Thinking On Paper

In Chapter 5 of Nexus, Yuval Noah Harari examines how information networks shape political power, from ancient empires and mass propaganda to social media and artificial intelligence.

Mark and Jeremy discuss why information doesn’t automatically produce truth, and how both democracies and totalitarian systems use stories, institutions and communication technologies to organise society. The central issue is who controls the network, who can challenge its claims and whether errors can be corrected.

In this episode, we discuss:

• How information networks create and maintain political power

• Why democracy depends on distributed information and self-correction

• How totalitarian systems centralise information and suppress feedback

• What ancient societies can teach us about collective decision-making

• How Roman rulers such as Nero used propaganda to protect authority

• Why pamphlets and printing helped destabilise established political systems

• How social media feeds influence political beliefs and public debate

• Whether AI-generated memes and propaganda could reshape political movements

• How artificial intelligence could strengthen democratic or authoritarian systems

• Why independent discussion groups and institutions still matter

Harari’s argument is that the struggle between democracy and totalitarianism is partly a struggle between different kinds of information networks. Democracies tolerate competing accounts and mechanisms for correcting mistakes. Dictatorships concentrate control and punish contradiction.

This conversation examines what happens when AI systems begin producing political messages, filtering information and influencing public opinion at a scale no earlier propaganda system could achieve.

--

TIMESTAMPS

(00:00) Disruptors and book lovers

(00:34) Chapter 5 summary

(01:14) Jeremy's first impressions

(01:53) The difference between democracy and totalitarianism

(03:58) Flipping democracy

(05:01) Populism

(09:34) A brief history of democracy

(12:23) Scale and Ancient Rome

(14:56) What is meaningful discourse?

(16:10) Do people like being governed?

(18:43) The rise of the pamphlet

(20:30) The spectrum of democracy

(24:27) Totalitarianism

(25:54) Don't be Stalin's general

(30:00) Will AI be democratic or a dictator?

(30:58) Book clubs as self-correcting mechanisms

(33:24) Will AI remove the human from the loop?

--

Read more books: www.thinkingonpaper.xyz

#AIethics #emergingtechnology #nexus #AI

Kathrin Spendier, Technical Prize Director at XPRIZE, joins Thinking on Paper to explain how quantum computing could contribute to climate science, medicine, energy and materials discovery.

She also discusses the $5 million Quantum XPRIZE, a three-year global competition designed to move quantum algorithms beyond theory and towards practical applications. The competition brings together hundreds of teams working on problems that could eventually benefit from quantum computation.

In this episode, we discuss:

• How quantum computing could support climate modelling and sustainability

• Whether quantum algorithms can accelerate drug discovery

• How quantum systems could help develop new materials

• The potential role of quantum computing in energy research

• Why natural processes such as photosynthesis are relevant to quantum science

• How Google Quantum AI is supporting the Quantum XPRIZE

• What hundreds of teams from dozens of countries are trying to build

• Why competition and technical constraints can accelerate innovation

• What makes useful quantum algorithms difficult to develop

• Which practical quantum applications may emerge first

  
  

Kathrin explains the difference between theoretical quantum advantage and applications that could create measurable value in the real world.

This conversation examines whether quantum computing can move from laboratory progress to useful tools for science, industry and some of humanity’s most difficult technical problems.

Please enjoy the show.

Timestamps

(00:00) - Welcome Disruptors and Curious Minds

(02:10) - Meet Kathrin Spendier: XPRIZE Technical Prize Director

(04:42) - Quantum vs. Classical Computing: What’s the Difference?(07:26) - What Is XPRIZE and How Does It Drive Innovation?

(14:50) - The Challenges Tackled by the Quantum XPRIZE

(17:20) - Quantum Algorithms(27:14) - Who’s Competing in the Quantum XPRIZE?(29:18) - The $5 Million XPRIZE Purse: Why That Amount?(32:17) - Hot Buttons: Rapid-Fire Quantum Insights(33:44) - Are We Living in a Simulation? (33:58) - Top Quantum News and Industry Updates(36:00) - Scaling Quantum Computing(38:12) - Google’s Willow Chip and the Importance of Error Correction(47:16) - Carry-Over Question: How Can Technology Serve Social Good?(49:42) - Mark & Jeremy Reflect: Backstage Thoughts on Quantum and XPRIZE--

Watch on YouTube: https://www.youtube.com/watch?v=SJi7aP5B69A&ab_channel=ThinkingOnPaper

In Chapter 4 of Nexus, Yuval Noah Harari examines how information systems shape belief, authority and our understanding of truth.

