
ferramentas de IA na publicidade, particularmente dentro do framework de Design
Thinking.
As suas respostas serão mantidas em confidencialidade e utilizadas apenas para fins
académicos. Agradecia se pudesse completar o questionário até ao final desta semana.

50
APPENDIX B: QUESTIONNAIRES

Questionnaire 1
Job: Art Director/Designer
Agency ID: AGEN1

Q1: Can you describe how you have integrated ChatGPT into your professional
workflow?
O uso do ChatGPT enquanto Director de Arte tem-me servido para a concepção de
textos explicativos e/ou raciocínios para melhor descrever um projecto ou ideia. Para
conseguir explicar uma ideia, normalmente costumo estruturar em 3 parâmetros:
Insight, Ideia e Resultado. Ao usar o ChatGPT consigo sintetizar de uma forma simples e
rápida, aquilo que teria de pensar e consequentemente levar mais tempo a fazer.
Q2: Considering what you described, where you find ChatGPT most beneficial?
Para mim ChatGPT é benéfico no momento da conceptualização da ideia e também em
brainstorm, muitas vezes para despistar se é ou não viável determinada ideia.
Q3: In which specific stages of the Design Thinking process do you see the integration
of ChatGPT being most beneficial?
Penso que o ChatGPT é bastante eficiente para a materialização de textos explicativos
em determinados projectos, uma vez que o resultado em si, é uma boa base para se
trabalhar em cima. Não acho que seja melhor na sugestão de ideias criativas em si, em
comparação a um humano, mas na questão na concepção de textos é bastante mais
rápido e eficiente.
Q4: What potential benefits do you foresee from incorporating ChatGPT-4 into Design
Thinking?
Optimização do tempo que levamos na criação de textos explicativos em determinados
projectos. Ao optimizar esse tempo, temos muito mais tempo para a parte criativa e da
real concepção do projecto em sim.
Q5: What challenges or limitations do you anticipate might arise from integrating
ChatGPT into Design Thinking?
A nível de motor de busca, para responder a determinadas questões, acaba por ser
limitado, visto o ChatGPT ter somente uma base de informação que só vai até ao ano
2021. Os filtros em si, também acabam por ser um pouco limitadores, visto que existem
várias restrições sobre a busca de resultados de informação de determinados
conteúdos/temas. Penso que tal como o google, deveria existir uma opção que pudesse
não ser desbloqueada, e não estar sujeita as filtros restritivos por definição.
Q6: In what ways do you think ChatGPT can enhance creativity and problem-solving
within Design Thinking?
Penso que a integração desta ferramenta a nível criativo pode ajudar na procura de
insight, e informação. A nível de tentar encontrar alguma resposta ou resultado criativo,
penso que a ferramenta em si ainda terá que aprender e melhorar bastante.

Questionnaire 2
Job: Account Manager
Agency ID: AGEN1

51
Q1: Can you describe how you have integrated ChatGPT into your professional
workflow?
I've integrated ChatGPT as a collaborative tool in my professional workflow, shaping my
thinking approach. But just as a help to mostly start the process.
Essentially, ChatGPT enhances my workflow by fostering my thinking process.
Q2: Considering what you described, where you find ChatGPT most beneficial?
In the initial strategy formulation phase. It aids in ideation, generating some insights and
creative concepts that serve as a valuable resource for my strategic planning.
Q3: In which specific stages of the Design Thinking process do you see the integration
of ChatGPT being most beneficial?
During "Ideate," offers subtle help, providing some general insights for market and
competition research. But just in a way for me to start and go into more in-depth
research. It helps me organize my thoughts. In the "Prototype" stage as well - in strategy
planning. A discreet integration ensures a more streamlined process, offering valuable
input without overshadowing the creative brief writing and strategic planning aspects
of my role.
Q4: What potential benefits do you foresee from incorporating ChatGPT-4 into Design
Thinking?
While ChatGPT-4 aids in the creative/problem-solving process, my approach is a
synthesis of external insights and personal knowledge. It serves as a helpful
complement, but the final decisions are based on my own thinking and research.
Q5: What challenges or limitations do you anticipate might arise from integrating
ChatGPT into Design Thinking?
Integrating ChatGPT into Design Thinking may present challenges like over-reliance on
AI-generated content and potential biases.
Q6: In what ways do you think ChatGPT can enhance creativity and problem-solving
within Design Thinking?
It aids in the Design Thinking process, subtly contributing to problem-solving while
avoiding over-reliance on AI-generated content and potential biases.

Questionnaire 3
Job: Copywriter
Agency ID: AGEN1

Q1: Can you describe how you have integrated ChatGPT into your professional
workflow?
Considero que fui, de certo modo, uma early adopter do ChatGPT, e a sua função no
meu método de trabalho tem se mantido muito semelhante. É uma ferramenta útil para
uma fase embrionária de qualquer ideia. Ajuda-me, especialmente a desbloquear outros
pontos de vista ou a dar "pistas" para outro caminho. No entanto, sinto que isso só
acontece com alguma interpretação. Por exemplo, se lanço um prompt como "Sugere
ideias para um headline sobre x", é muito raro que o output seja bom o suficiente para
escolher uma das hipóteses, no entanto, o que eu acabo por produzir pode ir buscar
alguma inspiração às sugestões.

