This post is essentially my evolving notebook. It brings together the writers, researchers, podcasters, bloggers, YouTubers and workshop leaders whose work has helped me understand AI over the past two years.

I put it together by going back through my YouTube, X and Spotify histories and revisiting the blogs and newsletters I read. The people here often disagree, sometimes fundamentally. I am not presenting a curriculum, a ranking or an argument for or against AI—I am trying to understand who these people are, what expertise they bring and how their views fit into the wider debate.

I pay for the top-tier plans on both ChatGPT and Claude and constantly test the latest models on my own work. I know I am still only scratching the surface, so if you are deeper into any of these areas, please suggest the people I should be reading, watching or listening to.

1. Where is AI now?

The easiest way I have found to understand the market is to separate frontier models from everything else. Frontier models are the strongest general-purpose systems available at a given moment; the rest of the market includes smaller, cheaper and more specialised models that may be better for a particular job. The Stanford AI Index is the clearest annual overview I have found of capabilities, investment, adoption and policy.

What real big problems of humanity could AI solve?

Science is where the promise becomes more concrete. AlphaFold predicts the three-dimensional shape of proteins—the tiny biological machines whose shape helps determine what they do. That can give researchers a much faster starting point for understanding disease and designing experiments. AlphaFold 3 extends the work to interactions among proteins, DNA, RNA, drugs and other molecules.

• Malaria: researchers are using predicted structures to explore which parts of a protein could go into a vaccine.

• Neglected diseases: DNDi has used AlphaFold in work on Chagas disease and leishmaniasis.

• Pollution: protein prediction can help scientists design enzymes that break down plastic.

• Antibiotic resistance: structural information can reveal ways to block bacterial resistance mechanisms.

• Cancer: AI is being studied across imaging, diagnosis, drug discovery and molecular interaction prediction. This review of AI in oncology is a useful reality check on both the possibilities and the limitations.

Where I would go to learn about AI and science

• Demis Hassabis — DeepMind’s work on frontier AI, AlphaFold, robotics and scientific discovery.

• Pushmeet Kohli — AI for science and the research behind AlphaFold.

• Regina Barzilay — machine learning for cancer detection and drug discovery.

• Daphne Koller — machine learning applied to biology and medicines.

• Eric Topol and Ground Truths — medicine, health research and AI, written for a wide audience.

• Google DeepMind: The Podcast — Hannah Fry’s accessible conversations with the scientists doing the work.

• Hugging Face Daily Papers and The Batch — for the geeks- regular ways to see new research without relying only on social-media summaries.

2. What does “open” versus “closed” AI actually mean?

A closed model is accessed through a company’s app or API while its weights and much of its training process remain private. An open-weight model makes its learned parameters available so others can run or adapt it. That is not always the same as fully open-source AI; the Open Source Initiative explains the stronger definition.

I find the politics striking. The United States built much of the modern open-source software movement, yet its leading AI labs chose closed frontier models. Chinese companies have become major suppliers of capable open-weight models. Open models widen access and local control; closed models can make safety, support and rapid product development easier. Both also concentrate different kinds of power.

Which models currently lead?

These rankings change quickly. In the Artificial Analysis Intelligence Index v4.3 published on 7 September 2026, the leading closed models included Claude Fable 5.1, GPT-6 Astra and Claude Opus 5. Astra appears to be leading much of the current conversation, although it was tied rather than alone at the top of that index.

The leading open-weight entries included GLM-5.3, Kimi K3 and GLM-5.3 Flash. Qwen3.8 is another important family to watch.

Where I go to learn about open AI

• Nathan Lambert’s Interconnects — open models, post-training, research culture and policy.

• Hugging Face — model pages, technical details, licences and a large open-model community.

• Goose — Block’s open-source AI agent, useful for seeing what an open ecosystem looks like in practice.

• Artificial Analysis — capability, speed and price comparisons across open and closed models.

3. How do we actually measure which model is best?

There is no single “best” model. One may lead at maths, another at coding, another in speed or price. Benchmarks can also be gamed, contaminated or disconnected from everyday work. I therefore look at several independent sources and then repeat the same real tasks on the models I pay for.

The benchmarking platforms I watch

• Artificial Analysis — a composite capability index alongside speed and price.

• LMArena — blind human preferences between model outputs; useful, but preference is not the same as factual reliability.

• Stanford HELM — transparent evaluation across many scenarios and metrics.

• LiveBench — frequently refreshed questions intended to reduce benchmark contamination.

• Epoch AI’s FrontierMath — difficult mathematical problems designed for frontier systems.

• Hugging Face Open LLM Leaderboard — comparisons focused on open models.

My own test is simpler: can the model handle the documents, research, writing, product work and follow-through I actually need, and can it do so reliably enough to trust?

4. What could the next ten years hold?

No one knows, so I find it more useful to compare named views than to offer one confident forecast.

If you ask Elon Musk

The direction is abundance built on intelligence, robotics and energy. Tesla’s “sustainable abundance” master plan connects autonomous systems and humanoid robots to a future in which goods and services become dramatically cheaper.