Mark and Jeremy discuss how religious texts, books and institutions created durable systems for organising knowledge and legitimising power. They also examine the difference between traditions that claim access to unquestionable truth and the scientific method, which treats human error as unavoidable and correction as essential.

In this episode, we discuss:

• How religious narratives create legitimacy and political authority

• Why written texts transformed the control and transmission of knowledge

• How books became tools for preserving doctrine and enforcing belief

• What the Oracle of Delphi reveals about information and power

• How witch hunts emerged from collective belief and institutional reinforcement

• Why false narratives can persist even when evidence is weak

• How the scientific method changed society’s approach to truth

• Why admitting error is central to reliable knowledge systems

• What historical information networks can teach us about AI and algorithms

• How automated systems could strengthen or distort modern narratives

  
  

Harari’s argument is that information systems don’t merely describe reality. They also create the stories, institutions and categories through which people interpret it.

This conversation examines what religion, books, witch trials and science reveal about the relationship between truth and power, and what those lessons mean in an age increasingly shaped by artificial intelligence.

--

TIMESTAMPS(00:00) Book Lovers(00:32) Shakespeare(01:53) The most important function of religion(02:19) Upsetting the catholic church(03:22) Fear, security and mechanisms of control(04:28) The Big Lebowski(05:11) God, religion and self-correcting errors(06:26) The Oracle Of Delphi(08:10) The Russian Doll / Telephone Game(09:04) Bible building by committee(11:31) The Ship Of Theseus (13:47) The role of women in church(15:52) Idiot Human(17:00) Heinrich Kramer and the birth of witch hunting(19:39) Satanic Conspiracy(23:44) Science and the discovery of ignorance(25:55) Parenting Lessons from science(31:24) To understand AI, do you have to understand religion?(35:00) Self-learning V self-correcting(39:04) Will AI ever have EGO?
--

Learn more: www.thinkingonpaper.xyz

Watch on YouTube: https://www.youtube.com/watch?v=IUBij_kyIEc&ab_channel=ThinkingOnPaper

Anthony Batt and Neville Spiteri, co-founders of Wevr, join Thinking on Paper to discuss how artificial intelligence and virtual reality could change entertainment, communication and digital creativity.

Wevr develops immersive experiences that combine interactive design, real-time computing and virtual environments. Drawing on careers spanning visual effects, gaming, VR and early internet platforms, Anthony and Neville explain what’s required to make digital experiences feel convincing rather than merely visual.

A central concept is presence: the feeling of occupying and interacting with a virtual environment. As AI improves the responsiveness of digital characters, worlds and interfaces, immersive media could become more adaptive, social and personalised.

In this episode, we discuss:

• How AI and virtual reality could transform entertainment

• What immersive media may look like over the next three years

• Why presence matters in virtual environments

• How AI could change human perception and communication

• The technical challenges involved in building convincing virtual worlds

• How intelligent digital characters could affect storytelling

• The role of VR in learning, creativity and collaboration

• Whether immersive experiences can increase empathy and connection

• How entertainment may shift from passive viewing to active participation

• What creators need to understand about the future of spatial media

  
  

Anthony and Neville also explain how advances in hardware, artificial intelligence and real-time rendering are bringing interactive digital experiences closer to mainstream use.

This conversation examines how AI and VR could reshape the way media is created, experienced and shared, and whether immersive entertainment can produce forms of connection that conventional screens cannot.

Please enjoy the show.

