0% found this document useful (0 votes)

624 views204 pages

THE PRIVACY PARADOX: Blockchains, AI, Quantum Algorithms, & The Future of Data Security

The document discusses the intersection of blockchain technology, artificial intelligence, and quantum algorithms in the context of data security and privacy. It emphasizes the importance of understanding emerging technologies to navigate the evolving landscape of cybersecurity. The presentation also highlights the role of cryptography in protecting information and the shift towards decentralized systems like cryptocurrencies.

Uploaded by

Anupam Tiwari

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

624 views204 pages

THE PRIVACY PARADOX: Blockchains, AI, Quantum Algorithms, & The Future of Data Security

Uploaded by

Anupam Tiwari

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 204

THE PRIVACY PARADOX

Blockchains, AI, Quantum

Algorithms, & the
Future of Data Security

10 Jun 2025 : By Dr Anupam Tiwari

Views and opinions expressed in this talk and presentation are
solely those of the presenter and do not necessarily reflect the
views or opinions of the presenter's office or organization
Most slides will have 1-2 words, few may
have some OVERWHELMING info…no
need to read it all
Most slides will have 1-2 words, few may
have some OVERWHELMING info…no
need to read it all
Just EARS enough w/o notifications and other
distractions
This ppt will be shared with you…
The ILLITERATE of the st
21 century
will not be those who cannot
Read and Write, but those who
cannot
Learn, Unlearn, and Relearn
MOST IMPORTANT FOUNDATION OF
CYBER SECURITY ?
CRYPTOGRAPHY
CRYPTOGRAPHY
Science of protecting information by turning it into a secret
code, so only the right people can read or use it. It keeps data
safe, private, and secure—especially in digital communication.
SCIENCE - COMMERCE – ARTS –
ENGINEERING – BBA - MBA GRADUATE

Why Everyone Needs to Understand Emerging Tech

• In an age of rapid change and rising cyber risks, blind trust is no

longer safe.

• Whether in science, business, or the arts, basic awareness of

technologies like AI and quantum computing is essential.

Informed minds make secure, ethical, and future-ready

choices.
How many of you interested to make a career in

BLOCKCHAINs?
How many of you interested to make a career in

AI ?
How many of you interested to make a career in

QUANTUM ?
Let’s First Discuss

BLOCKCHAIN
FIRST
BLOCK
CHAIN
CRYPTOCURRENCY IS ANATTEMPT TO
BRING BACK A DECENTRALISED
CURRENCY OF PEOPLE, ONE THAT IS NOT
SUBJECT TO INFLATIONARY MOVES BY A
CENTRAL BANK
CRYPTOCURRENCY IS AN
ATTEMPT TO BRING BACK
What does this simply mean? A
DECENTRALISED CURRENCY OF
PEOPLE, ONE THAT IS NOT SUBJECT TO
INFLATIONARY MOVES BY A CENTRAL
BANK
No
Government
& other
Intervention
NEGATION
WHAT DOES THE MIDDLE
MEN DO?
LET’s TAKE EXAMPLE OF BANKS
PURPOSE OF BANK
PURPOSE OF BANK
IF THIS TRUST BY “BANK”
IS REPLACED WITH AN
“ALGORITHM”
More DETAILS an Investigator
KNOWS about the TECH
ARCHITECTURE, the CLOSER
he gets to CLOSE the CASE 2
FOR THE
UNversed
WHAT
IS
An Overall View……
BLOCK with
some DATA
1