52
Q2: Considering what you described, where you find ChatGPT most beneficial?
Considero mais benéfico na parte de brainstorming. Como expliquei antes é útil para
gerar muitas ideia, para analisá-las e perceber se há alguma que possa ser aproveitada
ou se pode servir como ponto de partida para outras ideias.
Q3: In which specific stages of the Design Thinking process do you see the integration
of ChatGPT being most beneficial?
Considerando o método de design thinking, penso que pode ser mais útil na fase de
idealização, mas quase como uma pré-idealização. Como um primeiro passo de
brainstorming. Os outputs acabam por ajudar-me a desbloquear outras ideias.
Q4: What potential benefits do you foresee from incorporating ChatGPT-4 into Design
Thinking?
Considerando que o ChatGPT-4 é uma ferramenta mais fiável e "desenvolvida" pode ser
mais útil no processo criativo no sentido em que pode gerar respostas que se adaptam
melhor aos prompts. A versão gratuita do ChatGPT, de certo modo, não é uma
ferramenta de criatividade, a versão 4 pode aproximar-se mais disso.
Q5: What challenges or limitations do you anticipate might arise from integrating
ChatGPT into Design Thinking?
Sinto que, mais do que desafios, o ChatGPT ainda tem muitas limitações. Pode ser uma
ferramenta útil para desbloquear caminhos e até para procurar alguma informação
básica, e nesse sentido, pode poupar algum tempo, mas não acho que tenha o potencial
para cumprir, autonomamente, o processo completo.
Q6: In what ways do you think ChatGPT can enhance creativity and problem-solving
within Design Thinking?
É útil no processo de brainstorming e pesquisa e, nesse sentido, pode ajudar a encurtar
certas partes do processo. No entanto, penso que não é igualmente útil para todas as
tarefas criativas, depende da complexidade do pedido ou do tema.
E apesar de ainda não ser uma ferramenta perfeita, prefiro sempre usar, mesmo que os
resultados que me dá não sejam úteis.

Questionnaire 4
Job: Copywriter
Agency ID: AGEN2

Q1: Can you describe how you have integrated ChatGPT into your professional
workflow?
Initially, it was suggested to me by the Creative Director himself. We began to use it and
then it rapidly became a daily use tool.
Q2: Considering what you described, where you find ChatGPT most beneficial?
It has become a support tool for the initial brainstorm and for quickly completing some
micro-writing tasks (emails, letters, summary paragraphs). It always needs a touch of
human revision on my part, but it's a useful ally.
Q3: In which specific stages of the Design Thinking process do you see the integration
of ChatGPT being most beneficial?
Empathize and Define stages.

53
Q4: What potential benefits do you foresee from incorporating ChatGPT-4 into Design
Thinking?
GPT Chat can detect inconsistencies and predict problems in our line of thought, and is
an important assistant when it comes to building ideas that are creative but strategically
functional.
Q5: What challenges or limitations do you anticipate might arise from integrating
ChatGPT into Design Thinking?
The vast facilities offered by Chat GPT could lead to a certain inertia in the process of
conceiving ideas, generating a scenario of very similar campaigns with little degree of
differentiation.
Q6: In what ways do you think ChatGPT can enhance creativity and problem-solving
within Design Thinking?
It's hard to get an idea. But it's even harder to choose the best execution. AI is very useful
when it comes to suggesting practical and realistic executions for a bold and original
idea.

Questionnaire 5
Job: Account manager
Agency ID: AGEN2

Q1: Can you describe how you have integrated ChatGPT into your professional
workflow?
Summarizing information, look for specific information on a given subject
Q2: Considering what you described, where you find ChatGPT most beneficial?
Information gathering.
Q3: In which specific stages of the Design Thinking process do you see the integration
of ChatGPT being most beneficial?
Enphatize and define.
Q4: What potential benefits do you foresee from incorporating ChatGPT-4 into Design
Thinking ?
Gathering knowledge and assesing potential solutions.
Q5: What challenges or limitations do you anticipate might arise from integrating
ChatGPT into Design Thinking?
Information breaches (related with data security and rgpd), asking the right questions
and filtering information
Q6: In what ways do you think ChatGPT can enhance creativity and problem-solving
within Design Thinking?
Little impact.

Questionnaire 6
Job: Designer
Agency ID: AGEN2

Q1: Can you describe how you have integrated ChatGPT into your professional
workflow?

54
Enquanto Designer, o ChatGPT é uma ferramenta complementar no meu processo
criativo. Geralmente, utilizo para desenvolver copies ou fazer brainstormings de forma
rápida e consistente.
Q2: Considering what you described, where you find ChatGPT most beneficial?
O facto de apresentar vários conceitos, economiza o nosso tempo de trabalho e
aumenta a produtividade. Além disso, serve como fonte de inspiração, proporcionando
insights criativos para futuros projetos.
Q3: In which specific stages of the Design Thinking process do you see the integration
of ChatGPT being most beneficial?
Nas etapas iniciais dos projetos. Na Definição, recolhe informação clara e sucinta e
fornece recursos para chegar à melhor solução. Na idealização, ajuda a analisar a
questão em diferentes perspectivas.
Q4: What potential benefits do you foresee from incorporating ChatGPT-4 into Design
Thinking?
Apresenta caminhos mais inovadores, aumenta a diversidade de ideias, facilita a
explicação de conceitos mais complexos e adapta soluções de forma eficaz.
Q5: What challenges or limitations do you anticipate might arise from integrating
ChatGPT into Design Thinking?
Apesar do ChatGPT ser capaz de produzir qualquer tipo de conteúdo, pode limitar a
nossa autonomia e o pensamento mais crítico durante o processo criativo.
Q6: In what ways do you think ChatGPT can enhance creativity and problem-solving
within Design Thinking?
O ChatGPT pode aumentar a criatividade devido à quantidade de informação que
fornece e ao facto de dar feedback imediato sobre as ideias apresentadas.

