Scribd
Upload
3 views19 pages

Lab2 Ai

The document discusses the capabilities and limitations of artificial intelligence (AI), highlighting that AI can automate tasks that require minimal thought but struggles with complex reasoning. It emphasizes the importance of relevant and high-quality data for AI performance and notes that the boundary of what AI can do is still unclear. The ultimate goal of AI research is to create machines that can perfectly mimic human intelligence, although current AI systems can still be distinguished from humans.

Uploaded by

phtgaming123
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
3 views19 pages

Lab2 Ai

The document discusses the capabilities and limitations of artificial intelligence (AI), highlighting that AI can automate tasks that require minimal thought but struggles with complex reasoning. It emphasizes the importance of relevant and high-quality data for AI performance and notes that the boundary of what AI can do is still unclear. The ultimate goal of AI research is to create machines that can perfectly mimic human intelligence, although current AI systems can still be distinguished from humans.

Uploaded by

phtgaming123
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 19

GROUP 9

WHAT AI CAN
AND CANNOT
DO
1.INTRODUCTION
• Artificial
1 Introduction
intelligence is a
broad field, which
refers to the use of
technologies to
build machines and
computers that
have the ability to
mimic cognitive
functions
associated with
human intelligence
which theoretically
means that
anything we can do
AI can also do
1 • But the field of
AI research has
Introduction
only existed for (co
nt) about 100 years
and only started
advancing
recently so there
are still many
limitations when
it comes to what
AI can and
cannot do at this
current moment.
2.WHAT AI
CAN DO
2.1 Rule of
thumb
• An imperfect rule of
thumb to know what
an AI can do is that if
you can do it with 1
second of thought then
we can probably now
or soon automate it.

• Some examples could


be how you know
where something is by
looking at it , knowing
the state of an object
at a glance,…
2.2 What that
make it easier
• At its core AI requires
data to learn so the more
relevant data that it can
learn the better it will be
at doing its task.

• Another thing that helps


is if the task it’s trying to
do is “simple”
(something that doesn’t
require much
deduction/reasoning).
3.WHAT AI
CANNOT DO
3.1 Rule
• A rule of of
of thumb
what AI can do is
thumb
anything you can
think of doing in 1s
then it can do we can
stand to reason that
what ai can’t do is
the opposite which is
whatever you have to
think about for a long
time/predict to do is
what AI probably
can’t do.
• Such as accurately predict the
stock market using only the
price history.
3.2 What
make
• Things it
that can make
creating an AI harder
harder
and less efficient
include:

 Non relevant data


 Low amount of data
 Low quality data/data
set
 performing on new
types of data
4 OVERVIEW
• Overall the
4 OVERVIEW
boundary
between what AI
can or cannot is
still quite fuzzy
and you need
time and
technical
diligence to even
know what is
feasible and
what isn’t.
WHAT CAN AI
CURRENTLY DO
AND NOT DO?
Recognize a cars position to
know if you could pass it or not?
Recognize the traffic hand-
signs
Diagnose pneumonia

From 10,000 labeled images of pneumonia From 10 textbook images of pneumonia


with explanations
Fool a The ultimate
human ? goal of AI is to
create an
intelligence that
mimics that of a
humans
intelligence
perfectly.
Which also
include being
undisguisable
from a human.
There have been
many ways to
test this most
famously is the
cont
• Currently we are in a
sort of middle ground.
• There are many AI
which you could
communicate and talk
with or chat bots who
whole purpose is to
seem human.But they
can still be distinguish
somewhat easily
• The art of AI chat bots
has been evolving
rapidly these past few
years which could lead
to a perfected AI
personality someday in
the near future.
THANK YOU
FOR LISTENING

You might also like