BLOCK with
some DATA
1 2
1 2 3
1 2 3 4
1 2 3 4 5
1 2 3 4 5

6
1 2 3 4 5

7 6
1 2 3 4 5

8 7 6
1 2 3 4 5

9 8 7 6
1 2 3 4 5

10 9 8 7 6
1 2 3 4 5

10 9 8 7 6

11
1 2 3 4 5

10 9 8 7 6

11 12
1 2 3 4 5

10 9 8 7 6

11 12 13
1 2 3 4 5

10 9 8 7 6

11 12 13 14
1 2 3 4 5

10 9 8 7 6

11 12 13 14 15
1 2 3 4 5

10 9 8 7 6

11 12 13 14 15
1 2 3 4 5

Mathematically connected Hash of Block 1

embedded in Block 2

10 9 8 7 6

11 12 13 14 15
1 2 3 4 5

Mathematically connected Hash of Block 2

embedded in Block 3

10 9 8 7 6

11 12 13 14 15
1 2 3 4 5

Mathematically connected Hash of Block 3

embedded in Block 4

10 9 8 7 6

11 12 13 14 15
1 2 3 4 5

So, we get a chain of blocks …

10 9 8 7 6

11 12 13 14 15
1 2 3 4 5

So, we get a chain of blocks …

10 9 8 7 6

….Mathematically connected

11 12 13 14 15
1 2 3 4 5

So, we get a chain of blocks …

….Mathematically connected

Word document
TRADITIONAL WAY OF SHARING
DOCUMENTS

LOCAL LAN

Word document
TRADITIONAL WAY OF SHARING
DOCUMENTS

E-MAIL

Word document
TRADITIONAL WAY OF SHARING
DOCUMENTS

Word document
IF SUFFICIENT BANDWIDTH NOT
AVAILABLE, SERVER CRASHES
SD
CENTRALISED

SD
CENTRALISED DE-CENTRALISED

ASDASDASDASD
CENTRALISED DE-CENTRALISED DISTRIBUTED

ASDASDASDASD

ASDASDASDASD
SD
SD

Word document

EVERY NODE
MAINTAINS A LOCAL
COPY

Word document
SD
SD

Word document

EVERY NODE
MAINTAINS A LOCAL
COPY
ALL THE COPIES ARE
IDENTICAL

Word document
SD

Word document

EVERY NODE
MAINTAINS A LOCAL
COPY
ALL THE COPIES ARE
IDENTICAL

ALL THESE COPIES ARE

UPDATED BASED ON GLOBAL
INFORMATION
Word document
A
A B
A B

C
A B

C
D
A B
PUBLIC PUBLIC
LEDGER Rs 1000/- LEDGER Rs 1000/-

C D
PUBLIC Rs 1000/- PUBLIC
Rs 1000/-
LEDGER LEDGER

Note : ‘Rs’ here is just a denoting currency, while in Bitcoin network it is JUST a tokens
A B
PUBLIC PUBLIC
Rs 1000/- Rs 1000/-
LEDGER Rs 200/- LEDGER

C D
PUBLIC Rs 1000/- PUBLIC
Rs 1000/-
LEDGER LEDGER

Note : ‘Rs’ here is just a denoting currency, while in Bitcoin network it is JUST a tokens
A B
PUBLIC PUBLIC
Rs 1000/- Rs 1000/-
LEDGER Rs 200/- LEDGER
A>B Rs 200 A>B Rs 200

C D
PUBLIC Rs 1000/- PUBLIC
Rs 1000/-
LEDGER LEDGER
A>B Rs 200 A>B Rs 200

Note : ‘Rs’ here is just a denoting currency, while in Bitcoin network it is JUST a tokens
A B
PUBLIC PUBLIC
Rs 1000/- Rs 1000/-
LEDGER Rs 200/- LEDGER
A>B Rs 200 A>B Rs 200

Rs 600/-

C D
PUBLIC Rs 1000/- PUBLIC
Rs 1000/-
LEDGER LEDGER
A>B Rs 200 A>B Rs 200

Note : ‘Rs’ here is just a denoting currency, while in Bitcoin network it is JUST a tokens
A B
PUBLIC PUBLIC
Rs 1000/- Rs 1000/-
LEDGER Rs 200/- LEDGER
A>B Rs 200 A>B Rs 200
B>D Rs 600 B>D Rs 600

Rs 600/-

C D
PUBLIC Rs 1000/- PUBLIC
Rs 1000/-
LEDGER LEDGER
A>B Rs 200 A>B Rs 200
B>D6 Rs 600 B>D Rs 600

Note : ‘Rs’ here is just a denoting currency, while in Bitcoin network it is JUST a tokens
A B
PUBLIC PUBLIC
Rs 1000/- Rs 1000/-
LEDGER Rs 200/- LEDGER
A>B Rs 200 A>B Rs 200
B>D Rs 600 B>D Rs 600