Questionnaire 7
Job: Designer
Agency ID: AGEN3

Q1: Can you describe how you have integrated ChatGPT into your professional
workflow?
Utilizo como forma de desbloquear o pensamento criativo, através de palavras-chave
ajuda me na fase inicial do projeto como ferramenta de brainstorm
Q2: Considering what you described, where you find ChatGPT most beneficial?
Considero ser uma ferramenta de pesquisa onde a informação se encontra mais
focalizada, através de links e páginas de pesquisa consigo filtrar rapidamente ganhando
tempo
Q3: In which specific stages of the Design Thinking process do you see the integration
of ChatGPT being most beneficial?
Na fase inicial de pesquisa, recolha de informação e brainstorm.
Q4: What potential benefits do you foresee from incorporating ChatGPT-4 into Design
Thinking?
Muitas vezes as pessoas que trabalham na indústria criativa em determinadas fases do
projeto tem falta de inspiração criativa, sendo o chatgpt uma boa ferramenta para
aumentar os inputs

55
Q5: What challenges or limitations do you anticipate might arise from integrating
ChatGPT into Design Thinking?
As respostas sejam pré-formatados podendo qualquer pessoa alcançar o mesmo tipo de
respostas. Dando os mesmos inputs obter se sempre as mesmas respostas.
Por outro lado isso pode já acontecer uma vez que os criativas ontem já hoje inspiração
das mesmas fontes
Q6: In what ways do you think ChatGPT can enhance creativity and problem-solving
within Design Thinking?
Aumenta a produtividade, pesquisa mais direta e desbloqueio da criatividade

Questionnaire 8
JOB: Creative Director/Copywriter
Agency ID: AGEN2

Q1: Can you describe how you have integrated ChatGPT into your professional
workflow?
É uma ferramenta de apoio à pesquisa de insights e informação, servindo, por vezes, de
desbloqueador para encontrar ângulos inesperados em brainstorms. Nunca tomo como
certa a informação obtida. Não confio ainda a esse ponto. Para elaborar textos também
não costumo usar.
Q2: Considering what you described, where you find ChatGPT most beneficial?
Procura de insights e fazer brainstorms.
Q3: Q3: In which specific stages of the Design Thinking process do you see the
integration of ChatGPT being most beneficial?
Empatia e Idealização.
Q4: What potential benefits do you foresee from incorporating ChatGPT-4 into Design
Thinking?
Interação com grandes blocos de informação que surgem de forma orgânica e
espontânea no processo de pesquisa. Não é tão fácil sermos surpreendidos por
informação inesperada numa pesquisa tradicional em browser. Sobretudo isso, diálogo
casual com informação que parece infinita.
Q5: What challenges or limitations do you anticipate might arise from integrating
ChatGPT into Design Thinking?
Robotização de processos. Pouca espaço para poesia e instinto.
Q6: In what ways do you think ChatGPT can enhance creativity and problem-solving
within Design Thinking?
Ir para além do óbvio ao ter acesso a informação que — se o algoritmo não for enviesado
(que talvez o seja) — não é imediata.

Questionnaire 9
Job: Account and events manager
Agency ID: AGEN3

Q1: Can you describe how you have integrated ChatGPT into your professional
workflow?

56
I work in event building and management for a really established company in the
market, therefore a lot of the processes are well-established and structured. However,
since I do have some copywriting to do, as well as a lot of meetings to prepare for these
events, I use ChatGPT not only to get some ideas in terms of potential pitches that I'll
need to do but also to summarize information from these meetings to prepare for the
events. ChatGPT is just a tool I use for getting the most accurate and up-to-date
information or sources of information.
Q2: Considering what you described, where you find ChatGPT most beneficial?
ChatGPT is the most useful tool for stages like planning, structuring, and organizing
processes and information.
Q3: In which specific stages of the Design Thinking process do you see the integration
of ChatGPT being most beneficial?
In the empathize and define stages. Integrating ChatGPT into the Design Thinking
process can significantly enhance the Empathize and Define stages of planning different
scenarios. But also in the ideation stage to brainstorm ideas. In the planning stage,
ChatGPT can be utilized to explore various organizational strategies for product launches
or campaign launches. It can assist in envisioning different scenarios and potential
outcomes, providing a broad perspective on the possible impacts of a product or service
launch.
Q4: What potential benefits do you foresee from incorporating ChatGPT-4 into Design
Thinking?
ChatGPT can be employed to summarize key points and discussions, this not only aids in
creating concise and easy-to-understand notes but also offers considerable time-saving
benefits and serves as a good starting point for gathering information.
Q5: What challenges or limitations do you anticipate might arise from integrating
ChatGPT into Design Thinking?
Some problems that I feel are intrinsically connected with integrating chatgpt into
design thinking are not only the fact that you might be completely bombed with a lot of
non-worthy and accurate information. It might also create some confusion if you're not
100% sure of what you need and sometimes it might cause some chaos.
Q6: In what ways do you think ChatGPT can enhance creativity and problem-solving
within Design Thinking?
In general gathering and summarising information and also to test some ideas and
scenarios.

Questionnaire 10
Job: Copywriter
Agency ID: AGEN3

Q1: Can you describe how you have integrated ChatGPT into your professional
workflow?
Uso o ChatGPT pontualmente quando preciso de desbloquear alguma ideia com a qual
começo a ficar frustrada, para encontrar sinónimos ou palavras da mesma família e para
pesquisas simples que não exijam rigor científico (num processo de brainstorming ele é
muitas vezes mais prático para encontrar insights do que divagar pela SERP do Google).