If you ask Marc Andreessen

AI is a broad amplifier of human intelligence that could improve education, healthcare, productivity and creativity. Why AI Will Save the World is the clearest statement of his optimistic, pro-acceleration case.

If you ask Andrew Ng

The near-term opportunity is not one magical superintelligence but millions of useful applications. His talks on agentic workflows focus on what teams can build now, while warning against treating every task as suitable for automation.

If you ask Demis Hassabis

We may be in the “foothills of the singularity”: systems approaching more general intelligence, with major consequences for science, medicine and society. His view combines ambition with a strong emphasis on safety and scientific institutions.

Where I go to compare future views

• The Economist’s interview with Yuval Noah Harari — power, institutions, democracy and the historical significance of AI.

• Dwarkesh Patel — long-form interviews where lab leaders, economists and sceptics have room to explain their assumptions.

5. Why do some people think AI could end the world?

The catastrophic-risk argument is that a future system more capable than its operators could pursue the wrong objective, deceive people or gain access to critical systems before we know how to control it. The critics question how likely that is, whether today’s technology justifies the analogy, and whether the focus distracts from harms already happening.

The catastrophic-risk thinkers and their critics

• Eliezer Yudkowsky — the most severe case that sufficiently advanced, misaligned AI could be uncontrollable and fatal.

• Yoshua Bengio — catastrophic-risk concerns combined with technical safety research and governance proposals.

• Geoffrey Hinton — Nobel Prize-winning pioneer of neural networks. He warns that systems smarter than humans may become difficult to control and could pose an existential risk, while also stressing AI’s potential benefits.

• Center for AI Safety — a short statement showing how widely extreme-risk concerns are shared.

• AI as Normal Technology by Arvind Narayanan and Sayash Kapoor — a critique of treating AI as an autonomous historical force instead of technology deployed by institutions.

• Melanie Mitchell and Gary Marcus — sceptical examinations of claims about reasoning, reliability and rapid progress toward AGI.

• International AI Safety Report — the best starting point I have found for separating evidence, disagreement and uncertainty.

6. Why I don’t think AI will replace people in every domain

This is my current conclusion, not a prediction that disruption will be mild. A job is a bundle of tasks, relationships, judgment, responsibility, trust, physical action and local knowledge. AI can automate parts of that bundle without becoming the person who accepts liability, understands an unspoken concern or lives with the consequence.

The ILO–NASK global index estimates that one in four jobs has some exposure to generative AI, while finding transformation more likely than wholesale replacement. That still leaves serious questions about eliminated roles, entry-level work, bargaining power and who keeps the productivity gains.

Andrew Ng’s human context advantage

In Andrew Ng’s conversation with Silicon Valley Girl, he argues that judgment and taste often come from context the AI does not have: experience, relationships, organisational history and memories of what failed. He also raises the problem of cognitive offloading: AI can help us complete a task without helping us learn it.

People studying AI and work

• Erik Brynjolfsson — augmentation, productivity and how organisations capture value from technology.

• Daron Acemoglu — the economics of automation and the difference between genuinely useful tasks and cost-cutting.

• David Autor — expertise, job quality and how technology can restore or erode middle-class work.

7. Why are people protesting AI—and what do data centres and job losses have to do with it?

People are rarely objecting to one thing called “AI.” They are responding to particular costs: job loss and weaker bargaining power; creative work used without permission; surveillance; misinformation; the loss of entry-level learning; concentrated corporate power; and the electricity, water, land and noise required by data centres.

A data centre is a building full of computers. Frontier AI requires large numbers of specialised chips, plus electricity, cooling, networking and backup power. The International Energy Agency explains the global energy picture. A study in Nature Sustainability examines possible US water and carbon costs. The UK government’s public-engagement survey shows that concern also centres on inaccurate information, privacy, deepfakes, jobs and damage to creative or critical thinking.

Where I go to understand the opposition and the physical costs

• SemiAnalysis — chips, power, networking and data-centre economics.

• UN Environment Programme — the environmental impact of AI across its full life cycle.

• Ed Newton-Rex and Fairly Trained — copyright, consent and licensed training data for generative AI.

• Ada Lovelace Institute — public interest, regulation, biometrics, work and the social effects of data-driven systems.

• International Labour Organization — evidence on which occupations and tasks are most exposed.

These sources do not agree on whether the costs are justified. That is useful: the central questions are who benefits, who pays and who gets a say.

8. Where I learn how people are using ChatGPT

This is one of the practical centres of the article. I pay for ChatGPT’s top-tier plan and use the newest models for research, writing, document analysis, coding and product work. But most of my imagination has come from watching other people share their screens: seeing the problem they start with, the context they provide, the mistakes the model makes and the workflow they eventually keep.

YouTube: the individuals I learn from

• How I AI with Claire Vo — probably the most directly useful format I have found. Guests share their screens and show how they research, write, manage products, build agents and run parts of their companies.