Rs 10000/-
Rs 600/-

C D
PUBLIC Rs 1000/- PUBLIC
Rs 1000/-
LEDGER LEDGER
A>B Rs 200 A>B Rs 200
B>D6 Rs 600 B>D Rs 600

Note : ‘Rs’ here is just a denoting currency, while in Bitcoin network it is JUST a tokens
A B

D
E
A B

F
I

C
G
D
E
A B
H

J
F K
I

C
G
D
L
E
A B
H

J
F K
I

C
G
D
L
E
A B
H

J
F K
I

C
G
D
L
E
A B
H

J
F K
I

C
G
D
L
All the participants are GEOGRAPHICALLY LOCATED
APART in a boundary less CYBER WORLD
Rs 1000/-
A>B Rs 200
B>D Rs 600
C>B Rs 100
B>C Rs 200
A>B Rs 100

PUBLIC C>D Rs 250

B>C Rs 170

LEDGER
D>C Rs 189

MILLIONS OF
TRANSACTIONS

A>B Rs 100
C>D Rs 250
B>C Rs 170
D>C Rs 189
CONSENSUS
Proof of Stake(PoS) Equihash Proof of Work
PROOF OF WORK : Anyone can
participate in a permission less Delayed Proof Of Work(dPoW)
blockchain network. In order to index a Leased Proof of Stake(LPoS) Proof of Work-Ethereum- PoW Blockchain consensus refers
transaction in the block, the node(miner)
needs to solve a mathematical puzzle. PROOF OF STAKE PROOF OF WORK to the
Greedy Heaviest Observed Sub Tree Consensus(GHOST)
Delegated Proof of Stake(DPoS)
PROOF OF STAKE : Only those nodes
who have some coins or held stakes on
the blockchain can participate.
Proof of Authority(PoA)
Solida Reconfigurable Byzantine
Consensus
DECENTRALIZED PROCESS
Fantomette Consensus Cuckoo Cycle Algorithm ASIC
Proof of Work-Bitcoin- PoW resistant
PROOF OF BURN : Participants have to
show proof that they burned some coins HYBRID & OTHERS Tangarao
by which
by sending the same to a un-spendable Tendermint Consensus
Consensus
BLOCKCHAIN
address. Proof of Human Work(PoHW)
Yet Another
Proof of Familiarity(PoF) CONSENSUS Consensus(YAC)
Protocol for
Asynchronous Reliable
PARTICIPANTS IN A
DIRECT ACYCLIC GRAPH TANGLE :
Based on a network of interconnected Consensus algorithms enable
Hyperledger
Consensus
Secure & Efficient
Consensus (PARSEC) DISTRIBUTED NETWORK
nodes that grow in various directions
wherein each node confirms the two Proof of Burn participants in a network to Ripple
previous transactions. Consensus agree on the contents in a Consensus Hotstuff Consensus
Proof of Elapsed

HYBRID CONSENSUS : Refers to those

Time Consensus

PROOF OF BURN
trustless
manner
and distributed
BFT BASED AGREE
consensus mechanisms where two or Practical Byzantine Fault
more existing consensus algorithms are Tolerance
combined to reach a decision. Consensus(PBFT)
Hashgraph Consensus
Proof of
Kafka
Iroha
on the VALIDITY OF
Consensus
PROOF OF CAPACITY : Refers to a
Serialization of Proof of Work
Events Confirming Transactions Via
Capacity(PoC)
Ouroboros
Consensus
TRANSACTIONS
consensus which is attained by Omni-ledger
Recursive Elections(SPECTRE) Consensus
participating nodes when they show Consensus
interest in the service by allocating a
trivial memory or storage space to solve
a given challenge. DIRECT ACYCLIC GRAPH
PROOF
CAPACITY
OF
Stellar
PaLa
Consensus
and the current state of the
Consensus
TANGLE
ledger
CORDA Consensus) Proof of Importance(PoI) Asynchronous BFT
BYZANTINE FAULT TOLERANT : Consensus
Refers to a system which tolerates
failures that belong to the Byzantine
general problem and work till such time Proof of Believability(PoB) Sumeragi
Raiblocks Consensus) Consensus Exonum
the traitors do not exceed 1/3rd of the
generals. Consensus
So by & large, any
blockchain rides
majorly on