--TIMESTAMPS(00:24) Craigslist(01:58) VFX Gaming & Unreal Game Engine(06:00) 3 Years In AI (09:32) What Will You Be Doing With AI In 3 Years?(13:47) Wevr: 10 Year In Emerging Technology(14:17) The True Immersive Experience(16:12) Collapsing Virtual Space(18:01) When You Are Not You(20:00) Shared Immersive Experience(20:33) Presence: The Missing Ingredient Of Virtual Worlds(24:27) Swimming With Whales In VR(28:47) If We Live 300 Years, Will You Be Bored?(31:30) The Cinematic Virtual Experience(37:13) Storytelling Machines(40:29) Virtual Brand World Building(45:35) Curiosity(48:02) Can I Be In Reservoir Dogs?(50:11) Harry Potter: VR Modes--

Learn more:www.thinkingonpaper.xyz

Watch on YouTube: https://www.youtube.com/@thinkingonpaper/videos

In Chapter 3 of Nexus, Yuval Noah Harari examines how stories, documents and bureaucratic systems allow large societies to organise information and exercise power.

Mark and Jeremy discuss the shift from oral storytelling to written records, lists and administrative systems. Stories help people create shared beliefs, while documents turn those beliefs into laws, institutions, identities and systems of control.

In this episode, we discuss:

• How stories allow large groups of people to cooperate

• What Harari means by intersubjective reality

• Why writing transformed human societies

• How lists, records and documents enabled bureaucracy

• The relationship between information systems and political power

• How documentation supports institutions while limiting human judgement

• Whether bureaucracy makes societies more efficient or less adaptable

• Why people remember narratives more easily than administrative information

• What Shakespeare, Kafka and Alan Watts reveal about stories and systems

• How AI and misinformation are changing modern information networks

The central question is whether humans control the systems they create or gradually become shaped by their rules, categories and records.

This conversation examines how storytelling creates shared meaning, how bureaucracy converts meaning into organised power, and what these systems can teach us about artificial intelligence and the future of information.

--Timestamps(00:00) Book Lovers(00:39) William Shakespeare(02:46) Nations Are Intersubjective Realities(04:31) Dreams, Songs And Fantasies (05:53) Writing As The Catalyst For Bureaucracy(07:34) Stories Are Intersubjective Realities(09:39) Computer Says No(11:21) Alan Watts And A Conversation With Myself(14:34) Are Humans Organised?(19:30) Movies About Bureaucracy(21:31) Emergent Systems(24:44) Non Human Information Gets To Jeremy(30:10) Disinformation Debunking
--Read more: www.thinkingonpaper.xyz

--

Is history repeating itself? DeFi (Decentralized Finance) is disrupting the financial world, offering tools that bypass traditional banking and empower users with blockchain and cryptocurrency innovations. But is it the solution we’ve been waiting for—or a hyper-complex system destined to repeat the past?

In this episode of Thinking on Paper, Jeremy and Mark are joined by Kurapika, founder of Factor-Fi, to explore:

We’re unpacking the trust, innovation, and risks of decentralized systems while asking the big question: Can DeFi deliver on its promise of financial freedom, or will it fall into the same traps as traditional finance?

Whether you’re a crypto enthusiast, a finance professional, or just DeFi-curious, this episode breaks down the essentials and explores what’s next for money in the blockchain era.

⚠️ Disclaimer: Nothing here is financial advice—this is for educational purposes only.

Hit play, subscribe, and join us as we navigate the revolution shaping the future of finance.

--

TIMESTAMPS

(00:00 Disruptors And Curious Minds

(01:11) Bitcoin Serendipity

(02:29) The Complexity Of The Financial System

(03:23) Kurapika From Factor-Fi

(04:10) Is DeFi More Complex?

(07:23) The Building Lego Of Decentralized Finance

(07:54) Uniswap: Swap, Lend And Borrow

(09:01) What Are Liquidity Pools?

(14:25) The Risks Of Liquidity Pools

(16:52) Stupid Finance Question

(20:21) Undercollateralized Loans

(23:57) Soulbound Tokens, ZK-Proofs And Truth

(26:56) Creativity And Finance

(33:12) Rage Against The Machine

(36:00) Factor-Fi: More Yield On Your Assets

(41:38) Blockchain Financial Security

(47:06) AI, Crypto & Finance

Watch more shows: www.thinkingonpaper.xyz

--

#Bitcoin #Crypto #DeFi #Blockchain #Finance #AI #Factor #Cryptocurrency #FutureOfFinance #ThinkingOnPaper

In Chapter 2 of Nexus, Yuval Noah Harari examines how stories allow humans to build trust, organise large societies and create shared realities.