• Jeff Su — clear, edited walkthroughs of ChatGPT for research, projects, documents and everyday knowledge work. His ChatGPT Work walkthrough is a good introduction.

• Every and AI & I — conversations in which writers, founders and researchers demonstrate how they use ChatGPT and other models to think, create and build.

• Silicon Valley Girl — interviews about AI, careers and business. Her conversation with Andrew Ng broadened how I think about learning, judgment and human context.

• Kevin Stratvert — patient, beginner-friendly walkthroughs that are useful when a feature is unfamiliar and I want to see every click.

• Matt Wolfe — frequent experiments, comparisons and product news. I use him to discover what to test, rather than treating every first impression as a verdict.

Official reference: I use OpenAI’s channel to check release demonstrations and product announcements after seeing how independent teachers use the tools.

Substack, newsletters and written guides

• One Useful Thing by Ethan Mollick — research-informed experiments with ChatGPT at work and in education, including what happens when ordinary people are given frontier models.

• How I AI’s workflow library — searchable demonstrations organised by guest, company and tool. It is useful when I want an example for a particular kind of work rather than general advice.

• Every — writing about AI-assisted thinking, writing, coding and organisational design, produced by a team that also builds AI-native products.

• Simon Willison’s Weblog — detailed, reproducible testing of models, tools and security problems. It is one of the best places to understand what a new capability really does.

• The AI Exchange — Rachel Woods on turning responsibilities and processes into AI playbooks that teams can repeatedly use.

Official reference: OpenAI Learn is useful for checking current features and documentation, but it is not where most of my practical ideas come from.

Spotify: long-form examples and ideas

• How I AI — the audio version of Claire Vo’s practical workflow interviews. The video is better when the screen matters, but the conversations still generate ideas while walking or travelling.

• AI & I with Dan Shipper — writers, founders and creative people explaining how they use ChatGPT and Claude to think, write, code and make decisions.

• Practical AI — real-world implementations, architecture and the gap between an impressive model and a dependable system.

• The Cognitive Revolution — longer conversations with founders, researchers and policy thinkers that broaden the picture beyond one product or workflow.

Official audio: The OpenAI Podcast adds conversations with the people developing the models and with builders using them.

The limitation running through all of these sources is that fluent output is not the same as correct output. The demonstrations are starting points for experiments; important work still needs sources, privacy controls and human review.

9. Where I learn how people are using Claude

I also pay for Claude’s top-tier plan. I use it most for long documents, careful writing, analysis and, increasingly, Claude Code and agentic work. Claude becomes easier to understand when I watch somebody build an entire workflow with it—not only a polished final result, but the planning, context, permissions, corrections and review that made it work.

YouTube: people teaching Claude and Claude Code

• How I AI with Claire Vo — real Claude and Claude Code systems for product management, research, writing, company knowledge and agents, normally shown on screen rather than merely described.

• Jeff Su — accessible demonstrations of Claude for knowledge work and design. His Claude Design video shows how reusable context and a design system change the output.

• Brock Mesarich — explanations of Claude and Cowork for people who are not engineers. His Cowork concepts video covers context, memory, skills, connectors and subagents.

• Sabrina Ramonov — end-to-end Claude Code builds, agents, automation and reusable skills, often aimed at people learning to build without a traditional software background.

• IndyDevDan — opinionated, principle-led material on agentic engineering, planning, context files, subagents and production discipline.

• Cole Medin — technical builds combining Claude Code with agents, retrieval, MCP servers and automation.

• Every — interviews and demonstrations of how Claude Code is changing software, writing and small-team company building.

Official reference: After watching the independent teachers, I use Anthropic’s channel and the Claude Code team’s workflow video to compare those methods with the product team’s own approach.

Substack, newsletters and written guides

• Every — one of the richest collections I have found on Claude Code, AI-native products, compound engineering and the effect of agents on knowledge work.

• Simon Willison’s Weblog — careful testing of Claude releases, computer use, coding agents, security and prompt injection, usually with enough detail to reproduce the experiment.

• Latent Space — deeper writing and interviews for people moving from using Claude to engineering products and agent systems around models.

• Sabrina Ramonov — practical guides and reusable material for Claude, agent workflows and automation.

Official reference: Anthropic Engineering and Anthropic Academy are where I check the technical detail once somebody’s demonstration has given me an idea worth trying.

Spotify: Claude, agents and building

• Claude Code for normal people from How I AI — Grace Clarke demonstrates skills, voice mode and service-business tools she built with Claude.

• AI & I — Dan Shipper’s conversations with people using Claude and Claude Code for writing, software, research and running companies.

• OCDevel Claude Code — a technical, hands-on podcast covering session design, context files, hooks, permissions, skills, subagents and MCP.

• Latent Space: The AI Engineer Podcast — conversations about model infrastructure, agents and the engineering decisions behind AI products.

• You Need to Learn Claude Code in 2026 by Sandeep Swadia — an accessible account of the shift from chatting with a model to delegating multi-step work.