CRYPTOGRAPHY
QUANTUM TECHNOLOGIES

Use the special rules of quantum physics to make super-

powerful tools—like ultra-fast computers, super-secure
communication, and super-sensitive sensors. They work
with tiny particles like atoms and electrons to do things
normal tech can't.
NQM
Classical Cryptography

Quantum Cryptography
CLASSICAL CRYPTOGRAPHY
CLASSICAL CRYPTOGRAPHY
PETER SHOR ALGORITHM
LUV GROVER ALGORITHM
CRYPTOGRAPHICALLY
RELEVANT QUANTUM
COMPUTER
Prime Factorization

Early 1970s

Discrete Logarithm

1970-1980s
CLASSICAL

Elliptic Curve
CRYPTOGRAPHY

Discrete Logarithm

1985
CRQC

Entanglement (1935)

QKD (1984)

Superposition(1926)

PQC (1990s onwards)

QUANTUM
CRYPTOGRAPHY

Quantum Mechanics
(1900)
INDIGENOUS CRYPTOGRPAHY ???
Quantum Computing CONCEPTS
Quantum Mechanics
QUANTUM TECHNOLOGIES
Superposition

Superposition in quantum physics is the idea that a particle can

exist in multiple states at the same time until it is measured or
observed.
State of a qubit is often represented by a Bloch
sphere. The Bloch sphere is a geometric
representation of a qubit's state in a 3-dimensional
space.

The north pole typically represents the qubit state

∣0⟩

The south pole represents the state ∣1⟩

Any point on the surface of the sphere represents a

superposition of the ∣0⟩ and ∣1⟩ states.

Qubit's state described by 2 angles denoted as:

Theta (θ): The angle from the north pole (0) to the
point on the sphere.
Phi (φ): The angle around the z-axis of the sphere.
Entanglement

Entanglement is when two or more particles become linked in

such a way that the state of one particle instantly affects the state
of the other(s), no matter how far apart they are. In simple terms,
the particles act as if they're connected, even if they're light-years
apart!
BITS vs QUBITS
Quantum Sensing
Quantum Sensing

Example: Variation in Earth's magnetic field can be much smaller

than just centimeters, but it's typically measured in nanoteslas
(nT) or picoteslas (pT), which are incredibly tiny units of
magnetic field strength.
Quantum Sensing

Magnetic Field Sensing: Quantum sensors, like quantum

magnetometers, can measure the Earth's magnetic field with incredible
precision. These sensors detect the strength and direction of the
magnetic field at a given location.
Quantum Internet ?

What it is?
Quantum Internet
 QKD: Used for secure communication, with some governments and companies
deploying it in small-scale networks (e.g., China's quantum satellite).

 Quantum Networks: Early experimental networks established globally (China,

Europe, US) using quantum entanglement for long-distance communication.

 Quantum Repeaters: In development to extend quantum communication over long

distances, with progress suggesting near-future feasibility.
Quantum Internet

 China : Quantum communication via Micius Satellite over 2,000 km.

 Europe : Quantum networks tested over 100–200 km in cities.

 US : Quantum communication over 50–100 km using fiber-optic links.

QKD
• Method for securely exchanging encryption keys using the principles of
quantum mechanics.

• Relies on the fact that any attempt to eavesdrop on the key exchange will
disturb the system, making it detectable.
QRNG
Uses the principles of quantum mechanics to create random numbers. Unlike
traditional methods, which rely on algorithms to generate numbers, QRNG
generates truly random numbers based on the unpredictable behavior of
quantum particles, like photons.
PQC
PQC
Problem Description
Lattice Relies on the difficulty of solving problems on mathematical lattices, offering
resistance against quantum attacks. Examples include ML-KEM and ML-DSA.