57
Q2: Considering what you described, where you find ChatGPT most beneficial?
Como ponto de partida para gerar keywords num brainstorming e quando estou presa
num raciocínio e preciso de novas hipóteses ou pontos de vista.
Q3: In which specific stages of the Design Thinking process do you see the integration
of ChatGPT being most beneficial?
Considerando o rigor duvidoso das respostas que a ferramenta apresenta, considero-o
mais útil na fase de criar ideias — aí, ele é um bom aliado para explorar insights e
desconstruir conceitos. Na fase do protótipo, quando o projeto é de escrita criativa, o
ChatGPT também pode ser um bom apoio para encontrar a palavra certa.
Q4: What potential benefits do you foresee from incorporating ChatGPT-4 into Design
Thinking or any other creative/problem-solving process?
Acho que a ferramenta agiliza muitas vezes o processo de pesquisa, além de ser uma
boa forma de dinamizar um brainstorming.
Q5: What challenges or limitations do you anticipate might arise from integrating
ChatGPT into Design Thinking?
A falta de rigor científico das respostas é uma limitação de muito peso, é fácil assumir
uma justificação do ChatGPT como certa e ela não estar, o que faz da ferramenta uma
fonte que tem sempre que ser verificada e que muitas vezes nos pode induzir em erro.
Acho que o seu uso frequente pode ter um impacto negativo no desenvolvimento da
capacidade criativa pessoal. Se o assumirmos como um primeiro passo em qualquer fase
criativa estamos a privar o nosso cérebro de fazer esse processo, o que pode levar à
habituação e a um consequente "atraso" no desenvolvimento da habilidade criativa. É
como quando me passaram uma calculadora para as mãos nas aulas de matemática,
muito útil, mas esqueci-me da tabuada. Ainda, também é fácil ver o cidadão comum a
"abusar" deste recurso para escrever emails, artigos, projetos e outras coisas no seu dia
a dia. Isso faz com que esses pequenos textos não contemplem a personalidade e a
criatividade de quem os está a enviar, o que é uma perda preciosa para as relações
interpessoais.
Q6: In what ways do you think ChatGPT can enhance creativity and problem-solving
within Design Thinking?
Acho que esta ferramenta pode dinamizar o processo criativo e ajudar a explorar ideias.
Questionnaire 11
Job: Account Manager
Agency ID: AGEN1

Q1: Can you describe how you have integrated ChatGPT into your professional
workflow?
Chatgpt is a tool I use daily to simplify processes such as translating text, rewriting emails
or even searching for different ideas for certain topics. I also look for information, but
always check the sources used.
Q2: Considering what you described, where you find ChatGPT most beneficial?
Brainstorm, test some ideas or hypotheses, and synthesise information.
Q3: In which specific stages of the Design Thinking process do you see the integration
of ChatGPT being most beneficial?

58
In the "Empathize" and "Define" phases, ChatGPT can be useful for researching and
gathering information from various sources, as well as synthesizing it. In the "Ideate"
phase the tool can be useful for brainstorming and in the "Prototype" phase it also might
be useful for seeking feedback and alternative ideas.
Q4: What potential benefits do you foresee from incorporating ChatGPT-4 into Design
Thinking or any other creative/problem-solving process?
Faster responses in the process and more easily available information.
Q5: What challenges or limitations do you anticipate might arise from integrating
ChatGPT into Design Thinking?
Potentially wrong information and biased information. Uncreative writing and a great
need for human thought to accompany the information so that it doesn't become
strange or random.
Q6: In what ways do you think ChatGPT can enhance creativity and problem-solving
within Design Thinking?
It facilitates creative and problem-solving processes through rapid assistance and
availability of information, providing feedback on our ideas and finding some solutions
to our problems.

Questionnaire 12
Job: Creative Director
Agency ID: AGEN1

Q1: Can you describe how you have integrated ChatGPT into your professional
workflow?
Como profissional na área da criatividade, a minha utilização do ChatGPT prende-se
essencialmente a três pontos: 1) Em primeiro lugar, na sistematização e pesquisa de
informações, referências e publicações sobre determinado assunto;
2) Em segundo, na prototipação de ideias ou headlines possíveis para campanhas;
3) Por último, na criação de nomes. De salientar que, na verdade, tendo a usar o ChatGPT
como dupla criativa, de forma a abrir novas possibilidades e caminhos. Apesar de
considerar que, em alguns casos, é possível utilizar "ipsis verbis" o que é criado pela
ferramenta, ainda não estamos no ponto em que o GPT (ou qualquer outra ferramenta
de inteligência artificial) consegue substituir o trabalho de um criativo, sobretudo no
que diz respeito à inovação e disrupção.
Contudo, no que toca à performance e trabalho de uma agência de media, aí diria que
sim – é um meio rápido e eficaz de produzir mensagens que, não sendo especialmente
criativas, são funcionais e ajustadas a um target específico. Julgo, também, que a
ferramenta terá sido treinada para este efeito em particular, e que tem mais tendência
para criar copy "gerador de leads" do que pensar em ideias inovadoras.
Q2: Considering what you described, where you find ChatGPT most beneficial?
Como disse anteriormente, diria que é mais beneficial na pesquisa, para resolver
bloqueios, e para procurar formulações alternativas de headlines. Ainda estamos,
contudo, numa fase muito prematura. Seria espetacular se conseguisse produzir PDFs e
racionais criativos já paginados.

59
Q3: In which specific stages of the Design Thinking process do you see the integration
of ChatGPT being most beneficial?
Diria que a problematização estará sempre do lado humano – bem como o skill técnico
para "fazer a pergunta certa".
Mas a integração do ChatGPT no processo pode revelar-se incrivelmente útil das fases
2 - 4, ou seja, ajudando na definição do problema do target, procurando gerar soluções
e prototipando. Contudo, a sua utilidade no processo pode não ser tão linear quanto o
descrito: a fase de prototipagem pode ajudar a clarificar a definição do problema do
target, e a sistematização da pesquisa pode interferir no tipo de soluções propostas para
o problema. Ou seja, na minha opinião, diria que poderá ter um papel mais abrangente,
pelo menos nesta fase.
Uma ferramenta que amplifica o processo como um todo. Falando diretamente em
criatividade, é contudo uma ferramenta limitada, uma vez que ainda se baseia apenas
em texto. Pelo que diria que pode ser usado com um uma espécie de motor de busca
em esteróides, com capacidade de sintetização de informação e produção de algumas
soluções básicas. Mas essa sistematização revela-se extremamente útil porque, por
vezes, torna evidente algo que à partida não o era.
Q4: What potential benefits do you foresee from incorporating ChatGPT-4 into Design
Thinking or any other creative/problem-solving process?
Acho que, à medida que a ferramenta vai melhorando, ela pode de forma mais
deliberada e eficaz contribuir para todo o processo, sobretudo se começar a basear-se
em experiências passadas. Por outro lado, o facto de o modelo de "machine learning"
se basear em datasets predefinidos acaba por criar um "bias" que irá viciar sempre os
resultados. Diria que, por enquanto, as ciências exatas são o campo que mais irá
beneficiar da inteligência artificial, e que a criatividade e a inovação vão precisar sempre
do cunho humano.
Q5: What challenges or limitations do you anticipate might arise from integrating
ChatGPT into Design Thinking?
Se toda a gente usar o ChatGPT para tentar resolver um mesmo problema, e ninguém
construir sobre o assunto, provavelmente chegarão à mesma solução. E o facto de
existirem soluções idênticas para o mesmo problema vai tornar-se, a curto e médio
prazo, no problema em si. Então, a forma de resolver o problema já não passa pelo
caminho mais direto de A - B, mas sim na forma como podemos criar uma "volta maior".
E voltamos à criatividade de novo, que será cada vez mais especializada em fazer as
perguntas certas, mais do que encontrar as respostas. São os meus 50 cêntimos
relativamente ao futuro, pelo menos com base no que vemos atualmente.
Q6: In what ways do you think ChatGPT can enhance creativity and problem-solving
within Design Thinking?
Julgo que a sua principal utilidade passa pela sistematização e pela criação rápida de
protótipos, do ponto de vista da formulação de frases e ideias. Acho que é
extremamente útil integrar o ChatGPT como parte integrante do brainstorm – no fundo,
como mais um local onde é possível partilhar ideias, explorar caminhos e utilizar para
reformular e rever alguns pressupostos que fazem parte de todo o processo.