• Claude Code AI Subagents by Jake Morrison — an AI-narrated audiobook about specialised agent teams. I include it for discovery, while treating its performance and cost claims as ideas to verify.

Tutorials date quickly because Claude changes quickly. I use the independent sources for ideas and working methods, then return to Anthropic’s documentation and my own experiments before adopting a workflow.

10. Where do Microsoft Copilot and Google Gemini fit?

I have spent much less time with Copilot and Gemini, so I do not want to pretend to offer a deep personal comparison. Copilot is most relevant where work already lives inside Microsoft 365, Windows, GitHub or Azure; Microsoft’s Copilot documentation and the Phi model family are the useful primary sources.

Gemini sits across Google Search, Workspace, Android, Cloud and DeepMind’s model work. The Gemini overview shows the product family, while Sundar Pichai’s conversation with Rowan Cheung is a better source for Google’s wider strategy.

From learning about AI to building with it

Over the last few weeks my learning has shifted from reading and experimenting to building. These are the projects where I am testing what the technology can and cannot do:

• CompanyBoard.org — a dashboard that brings UK company deadlines, directors, filings, documents and compliance information into one place.

• Loop AI — a relationship layer for subscription and membership businesses that identifies customers needing attention and drafts a personal follow-up for human approval.

• Buy Back Brixton and BrixtonCulture.com — connected work around community ownership and making Brixton’s culture more visible and economically powerful.

The simplest doorway to all of this work is gvdp.co.uk.

What I am learning now: deploying agents at scale

I am now looking at how several agents can use tools and company data, divide work, check one another and escalate consequential decisions to people. The difficult questions are permissions, security, evaluation, cost, monitoring and recovery when something fails.

• Model Context Protocol specification — the authoritative description of the open standard that connects AI applications to tools and data.

• Zapier MCP — a practical example of connecting services such as calendars, email and CRM to ChatGPT and Claude.

• Anthropic on trustworthy agents — research on autonomy, safeguards, permissions and prompt-injection risk.

Building AI-native organisations

“AI-native” interests me as organisational design, not as adding a chatbot to an old process. These sources explore how companies might redesign work around context, feedback, evaluation, customer trust and human approval:

• Garry Tan: New Rules for Founders — leverage, reusable instructions and building companies in the AI era.

• Paul Graham on startups and great founders — why AI progress can be astonishing in one task and unreliable in the next.

• Andrew Ng: Generative AI for Everyone — a non-technical course on capabilities, limitations, business opportunities and social consequences.

Marketing and sales

The useful question here is not only how to produce more content. It is how AI can help teams understand customers, analyse conversations, explore positioning, create variants and run better experiments without generating more spam.

• Marketing AI Institute — structured education for marketing and business leaders.

• Claire Vo’s How I AI — screen-shared demonstrations of real workflows.

• Gong Labs — analysis of sales conversations and deal activity; useful company research that should be read with Gong’s commercial interest in mind.

Where I go to keep learning

These are the sources I return to, grouped by what they help me understand.

My current top three AI podcasts

If I had to reduce a very long list to three, these are the ones that currently give me the widest spread of perspectives:

1. Dwarkesh Podcast — my choice for long, deeply prepared conversations with the people building, studying and criticising frontier AI. It is where I go when I want an argument developed properly rather than compressed into a clip.

2. Google DeepMind: The Podcast — Hannah Fry makes frontier research, AlphaFold, agents, robotics and safety understandable without removing the science. It is produced by DeepMind, so I treat it as an excellent primary source rather than independent coverage.

3. How I AI — Claire Vo asks people to show the real workflows they use. It is the most practical of the three; the video edition is especially useful because the screen-sharing is part of the lesson.

Newsletters and podcasts for practical AI

• One Useful Thing — Ethan Mollick’s research-informed writing about AI, work, education and everyday practice.

• How I AI — Claire Vo’s screen-shared workflows. Its strength is seeing exactly how somebody does the work.

• Sabrina Ramonov — tactical material on Claude, agents and automation, sometimes presented alongside bold commercial claims.

• Elena’s Growth Scoop — AI through the lens of growth, product and marketing.

• Lenny’s Newsletter — broader product and growth thinking with real implementation stories.

Newsletters and podcasts for understanding the frontier

• Interconnects — Nathan Lambert on models, post-training, open AI and frontier research.

• Understanding AI — Timothy B. Lee and colleagues explaining technical developments with limited jargon.

• Import AI — Jack Clark’s detailed research and policy digest. Clark is an Anthropic co-founder, so it is insider analysis with an identifiable institutional perspective.

• Dwarkesh Podcast — deeply researched interviews rather than quick news.

• The Algorithmic Bridge — Alberto Romero on AI, culture, philosophy and society.

• SemiAnalysis — chips, compute, data centres, networking and the economics beneath the models.

• AI as Normal Technology — a rigorous counterweight to inevitability, hype and simplistic AGI narratives.

YouTube for using the products and building

• Jeff Su — unusually clear, structured product walkthroughs. The two videos linked above cover ChatGPT Work and Claude Design.