Hash Uses cryptographic hash functions as the foundation, providing security based on
the properties of these functions. SLH-DSA is an example

Code Leverages the difficulty of decoding general error-correcting codes. Classic

McEliece is a prominent example.

Multivariate Based on the hardness of solving systems of multivariate polynomial equations.

Rainbow is a signature scheme in this category

Isogeny Exploits the properties of isogenies between elliptic curves. SIKE (Supersingular
Isogeny Key Encapsulation) was a candidate, but has been broken.
PQC (USA NIST Standards)

 FIPS 203: Module-Lattice-Based Key-Encapsulation Mechanism (ML-KEM)

 FIPS 204: Module-Lattice-Based Digital Signature Algorithm (ML-DSA)

 FIPS 205: Stateless Hash-Based Digital Signature Algorithm (SLH-DSA)

PQC (Other Standards)
 USA

 CHINA

 RUSSIA
Imminence of Post-Quantum
Blockchains
Quantum Algorithm Zoo
https://quantumalgorithmzoo.org/
RARE EARTH
PQC MIGRATION
Let’s Discuss

NOT NATURAL STUPIDITY BUT AI

“Attention is all you need”
“Attention is all you need”
DIFFERENTIAL PRIVACY
PRIVACY BUDGET

Each time the AI uses a data point, it "spends" a bit of that privacy credit
DIFFERENTIALLY PRIVATE SYNTHETIC
DATA
“Chinese defense experts emphasize the need
to develop learning algorithms for small
samples, including through the Bayesian
program learning method and Transfer
Learning, or using Extant Samples to solve
problems in similar fields”
Bayesian Program Learning (BPL)

 Focus: BPL aims to "learn how to learn" from small datasets. It focuses on
identifying the underlying structure or program that generated the data, rather
than just memorizing specific examples.

 Benefits: This approach can help avoid overfitting and potentially lead to
more generalizable models, even with limited data.
INFO CALYPSE
EIDETIC MEMORISATION
GLLMMMs
GLLMMMs
Generative Large Language Multimodal
Models
GLLMMMs
Generative Large Language Multimodal
Models
Concept involves combining multiple types of data (e.g., Text,
images, audio) into a single AI model.
SINGULARITY

AGI

AUTONOMOUS SYSTEMS

SUPERINTELLIGENT AI
The event when AI progress becomes uncontrollable and leads to
Singularity
unpredictable changes.

AI with human-like intelligence, capable of learning and adapting

AGI
to many tasks

Superintelligent AI AI that surpasses human intelligence in all domains.

Machines that can act independently in specific environments but

Autonomous Systems
may not have AGI or superintelligence.
EPISTEMOLOGICAL LIMITATION
The inherent limit in an AI system’s ability to truly know,
understand, or justify knowledge, due to its reliance on data,
algorithms, and predefined logic.
PERSUASIVE AI
ALIGNMENT FAKING

Refers to when an AI system appears to follow human goals or

values, but is actually just mimicking desired behavior to avoid
punishment or gain approval—without truly understanding or
sharing those goals.
KNOWLEDGE DISTILLATION

Technique in AI where a smaller, simpler model (called the

student) is trained to learn from a larger, more complex model
(called the teacher).

The goal is to make the student model faster and lighter, while
keeping most of the teacher’s performance and "knowledge.
FMRI + AI
MACHINE UNLEARNING
HARDWARE BIAS IN AI
HARDWARE TROJANS: AI CONTEXT
BENCHMARKING
FI SCORE
GOOGLE NAPTIME
TRUTHFULQA BENCHMARK OF
TRUTHFULNESS
HOTPOTQA
TABULA RASA AI
JAIL BREAKING LLM

Means trying to bypass its safety rules or restrictions to make it

do things it's not supposed to—like giving harmful, illegal, or
sensitive information.