60
 APPENDIX C: CODEBOOK

CODEBOOK

CODE GROUP CODE CODE DESCRIPTION INTERVIEW QUOTATIONS

"O uso do ChatGPT enquanto Director de Arte tem-me

servido para a concepção de textos explicativos e/ou
raciocínios para melhor descrever um projeto ou ideia."
"For quickly completing some micro-writing tasks (emails,
letters, summary paragraphs)."
"...desenvolver copies..."
Utilizing ChatGPT to "... para elaborar textos também costumo usar."
produce creative "... simplify processes such as translating text, rewriting
Efficient content and emails..."
Copywriting explanatory texts “o ChatGPT é bastante eficiente para a materialização de
rapidly, improving textos explicativos em determinados projectos”
communication and “facilita a explicação de conceitos mais complexos e adapta
project descriptions. soluções de forma eficaz”
“Optimização do tempo que levamos na criação de textos
explicativos em determinados projectos.”
“na concepção de textos é bastante mais rápido e eficiente”
“... headlines possíveis para campanhas...”
“... é um meio rápido e eficaz de produzir mensagens...”
“... procurar formulações alternativas de headlines...”
“É uma ferramenta útil para uma fase embrionária de
qualquer ideia”
Enhancement
“… útil para gerar muitas ideias…”
of Creative
“Considerando que o ChatGPT-4 é uma ferramenta mais
Processes
Employing ChatGPT in fiável e "desenvolvida" pode ser mais útil no processo
the early stages to criativo no sentido em que pode gerar respostas que se
generate and develop adaptam melhor aos prompts.”
Conceptualizat
new ideas, enhancing “… conceptualização da ideia…”
ion of Ideas
the creative process by “… ideation, generating some insights and creative
exploring various concepts...”
possibilities. “…important assistant when it comes to building ideas that
are creative but strategically functional.”
“… apresentar vários conceitos...”
“... procura de insights...”
“... prototipação de ideias...”
“Penso que a integração desta ferramenta a nível criativo
pode ajudar na procura de insigh…”
Using ChatGPT to
“... proporcionando insights criativos...”
gather insights and
“... aumentar a criatividade....”
relevant information,
Creative “... em determinadas fases do projeto tem falta de
aiding in the creative
Insights inspiração criativa, sendo o chatgpt uma boa ferramenta
process and the
para aumentar os inputs.”
exploration of new
“... ferramenta de apoio à pesquisa de insights...”
ideas.
“Não é tão fácil sermos surpreendidos por informação
inesperada numa pesquisa tradicional em browser.”

61
 “Ir para além do óbvio ao ter acesso a informação que — se
o algoritmo não for enviesado (que talvez o seja) — não é
imediata.”
“I use ChatGPT not only to get some ideas...”
“...processo de brainstorming ele é muitas vezes mais
prático para encontrar insights…”
“... ferramenta pode dinamizar o processo criativo e ajudar
a explorar ideias...”
“... na criação de nomes...”
“… é uma boa base para se trabalhar em cima.”
“It serves as a helpful complement…”
Viewing ChatGPT as a
Complementar “help to mostly start the process”
supportive tool that
y Tool for “… important assistant when it comes to building ideas that
complements creative
Creative are creative but strategically functional.”
efforts, enhancing the
Enhancement “… desbloqueio da criatividade…”
overall creative output.
“… ferramenta pode dinamizar o processo criativo e ajudar a
explorar ideias…”
“…brainstorm...”
“... considero mais benéfico na parte de brainstorming...”
“... processo de brainstorming e pesquisa…”
“…support tool for the initial brainstorm…”
“…brainstormings...”
“...ajuda me na fase inicial do projeto como ferramenta de
brainstorm...”
Viewing ChatGPT as a “...fazer brainstorms...”
supportive tool that “...desbloqueador para encontrar ângulos inesperados em
Brainstorming complements creative brainstorms...”
efforts, enhancing the “...brainstorm ideas...”
overall creative output. “...processo de brainstorming ele é muitas vezes mais
prático para encontrar insights.”
“... ponto de partida para gerar keywords num
brainstorming...”
“...boa forma de dinamizar um brainstorming...”
“...brainstorm…”
“… the tool can be useful for brainstorming…”
“… parte integrante do brainstorm…”
“... o que eu acabo por produzir pode ir buscar alguma
inspiração às sugestões.”
“... fonte de inspiração...”
“... em determinadas fases do projeto tem falta de
Using ChatGPT as a
inspiração criativa, sendo o chatgpt uma boa ferramenta
source of inspiration,
Inspiration para aumentar os inputs...”
exploring various
Search “Procura de insights...”
concepts and ideas to
“...quando estou presa num raciocínio e preciso de novas
spark creativity.
hipóteses ou pontos de vista.”
“... abrir novas possibilidades e caminhos...”
“... sistematização e pela criação rápida de protótipos, do
ponto de vista da formulação de frases e ideias.”
Using ChatGPT to “Apresenta caminhos mais inovadores…”
Diversity of
discover new and
ideas
innovative approaches

62
 to creative projects and
problem-solving.