• Silicon Valley Girl — Marina Mogilko’s interviews on AI, business and careers. Her conversation with Andrew Ng is where I encountered the ideas about cognitive offloading and human context advantage. Some titles use aggressive “top 1%,” income and future-of-work framing, so I focus on what the guests actually say.

• Brock Mesarich — accessible explanations of Claude and Cowork. I have linked his Claude Cowork concepts video in the Claude section.

• Matt Wolfe — frequent AI news, tool testing and practical demonstrations. He also operates FutureTools, so I use him for discovery rather than treating a product mention as an independent verdict.

• Y Combinator — founder interviews, startup advice and discussions about building AI-native companies. It is also an investor’s view of the market.

YouTube for frontier research and deeper conversations

• AI Explained — independent commentary on model releases, research papers, benchmarks and the path towards more general systems.

• Two Minute Papers — visual, enthusiastic explanations of research in AI, graphics and simulation. It is an accessible doorway into a paper, not a substitute for reading the paper.

• Dwarkesh Patel — long, heavily prepared interviews with researchers, lab leaders, economists and critics.

• Google DeepMind — research demonstrations, explainers and video editions of its podcast. It is a primary company source, not independent coverage.

• Latent Space — models, agents, tools and infrastructure for people who want to move from general interest into AI engineering.

• Machine Learning Street Talk — long technical conversations about machine learning, cognition and AI philosophy. This is the deep end rather than a beginner channel.

Spotify and audio on Claude Code and agents

• You Need to Learn Claude Code in 2026 — Sandeep Swadia’s episode on the move from chat-based assistance towards agentic coding.

• Claude Code AI Subagents — Jake Morrison’s independently published, AI-narrated audiobook. I have included it for discovery, but its performance and cost claims still need checking against primary sources and direct experiments.

A wider map of people I follow

Some of these people appear earlier in the article. I have brought them together here by the part of the AI conversation they help me understand. This is not a ranking, and the categories inevitably overlap.

Intellectuals and lab leaders

• Dario Amodei — essays on frontier capabilities, interpretability, safety and the possible scientific and economic benefits of powerful AI.

• Demis Hassabis — AGI, scientific discovery, AlphaFold, robotics and DeepMind’s long-term research direction.

• Sam Altman — OpenAI’s case for abundant intelligence, infrastructure and broad access; his longer writing is at samaltman.com.

• Geoffrey Hinton — the foundations of deep learning and his warnings about systems becoming more intelligent than their operators; the Nobel interview is a useful introduction.

• Yann LeCun — open research, world models and a prominent critique of near-term AGI and extinction predictions.

• Yoshua Bengio — deep learning, AI safety, governance and research on controlling increasingly capable systems.

• Fei-Fei Li — computer vision, spatial intelligence and human-centred AI; she also co-founded Stanford HAI.

• Mustafa Suleyman — consumer AI, the politics of powerful technology and how Microsoft is turning models into widely used products.

• Yuval Noah Harari — how AI may reshape institutions, democracy, stories, power and human agency.

• Erik Brynjolfsson — productivity, augmentation and how organisations capture—or fail to capture—the value of new technology.

• Daron Acemoglu — automation, labour markets, inequality and the economic choices behind how AI is deployed.

Practitioners and educators

• Ethan Mollick — research-informed experiments with AI at work, in education and as a creative collaborator.

• Andrew Ng — accessible courses covering foundations, business use, machine learning and agentic systems.

• Dror Poleg — AI’s effects on work, organisations, cities and markets, alongside practical workshops.

• Claire Vo — screen-shared demonstrations of how people use AI in real professional workflows.

• Allie K. Miller — workshops and AI-first operating approaches for leaders and teams.

• Rachel Woods — AI operations and turning organisational processes into documented, repeatable playbooks.

• Dan Shipper — experiments, writing and products created by an organisation trying to work AI-natively.

• Paul Roetzer — marketing education, organisational adoption and practical AI workshops.

• Sabrina Ramonov — tactical material on Claude, agents, automation and AI-assisted building.

• Elena Verna — AI through the lens of growth, product, retention and marketing.

Builders and researchers

• Andrej Karpathy — language models, AI coding and clear technical education; his YouTube lectures are especially useful.

• Simon Willison — hands-on model testing, AI-assisted development, security and prompt injection.

• Jeff Dean — large-scale systems, research infrastructure and Google’s work on machine learning.

• Nathan Lambert — open models, reinforcement learning, post-training and research culture.

• swyx — AI engineering, agents, developer tools and conversations with the people building them.

• Jim Fan — agents, robotics, simulation and embodied AI at NVIDIA.

• Dylan Patel — chips, compute, networking, data centres and the economics beneath frontier AI.

• Ilya Sutskever — foundational model research and the pursuit of safe superintelligence through SSI.

Operators, founders and investors

• Tobi Lütke — Shopify’s push to make reflexive AI use part of everyday company operations.