Mark Fielding and Jeremy Gilbertson discuss why religions, nations, corporations, currencies and political systems depend on collective belief. These institutions aren’t imaginary in the sense of being unimportant. They become powerful because enough people accept the same story and act accordingly.

Guest Mustafa Quadri joins the discussion to explore how narratives connect strangers and coordinate behaviour across large groups.

In this episode, we discuss:

• How stories create shared realities

• Why storytelling helped humans cooperate at scale

• What Harari means by intersubjective reality

• Whether nations and corporations are shared myths

• How religions use narratives to build trust and identity

• Why brands such as Coca-Cola depend on collective belief

• How Bitcoin derives value from a shared network and narrative

• Why some stories become widely accepted

• How narratives influence behaviour and decision-making

• How stories can unite groups while excluding others

The central argument is that large societies don’t function through facts alone. They also depend on stories that give institutions meaning, create expectations and persuade strangers to cooperate.

This conversation examines how shared narratives become social, political and economic power, and why understanding them matters in an age of digital platforms and artificial intelligence.

Please enjoy the show.

--TIMESTAMPS(00:00) Disruptors and Curious Minds(01:55) Books And Your Thinking Space(02:25) Dust Jacket Vote(02:51) A Religious Disclaimer(03:34) Mustafa On Why We Need Books(07:53) Can We Get Yuval On The Show?(09:21) Sapiens, Nexus And The Power Of Story(12:00) The Last Supper(13:02) Stories Extend Biological Bonds(16:04) Lawyers, Guns & Money(18:41) Inter-subjective realities(22:13) Plato, The Nobile Lie and The 10 Commandments(28:23) Starting A Nation(34:25) Kung Fu Brotherhood(42:00) Good Stories(45:00) Do You Think Unhappiness Make You Susceptible To Story(50:26) Different Believers--Watch Chapter 1: https://www.youtube.com/watch?v=9BhXeNQkbkY&ab_channel=WhereEmergingTechnologiesMeetHumanity--

Learn more at Thinking on Paper

In this episode, we get into the weeds of Martian, a novel AI routing platform designed to help developers select the most appropriate AI model—whether it’s Claude, OpenAI, or others—for specific tasks. With Martian’s innovative technology, developers can optimize AI model performance, ensuring that the best tools are used for the right job.

We explore why AI routing is poised to become the next big thing in software development, how it parallels creative fields like football and quantum pathfinding, and why balancing productivity boosts with critical thinking remains crucial in the age of AI.

All killer, no filler.

Click subscribe and enjoy the show.

In the opening episode of the Thinking on Paper Book Club, Mark Fielding and Jeremy Gilbertson begin reading Yuval Noah Harari’s Nexus and examine its central question: what is information, and how has it shaped human history?

The discussion moves from ancient myths and religious stories to political institutions, mass media and artificial intelligence. Harari argues that human societies are built through information networks that allow large groups to cooperate, but these networks don’t always produce truth.

In this episode, we discuss:

• What information is

• How information networks create power

• Why truth and information aren’t the same thing

• How stories and myths help humans cooperate

• The relationship between information, reality and belief

• Why symbols can carry different meanings in different systems

• How networks shape societies for better and worse

• What artificial intelligence changes about the production and distribution of information

• Whether Harari’s account of human history is convincing

• How Nexus connects ancient information systems with modern AI

This is a live and unscripted book-club discussion. Mark and Jeremy test Harari’s ideas rather than simply summarising them, asking where his arguments hold up and where they require more scrutiny.

The episode introduces the main themes of Nexus: information, networks, truth, power and the growing role of AI in human society.

Please enjoy the show.