Employing ChatGPT to “… aumenta a diversidade de ideias...”

generate a wide range “... it also might be useful for seeking feedback and
Innovative
of ideas, fostering alternative ideas.”
paths
diversity in creative
thinking and solutions.
"Ao usar o ChatGPT consigo sintetizar de uma forma simples
e rápida, aquilo que teria de pensar e consequentemente
levar mais tempo a fazer."
"While ChatGPT-4 aids in the creative/problem-solving
process, my approach is a synthesis of external insights and
personal knowledge."
“for quickly completing some micro-writing tasks (emails,
letters, summary paragraphs)”
Employing ChatGPT to
“facilita a explicação de conceitos mais complexos e adapta
condense complex
soluções de forma eficaz”
concepts and
“ChatGPT can be employed to summarize key points and
Efficient discussions into
discussions”
Synthesis of simplified, easily
“synthesize information”
Ideas understandable
“as well as synthesizing it”
summaries, thereby
“Summarizing information…”
accelerating the
“…also to summarize information from these meetings to
creative process.
prepare for the events.”
“…gathering and summarising information…”
“… a fase de prototipagem pode ajudar a clarificar a
definição do problema do target, e a sistematização da
Efficiency and pesquisa pode interferir no tipo de soluções propostas para
Workflow o problema.”
Enhancement “... sistematização e pela criação rápida de protótipos, do
ponto de vista da formulação de frases e ideias.”
Integrating ChatGPT to “I've integrated ChatGPT as a collaborative tool in my
enhance collaboration professional workflow, shaping my thinking approach.”
Collaborative
among team members, “... ferramenta complementar no meu processo criativo...”
Workflow
facilitating a more
Efficiency
efficient and dynamic
workflow.
Leveraging ChatGPT for “It helps me organize my thoughts.”
better planning and “ChatGPT is the most useful tool for stages like planning,
Planning and
organizing of creative structuring, and organizing processes and information.”
Organizing
projects, enhancing
Processes
project management
and efficiency.
“A discreet integration ensures a more streamlined process,
Integrating ChatGPT to offering valuable input without overshadowing the creative
streamline creative and brief writing…”
Streamlined
operational processes, “…help to mostly start the process…”
Process
making them more “... facilita a explicação de conceitos mais complexos e
efficient and effective. adapta soluções de forma eficaz...”
“... agiliza muitas vezes o processo de pesquisa...”

63
 “… optimizar esse tempo, temos muito mais tempo para a
parte criativa…”
Utilizing ChatGPT to “...pode poupar algum tempo...”
save time on routine “... pode ajudar a encurtar certas partes do processo...”
Efficiency in
tasks, allowing more “... economiza o nosso tempo...”
Time
focus on creative and “... consigo filtrar rapidamente ganhando tempo...”
Management
strategic activities. “… offers considerable time-saving benefits.”
“Faster responses in the process…”
“Optimização do tempo que levamos na criação de textos
explicativos em determinados projectos.”
Highlighting the rapid “We began to use it and then it rapidly became a daily use
integration of ChatGPT tool.”
Quick
into daily workflows, “Chatgpt is a tool I use daily…”
Adoption and
becoming an essential
Integration
tool for creative and
operational tasks.
“Penso que a integração desta ferramenta a nível criativo
pode ajudar na procura de insight, e informação.”
“procurar alguma informação básica”
“pesquisa mais direta”
“gathering and summarising information”
“ferramenta pode dinamizar o processo criativo e ajudar a
explorar ideias”
“It facilitates creative and problem-solving processes
through rapid assistance and availability of information”
“… look for specific information on a given subject…”
“... ferramenta de apoio à pesquisa de insights e
informação...”
“ChatGPT is just a tool I use for getting the most accurate
and up-to-date information or sources of information.”
Using ChatGPT to
“... para pesquisas simples que não exijam rigor científico...”
Information conduct targeted
“… searching for different ideas for certain topics…”
Search searches for relevant
“...Information gathering.”
information.
“Gathering knowledge...”
“... ferramenta de pesquisa onde a informação se encontra
mais focalizada...”
“ChatGPT is the most useful tool for stages like planning,
structuring, and organizing processes and information.”
“…serves as a good starting point for gathering
information…”
“...agiliza muitas vezes o processo de pesquisa...”
“ChatGPT can be useful for researching and gathering
information from various sources
more easily available information…”
“… sistematização e pesquisa de informações, referências e
publicações sobre determinado assunto...”
“...beneficial na pesquisa...”
Using ChatGPT to “… feedback imediato sobre as ideias apresentadas…”
Instant receive immediate “... providing feedback on our ideas and finding some
feedback feedback on ideas, solutions to our problems.”
helping to refine and

64
 improve creative
concepts.