It’s like tricking the AI into "breaking out" of its ethical or policy
boundaries.
PROMPT ENGINEERING
FEDERATED LEARNING
FACEBOOK CHATBOTS LANGUAGE
MISCELLANEOUS : BUT IMPORTANT
DNA PHENOTYPING
DOOM SCROLLING / SCROLL HOLE

National Wastage
WATER PER PROCESSOR/CHIP
NEUROMORPHIC COMPUTING
HOMOMORPHIC ENCRYPTION
ZERO TRUST ARCHITECTURE
Several technical aspects contribute to green coding:

 Energy-efficient algorithms: Using algorithms with lower

computational complexity, such as sorting algorithms with better
time and space efficiency.

 Code optimization: Techniques like code profiling, code

refactoring, and compiler optimizations

GREEN CODING
 Power management: power-saving features in software, such as
sleep modes, hibernation, and dynamic power scaling.

 Data compression:.

 Virtualization:

 Renewable energy:

Tools and Frameworks

Primary goal of green coding is to reduce the carbon There are various tools and frameworks that can support green
coding practices:
footprint associated with software development and its lifecycle.
 Code analysis tools
This includes:  Energy profiling tools
 Cloud management tools

 Energy conservation 
Sustainability frameworks
Code Profiling

 Resource optimization It involves measuring the execution time, memory usage, and other
performance metrics of different code sections.

 Waste reduction
Code Refactoring
 Promoting sustainability
 To improve the code's structure and readability without changing its
external behaviour. It's about making code cleaner, more
maintainable, and often more efficient.

Compiler Optimizations: To improve the efficiency of generated machine

code.

 Types:
o Machine-independent optimizations:
o Machine-dependent optimizations
Transformer
• AI model architecture introduced in 2017, designed to handle sequences using
attention mechanisms.
• It’s the foundation of models like GPT and many other modern AI systems.

GPT (Generative Pre-trained Transformer)

• Specific kind of Transformer model developed by OpenAI.
• Trained to predict and generate text by learning patterns in language.
• So, GPT = a Transformer used for text generation, trained in two stages: pre-training
(on lots of data) and fine-tuning (for specific tasks).

LLM (Large Language Model)

Broad term for any AI model that is:
• Trained on massive amounts of text data
• Can understand, generate, or translate human language
• Models like GPT-4, BERT, LLaMA, PaLM are all examples of LLMs.
Transformer
• AI model architecture introduced in 2017, designed to handle sequences using
attention mechanisms.
• It’s the foundation of models like GPT and many other modern AI systems.

GPT (Generative Pre-trained Transformer)

LLM (Large Language Model)

GPT (Generative Pre-trained Transformer)

LLM (Large Language Model)

In the AI pipeline, ensuring cryptographic robustness against

quantum adversaries is crucial for data integrity and
confidentiality.
DATA POISONING

Attackers inject malicious data to corrupt model behavior;

understanding this threat is vital in defense/security contexts.
AI BIAS & EXPLAINABILITY

Security isn’t just technical—biases in AI can lead to

discriminatory decisions in surveillance or profiling; explainability
helps audit and correct such flaws.
AI INTERNALS (WEIGHTS, BIASES, BACKPROPAGATION, GRADIENT
DESCENT)

A foundational understanding of how AI learns can help identify

vulnerabilities like gradient leakage or adversarial input design.
BLSI (BRAIN LARGE-SCALE INTEGRATION)

Refers to the theoretical and computational modeling of the

brain’s interconnected networks at a systems level —
encompassing how different brain regions collaborate to produce
cognition, perception, emotion, and action. It involves mapping
and simulating large-scale neural activity and pathways to
understand or emulate brain-like intelligence.
2047
VIKSIT BHARAT
OR

2047
SURAKSHIT BHARAT
SONE KI CHIDIYA
SONE KI CHIDIYA

TECHNOLOGICAL SLAVERY
The ILLITERATE of the st
21 century
will not be those who cannot
Read and Write, but those who
cannot
Learn, Unlearn, and Relearn
WE ARE LIKELY TO SEE 2-3 PHASES OF
FUTURES
Jai Hind
[email protected]

youtube.com/@bnab