• Jack Dorsey — open-source software, decentralisation and user-controlled agents such as Goose.

• Garry Tan — AI-native startups, founder behaviour and what very small teams can now build.

• Aaron Levie — enterprise agents, organisational data and the future of workplace software.

• Sebastian Siemiatkowski — Klarna’s highly visible experiments with AI in customer service, operations and employment.

• Farhan Thawar — how Shopify deploys AI tools, shared infrastructure and MCPs across an organisation; this Shopify case study is the best starting point.

• Marc Andreessen — the accelerationist case for AI, the investment landscape and technology-led abundance.

Critics and counterweights

• Arvind Narayanan and Sayash Kapoor — evidence-led scepticism and the argument that AI should be understood as a powerful but socially embedded technology.

• Melanie Mitchell — the gap between impressive performance, reliable reasoning and general intelligence.

• Gary Marcus — persistent criticism of language-model reliability, industry claims and the route to AGI.

• Timnit Gebru — training data, bias, labour, concentrated power and the social costs often excluded from capability debates; she is active on Bluesky.

• Emily M. Bender — computational linguistics, the limits of language models and resistance to anthropomorphic AI claims; she co-hosts Mystery AI Hype Theater 3000.

• Ed Zitron — an intentionally combative critique of AI economics, corporate incentives and industry hype, also available through the Better Offline podcast.

Curators and journalists

• Dwarkesh Patel — deeply prepared, long-form interviews with frontier researchers, lab leaders, economists and critics.

• Jack Clark — a detailed research and policy digest written from inside the frontier-lab world.

• Timothy B. Lee — technical and policy developments explained with limited jargon.

• Alberto Romero — AI, culture, philosophy and society.

• Azeem Azhar — the interaction between exponential technologies, economics, institutions and geopolitics.

• Ben Thompson — the business strategy, platform economics and competitive structure of the AI industry.

• Rowan Cheung — a fast-moving daily overview of launches and demonstrations, useful for discovery rather than final judgment.

• Kevin Roose and Casey Newton — accessible technology journalism across the Hard Fork archive. They ended the New York Times show in August 2026 and are building a new independent programme; Newton continues to publish Platformer.

On my list—not yet read or watched

This is my queue, not a list of endorsements. I have collected these sources because they may add something missing from the picture, but I have not yet examined them closely enough to place them in the main sections.

Material already collected

• The Moonshots roundtable with Peter Diamandis and others, for speculative conversations about agents, mathematics, scientific discovery and AGI.

• Charlie Chang’s 12 Life-Changing Ways to Use ChatGPT, a broad survey of personal and business uses that also touches on automation.

Perspectives I want to add

• African AI: Masakhane on open research for African languages; Deep Learning Indaba on African AI research communities; and Lelapa AI on resource-efficient language technology designed in Africa.

• Britain and Europe: Azeem Azhar’s Exponential View; the Ada Lovelace Institute; the UK AI Security Institute; and the European Commission’s AI Act material.

• China and technology policy: Jordan Schneider’s ChinaTalk.

• Copyright and creative work: Ed Newton-Rex’s Fairly Trained, which advocates licensed training data and certifies qualifying AI companies.

An evolving record

Looking across these sources, I do not see one settled story about AI. I see overlapping arguments about capability, productivity, creativity, scientific progress, corporate power, employment, environmental cost, safety and control.

That is why this page will never really be finished. Models will change, new voices will emerge and some of today’s commentary will age quickly. I will keep returning to the post to add people, replace outdated material and record what I am reading and learning next.

What I’m still looking for

This list is still heavy on US labs and light on almost everything else. If you follow someone strong on AI in Africa, China or Europe; on copyright and creative work; on healthcare or government use; or on projects that did not work, reply to this email or leave a comment. I read everything and will add the best suggestions to the queue above, with credit.

Impact Brixton had just been through one of the most difficult summers in our history.

We had lost three of our largest customers and, with them, close to 30% of our revenue.

The reasons were understandable. As companies grow, their needs change. They want things like lifts, bike storage, larger facilities and the sort of infrastructure that bigger coworking operators can provide.

We couldn’t.

Impact Brixton has five private offices. We don’t have a lift. We don’t have a bike shed. There are physical constraints to what our building can ever become.

And when your revenue has just fallen by almost half, those constraints feel considerably less philosophical.

So you can probably understand why, the night before an 8am team away day, I was thinking:

Why am I going to spend half a day climbing walls when my business is close to failing?

I had emails to answer. Decisions to make. Problems everywhere.

Climbing felt very low down the list.

Thankfully, our General Manager Rachael wasn’t having it.

She reminded me that I’m constantly talking about the importance of us spending proper time together as a team.

Essentially: Gerald, we’re going.

She was right.

And I’m incredibly glad she was.

It started with croissants

We had originally been introduced to Leila at Substation through a friend and member, Zac Munroe, an Impact Brixton client and architect who had experienced it himself and couldn’t recommend it highly enough.

Leila and I started talking about how two Brixton businesses could work together.