--

Timestamps

TIMESTAMPS(00:00) - Welcome: Disruptors and Curious Minds, CEOs, Founders, and Book Lovers(00:50) - Revisiting Previous Books in the Thinking On Paper Book Club(01:08) - Exploring the Brief History of Information Networks(01:33) - Debate: Dust Jackets on Books - Yes or No?(02:40) - Insights from the Nexus Prologue(03:08) - Is Yuval Noah Harari’s Perspective Too Dramatic?(04:54) - The Naïve View of Information Explained(05:58) - The Growing Spectre of Artificial Intelligence(06:48) - Understanding Delusional Networks in Human History(10:03) - Exploring the Relationship Between Truth, Wisdom, and Power(13:52) - Yuval Noah Harari's Definition: What Is Information?(16:58) - The Story of Cher Ami: Information as the Transfer of a Story(17:36) - Most Information Does Not Represent Anything – What Does That Mean?(19:49) - Misinformation, Disinformation, and the Dove of Peace(21:37) - Why Truth Is Not the Same as Reality(28:00) - Music as a Medium of Information and Communication(29:26) - AI-Generated Art and the Importance of Micro Decisions(31:07) - Music, Silence, and the Information You Don't Say(34:57) - Information in the Context of Quantum Mechanics

--

Leonard Tang, founder and CEO of Haize Labs, joins Thinking on Paper to explain how AI models are tested for safety, reliability and predictable behaviour.

Systems such as ChatGPT and Claude can perform well under standard evaluations while still failing in unexpected ways when deployed in healthcare, finance, education or other high-stakes environments. Haize Labs develops testing and evaluation tools designed to identify those vulnerabilities before AI systems reach users.

In this episode, we discuss:

• How AI safety testing and red teaming work

• What Haize Labs means by an AI robustness and safety layer

• How researchers uncover hidden failure modes in language models

• Why benchmark performance doesn’t guarantee real-world reliability

• How adversarial testing exposes weaknesses before deployment

• Why different industries need specific AI safety standards

• How organisations can evaluate whether an AI model is suitable for a particular use case

• The difference between Silicon Valley’s AI claims and adoption in established industries

• How language models could affect communication across cultures

• What it means to align an AI system with human needs

Leonard explains why AI safety can’t be reduced to a single test or universal code of conduct. A model used for medical advice faces different risks from one used in education, financial services or customer support.

This conversation examines how developers and organisations can test AI systems more rigorously, identify failures before deployment and build models that behave more reliably in real-world conditions.

--

TIMESTAMPS

(00:00) - Disruptors and Curious Minds(01:07) - Our Sponsor: Conviction (01:50) - Introducing Leonard Tang: AI CEO and Founder(03:37) - The Importance of AI Safety: What’s at Stake in AI Development?(06:21) - Using Mathematics and Modeling to Understand Human Behaviour in AI(08:12) - Why Are Technologists So Often Musicians?(11:06) - Language, Culture, and AI(17:05) - Common Misconceptions About AI: What People Get Wrong(19:20) - The Dartmouth Conference: Birth of AI and Its Lasting Impact(19:55) - Claude and ChatGPT Pre-training: What Do The Models Go Through?(25:20) - An Alan Watts AI Model for Enhanced Understanding(28:33) - Claude vs ChatGPT: Comparing AI Models and Performance(31:44) - AI Jailbreak Detection(33:25) - How Dreamlike Images Enhance AI Safety and Trustworthiness(38:20) - Top-Down vs Bottom-Up AI Development: Approaches to Building Safer AI(42:55) - Protecting Artists, Intellectual Property, and Art in the Age of AI(48:20) - Developing an AI Code of Conduct for Ethical AI Usage(49:45) - A Message for Veteran AI Stars(52:35) - Restructuring Education for Critical Thinking in the Age of AI(54:16) - Book Club Live--

Quotes from the show:"We need to rigorously test AI models to discover all their vulnerabilities, failure modes, and gotchas before they get deployed in production.""AI is a technology of language, and inevitably, it will empower us to merge cultures.""We’re trying to get AI to be a little more mature, a little more sophisticated, and just more reliable.""What we’re interested in is enforcing an AI code of conduct for specific applications, making AI systems tightly aligned with the needs of their use cases.""People in legacy industries are underestimating AI’s potential, while Silicon Valley is often overhyping it."--🔗 More:Visit Haize Labs: ⁠⁠⁠https://haizelabs.com/⁠⁠⁠Visit Thinking On Paper: ⁠⁠⁠https://www.thinkingonpaper.xyz/⁠⁠⁠

⁠⁠⁠⁠⁠⁠Instagram⁠⁠⁠⁠⁠⁠

⁠⁠⁠⁠⁠⁠Twitter⁠⁠⁠⁠⁠⁠