Highlighting the role of “… aumenta a produtividade”

ChatGPT in enhancing “Aumenta a produtividade…”
Productivity
productivity through
Boost
efficient task execution
and idea generation.
“… quantidade de informação que fornece…”
Highlighting the vast “… pesquisa mais direta…”
amount of easy access “It facilitates creative and problem-solving processes
Accessible to information ChatGPT through rapid assistance and availability of information…”
information can provide, aiding in “… more easily available information…”
research and creative “Interação com grandes blocos de informação que surgem
exploration. de forma orgânica e espontânea no processo de pesquisa.”
“...diálogo casual com informação...”
“… raciocínios para melhor descrever um projecto ou ideia.”
Utilizing ChatGPT to “... é uma boa base para se trabalhar em cima.”
improve the problem- “... ChatGPT enhances my workflow by fostering my thinking
Enhanced
solving process, process.”
Creative
providing new “...ferramenta complementar no meu processo criativo...”
Problem-
perspectives and “... desbloquear o pensamento criativo...”
Solving
solutions to creative “Uso o ChatGPT pontualmente quando preciso de
challenges. desbloquear alguma ideia com a qual começo a ficar
frustrada...”
Applying ChatGPT in “In the initial strategy formulation phase. It aids in ideation,
strategic planning, generating some insights and creative concepts that serve
Strategy using it to generate as a valuable resource for my strategic planning.”
formulation insights, creative “ChatGPT can be utilized to explore various organizational
concepts, and strategic strategies…”
directions.
Strategic Employing ChatGPT to “… subtly contributing to problem-solving…”
Planning and contribute to solving “… providing feedback on our ideas and finding some
Optimizing
Problem- problems more solutions to our problems.
problem-
Solving efficiently, providing
solving
feedback and potential
solutions.
“Ajuda-me, especialmente a desbloquear outros pontos de
vista ou a dar "pistas" para outro caminho.”
Using ChatGPT to
“... desbloqueador para encontrar ângulos inesperados...”
explore different
Unlocking “... quando estou presa num raciocínio e preciso de novas
perspectives and
Alternative hipóteses ou pontos de vista.”
alternatives, enhancing
Viewpoints “…test some ideas or hypotheses…”
creativity and problem-
“… it also might be useful for seeking feedback and
solving.
alternative ideas.”
“… resolver bloqueios…”
Detect and Using ChatGPT to “… muitas vezes para despistar se é ou não viável
predict identify potential issues determinada ideia…”
problems and predict problems “GPT Chat can detect inconsistencies and predict problems
before they arise, in our line of thought…”

65
 aiding in proactive “It can assist in envisioning different scenarios and potential
problem-solving. outcomes, providing a broad perspective on the possible
impacts of a product or service launch.”

“AI is very useful when it comes to suggesting practical and

realistic executions for a bold and original idea.”
Employing ChatGPT to “It can assist in envisioning different scenarios and potential
Practical suggest realistic and outcomes, providing a broad perspective on the possible
Execution actionable steps for impacts of a product or service launch.”
Suggestions implementing creative “…to test some ideas and scenarios.”
ideas and strategies. “… test some ideas or hypotheses…”
“… providing feedback on our ideas and finding some
solutions to our problems.”
“… assessing potential solutions…”
Leveraging ChatGPT to
“It can assist in envisioning different scenarios and potential
explore and evaluate
Potential outcomes, providing a broad perspective on the possible
potential solutions to
Solutions impacts of a product or service launch.”
creative and
“…test some ideas or hypotheses…”
operational challenges.
Acknowledging the “A nível de motor de busca, para responder a determinadas
constraints of ChatGPT questões, acaba por ser limitado...”
as a search tool, “Os filtros em si, também acabam por ser um pouco
Search Engine
including its limitations limitadores, visto que existem várias restrições sobre a
Limitations
in accessing real-time busca de resultados de informação de determinados
or highly specific conteúdos/temas.”
information.
“… challenges like over-reliance on AI-generated content
and potential biases…”
Recognizing and “Nunca tomo como certa a informação obtida. Não confio
addressing the biases ainda a esse ponto.”
Awareness of
that may be present in “... bombed with a lot of non-worthy and accurate
Potential
AI-generated content, information…”
Biases
ensuring a balanced “… biased information…”
and fair perspective. “... o facto de o modelo de "machine learning" se basear em
Challenges and datasets predefinidos acaba por criar um "bias" que irá
Limitations viciar sempre os resultados…”
“Não acho que seja melhor na sugestão de ideias criativas…”
“A nível de tentar encontrar alguma resposta ou resultado
criativo, penso que a ferramenta em si ainda terá que
aprender e melhorar bastante.”
“… é muito raro que o output seja bom o suficiente para
escolher uma das hipóteses…”
Recognizing
ChatGPT's lack of “A versão gratuita do ChatGPT, de certo modo, não é uma
Creative
capabilities for creative ferramenta de criatividade
Boundaries of
responses. não acho que tenha o potencial para cumprir,
AI
autonomamente, o processo completo.”
“... penso que não é igualmente útil para todas as tarefas
criativas, depende da complexidade do pedido ou do
tema...”
“Pouca espaço para poesia e instinto.”
“Uncreative writing…”

66
 “Apesar de considerar que, em alguns casos, é possível
utilizar "ipsis verbis" o que é criado pela ferramenta, ainda
não estamos no ponto em que o GPT (ou qualquer outra
ferramenta de inteligência artificial) consegue substituir o
trabalho de um criativo, sobretudo no que diz respeito à
inovação e disrupção.”
“Falando diretamente em criatividade, é contudo uma
ferramenta limitada, uma vez que ainda se baseia apenas
em texto.”
“… a criatividade e a inovação vão precisar sempre do cunho
humano.”
“… final decisions are based on my own thinking and
research…”
“It always needs a touch of human revision on my part…”
Emphasizing the “… asking the right questions...”
importance of human “... faz da ferramenta uma fonte que tem sempre que ser
Necessity of review and intervention verificada e que muitas vezes nos pode induzir em erro.”
Human in refining and “… great need for human thought to accompany the
Oversight validating AI-generated information so that it doesn't become strange or random.”
content. “Diria que a problematização estará sempre do lado
humano – bem como o skill técnico para "fazer a pergunta
certa"”
“... a criatividade e a inovação vão precisar sempre do
cunho humano.”
Recognizing the “… information breaches (related with data security and
importance of data rgpd)”
Data Security
security and privacy in
the use of AI tools like
ChatGPT.
Addressing the “… limitar a nossa autonomia...”
limitations of ChatGPT “... Dando os mesmos inputs obter se sempre as mesmas
Autonomy in fostering autonomy respostas.”
Limitations in creative processes,
due to its reliance on
input.
Discussing the potential “... pensamento mais crítico durante o processo criativo...”
impact of frequent AI “It might also create some confusion if you're not 100% sure
Limitations on
use on the of what you need and sometimes it might cause some
Critical
development and chaos.”
Thinking
exercise of critical “...uso frequente pode ter um impacto negativo no
thinking skills. desenvolvimento da capacidade criativa pessoal...”
Examining the impact “Robotização de processos…”
of automating routine “Se toda a gente usar o ChatGPT para tentar resolver um
Robotization tasks with ChatGPT, mesmo problema, e ninguém construir sobre o assunto,
of processes focusing on efficiency provavelmente chegarão à mesma solução. E o facto de
and potential existirem soluções idênticas para o mesmo problema vai
drawbacks. tornar-se, a curto e médio prazo, no problema em si.”
Risk and The consequences of “Se o assumirmos como um primeiro passo em qualquer
Impact of excessive reliance on AI fase criativa estamos a privar o nosso cérebro de fazer esse
Overreliance for creative tasks and processo, o que pode levar à habituação e a um
on AI Tools routine operations, consequente "atraso" no desenvolvimento da habilidade
caution against the criativa.”