Impact Brixton has its rum-tasting experience. Substation has its corporate climbing and strategy experiences.

Why not swap?

More importantly, why not find ways for the communities around our businesses to know one another?

Substation generously invited our team to experience one of its corporate mornings.

Eight of us arrived at 8am.

Some enthusiastic.

Some sceptical.

Me, perhaps the most distracted of everyone.

The first thing I remember is breakfast.

Croissants, pains au chocolat, coffee. It felt somewhere between a French boulangerie, a buzzing café and a gym.

But there was something else I noticed almost immediately.

People knew each other.

People talked.

It didn’t feel like a room full of anonymous gym memberships. It felt like a community that happened to have climbing walls in it.

I put my laptop down on a table.

Someone essentially told me it was perfectly safe there.

That tiny interaction told me something about the place.

There was trust.

You could feel that whoever had built Substation cared about more than getting people through the door.

Of course, I then ruined the zen by repeatedly going back to my laptop.

Emails.

Messages.

Problems.

Business.

Leila noticed.

She didn’t tell me off. She smiled.

She acknowledged that there was obviously a lot going on in my head and said something that I didn’t fully believe at the time:

Once you start climbing, you’re going to become incredibly present.

You won’t be thinking about the emails.

She was absolutely right.

Then we started climbing

I had never bouldered before.

And I had massively underestimated it.

Looking at somebody else climb, there is a temptation to think:

How difficult can this actually be?

Quite difficult, as it turns out.

We were split into two groups of four and taken through the fundamentals by the Substation team.

The first revelation was that climbing isn’t simply about strength.

It’s a problem.

Where does your foot go?

Where should your weight be?

Which hand should move next?

Can you reach that hold?

Should you even be trying to reach it from that position?

The stronger person isn’t necessarily the better climber.

Technique matters.

Balance matters.

Planning matters.

And attention really matters.

We got through some of the easier routes. Then we started meeting problems that weren’t so easy.

We failed.

We tried again.

We failed differently.

Someone on the floor would spot something you couldn’t see from the wall.

“Move your foot there.”

“Try that hold.”

“No, the other one.”

And slowly eight colleagues who had arrived at an early-morning corporate away day started laughing, shouting instructions at one another and trying to solve these strange physical puzzles together.

Daniel surprised all of us. He absolutely took to it.

Kylie had started the morning by joking that the team seemed to be full of injuries. She has problems with her shoulder. Others had their own aches and limitations.

Yet once we were engrossed in climbing, something interesting happened.

People stopped defining themselves by what they couldn’t do and started concentrating on the next move they could make.

You can’t worry about Tuesday while you’re hanging from a wall

This was the big revelation for me.

Leila had been watching me return to that laptop all morning.

Then I started climbing.

And I didn’t touch my laptop or phone for the next two hours.

Not because somebody had confiscated them.

I forgot about them.

For somebody whose brain is nearly always running across multiple problems, businesses, emails, decisions and ideas, that is incredibly unusual.

Bouldering demands your attention.

Not most of it.

All of it.

When you’re on that wall trying to work out where your left foot goes while your fingers are telling you they would quite like you to make a decision soon, there isn’t much cognitive capacity left for an email somebody sent you yesterday.

You cannot worry about Tuesday while you’re hanging from a wall.

The consequence of losing focus is immediate.

That changes everything.

We talk a lot about “being present”.

We meditate.

We put our phones away.

We install focus apps.

We go on retreats.

We tell ourselves to concentrate harder.

Bouldering gave me a completely different experience of presence.

It didn’t ask me to ignore everything else.

For those moments, everything else simply ceased to exist.

And I think this is where the experience became much more interesting than team building.

Because perhaps one of the reasons so many corporate strategy sessions are terrible isn’t that people lack ideas.

Maybe we’re asking people to think deeply while their minds are still somewhere else.

First movement. Then stillness.

After climbing, we went into yoga.

The contrast was wonderful.

We had gone from hanging from walls, falling onto mats and shouting instructions at one another to slowing everything down.

None of us were particularly accomplished yogis.

It didn’t matter.

The climbing had made us present.

The yoga centred that energy.

By the time we came together afterwards to talk about Impact Brixton, something had changed.

No phones.

No laptops.

No usual distractions.

And, perhaps more importantly, the normal dynamics of the workplace seemed to have shifted as well.

We had spent the morning looking slightly ridiculous together.

Failing together.

Learning together.

Being beginners together.

I was no longer the CEO with all the answers.

I was just another person trying to work out how the hell to get up a wall.

That vulnerability matters more than I realised.

Then we had one of our best strategy conversations in years

The question facing Impact Brixton was serious.

What do we do now?

Losing nearly half your revenue naturally makes you focus on what you’re missing.

We don’t have enough private offices.

We don’t have a lift.

We don’t have bike storage.

We cannot provide every facility that a growing company might eventually require.

The instinct is to view all of those things as disadvantages.

But sitting together after that morning at Substation, we started looking at the problem differently.