67
 potential for reduced “Ainda, também é fácil ver o cidadão comum a "abusar"
creative skill deste recurso para escrever emails, artigos, projetos e
development and outras coisas no seu dia a dia.”
critical thinking. “Isso faz com que esses pequenos textos não contemplem a
personalidade e a criatividade de quem os está a enviar, o
que é uma perda preciosa para as relações interpessoais.”
“...visto o ChatGPT ter somente uma base de informação
que só vai até ao ano 2021.”
“... bombed with a lot of non-worthy and accurate
information...”
Highlights the issues
Information “… rigor duvidoso das respostas que a ferramenta
with outdated,
Quality apresenta…”
confusing, and
Challenges “...falta de rigor científico das respostas é uma limitação de
incorrect information.
muito peso...”
“Potentially wrong information...”
“...great need for human thought to accompany the
information so that it doesn't become strange or random.”
“… facilities offered by Chat GPT could lead to a certain
inertia in the process of conceiving ideas, generating a
scenario of very similar campaigns with little degree of
differentiation.”
“As respostas sejam pré-formatados podendo qualquer
pessoa alcançar o mesmo tipo de respostas.”
Addressing the
“... isso faz com que esses pequenos textos não contemplem
challenge of achieving
Lack of a personalidade e a criatividade de quem os está a enviar, o
unique and
Creative que é uma perda preciosa para as relações interpessoais.”
differentiated creative
Differentiation “Pouca espaço para poesia e instinto...”
outputs when relying
“...tem mais tendência para criar copy "gerador de leads" do
on AI tools.
que pensar em ideias inovadoras...”
“Se toda a gente usar o ChatGPT para tentar resolver um
mesmo problema, e ninguém construir sobre o assunto,
provavelmente chegarão à mesma solução. E o facto de
existirem soluções idênticas para o mesmo problema vai
tornar-se, a curto e médio prazo, no problema em si.”
Utilizing ChatGPT in the “… start and go into more in-depth research…”
Empathize stage of “Empathize and Define stages.”
Empathize Design Thinking to gain “Enphatize and define.”
Stage deeper insights into “Empatia…”
user needs, emotions, “In the empathize and define stages.”
and experiences. “Empathize”
“o ChatGPT é bastante eficiente para a materialização de
AI Design textos explicativos em determinados projectos”
Thinking Applying ChatGPT in “Empathize and Define stages.”
Stages the Define stage of “Enphatize and define.”
Design Thinking to “Na Definição, recolhe informação clara e sucinta e fornece
Define Stage clarify project goals, recursos para chegar à melhor solução.”
understand user needs, “Na fase inicial de pesquisa, recolha de informação”
and define problems “In the empathize and define stages.”
more clearly. “Define”
“Mas a integração do ChatGPT no processo pode revelar-se
incrivelmente útil das fases 2 – 4...”

68
 “During "Ideate," offers subtle help, providing some general
insights for market and competition research.”
“… pode ser mais útil na fase de idealização, mas quase
Employing ChatGPT como uma pré-idealização...”
during the Ideate stage “Na idealização, ajuda a analisar a questão em diferentes
of Design Thinking to perspectivas.”
Ideate Stage
generate a wide range “...Brainstorm...”
of ideas and foster “...Idealização...”
creative brainstorming. “... útil na fase de criar ideias...”
“…Ideate…”
“Mas a integração do ChatGPT no processo pode revelar-se
incrivelmente útil das fases 2 – 4...”
“In the "Prototype" stage as well - in strategy planning. A
Using ChatGPT in the discreet integration ensures a more streamlined process,
Prototype stage to offering valuable input without overshadowing the creative
assist in developing brief writing and strategic planning aspects of my role.”
Prototype “…fase do protótipo...”
prototypes, offering
Stage “Prototype...”
suggestions for
improvement and “Mas a integração do ChatGPT no processopode revelar-se
iteration. incrivelmente útil das fases 2 – 4...”

69
70
 ProQuest Number: 31343616

INFORMATION TO ALL USERS

The quality and completeness of this reproduction is dependent on the quality
and completeness of the copy made available to ProQuest.

Distributed by ProQuest LLC ( 2024 ).

Copyright of the Dissertation is held by the Author unless otherwise noted.

This work may be used in accordance with the terms of the Creative Commons license
or other rights statement, as indicated in the copyright statement or in the metadata
associated with this work. Unless otherwise specified in the copyright statement
or the metadata, all rights are reserved by the copyright holder.

This work is protected against unauthorized copying under Title 17,

United States Code and other applicable copyright laws.

Microform Edition where available © ProQuest LLC. No reproduction or digitization

of the Microform Edition is authorized without permission of ProQuest LLC.

ProQuest LLC
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, MI 48106 - 1346 USA