What if the constraint wasn’t something to fight?

What if it was telling us what Impact Brixton should actually become?

We realised that we’d been trying to be too many things to too many people.

Perhaps we shouldn’t be trying to compete for larger companies that eventually need the infrastructure of a WeWork-sized building.

Perhaps our job is to be brilliant for the people who need us most:

Freelancers.

People starting businesses.

Small teams.

People registering their first company.

Entrepreneurs who need community before they need a corporate headquarters.

Our five private offices suddenly weren’t evidence that we needed more private offices.

They were evidence that perhaps private offices shouldn’t define our future at all.

That was a fairly profound shift.

Instead of asking:

How do we overcome our constraints?

we started asking:

What are our constraints trying to tell us?

The quietest voices got louder

Something else happened around that table.

People who don’t always speak the most in our strategy meetings started contributing much more.

Cooper was one of them.

She helped push the conversation beyond the physical boundaries of Impact Brixton itself.

If our mission is to help people build businesses, why should every solution depend on someone physically occupying a desk?

That opened up a much bigger conversation around virtual office services and what it might mean to support a business from the very first day it exists — potentially creating ways for people to establish and connect their companies to Impact Brixton without requiring another square foot of office space.

That thinking is still evolving.

But that’s precisely the point.

The constraints of the building stopped limiting our imagination about the business.

And watching Cooper contribute so strongly also made me notice something else.

Changing the environment had changed who felt able to speak.

The normal hierarchy had softened.

People were relaxed.

Optimistic.

Open.

Energised.

And I found myself listening differently too.

We had spent a morning helping one another do something none of us were particularly good at.

By the time we talked strategy, contribution wasn’t determined by job title.

It was simply about whether someone could see the next move.

Which, thinking about it now, is exactly what we’d been doing on the climbing wall.

We walked in worrying about the business. We left redesigning it.

I don’t want to overstate what can happen during one team away day.

Impact Brixton’s challenges didn’t disappear because we went bouldering.

The revenue didn’t magically reappear while I was doing yoga.

There is still difficult work ahead.

But something genuinely significant did happen.

We stopped seeing our limitations purely as failures.

We started seeing them as information.

And that is already changing how I think about the next version of Impact Brixton.

The team that walked into Substation was asking:

How do we replace what we’ve lost?

The team that walked out was starting to ask:

What if we build around what we’re uniquely good at instead?

Those are very different questions.

And strategy is often nothing more—and nothing less—than finding the better question.

Maybe we’re doing corporate away days wrong

The experience also made me question what we normally call “team bonding”.

We default to drinks.

Dinner.

The pub.

Christmas parties.

Perhaps an activity followed by alcohol.

There’s nothing inherently wrong with any of that.

But I now think we’re overlooking something much more powerful:

state.

If you want people to connect differently, change their state.

If you want them to think differently, don’t necessarily give them another brainstorming framework.

Change their environment.

Make them move.

Make them learn something.

Make them slightly vulnerable.

Let them fail at something completely unrelated to work.

Make the CEO a beginner again.

Get everyone off their phones.

Bring them completely into the present.

Then ask the difficult question.

Substation describes its corporate retreats as moving teams through physical challenge, stillness and conversation. Its strategy experiences combine climbing with activities such as yoga before teams move into dedicated time to think, plan and align. That sounds like marketing copy until you experience why the sequence works.

I experienced it.

It works.

And maybe Brixton businesses can do more of this too

There’s another reason I’m grateful to Leila and the Substation team.

This started because somebody in our community made an introduction.

One Brixton business introduced another.

We exchanged experiences.

Now we’re talking about what a deeper partnership could look like.

That’s how I want more of Brixton to operate.

Not simply as hundreds of businesses occupying neighbouring postcodes.

As an ecosystem.

Impact Brixton has a community.

Substation has a community.

Other businesses around us have theirs.

Imagine what becomes possible when those communities start crossing over, collaborating and genuinely recommending one another.

That’s something I want to explore much more through Impact Brixton and Brixton Culture Capital.

Local economic development doesn’t always require another grand programme.

Sometimes it begins with:

You should meet Leila. I think you’ll like what they’re building.

Someone did that for me.

I’m glad they did.

So, would I recommend it?

Completely.

If you have a team of five, ten, twenty or more and you’re facing a problem that genuinely matters, I’d particularly consider something like this.

Not because climbing will magically solve your strategy.

Because the quality of strategy isn’t determined by how many hours you spend in a meeting room.

It’s determined by the quality of insight you can reach together.

And sometimes the fastest route to that insight is to stop talking about the problem for a while.

Go somewhere unfamiliar.

Move your body.

Become a beginner.

Laugh at yourselves.

Put the laptops away.

Then come back to the question.

Substation currently offers corporate retreat formats ranging from movement-focused sessions to half-day and full-day strategy experiences, combining climbing, mindful movement and dedicated time for teams to work through what matters to them.

We only did part of what is possible.

Next time, I’d love to do the full strategy experience.

And there will definitely be a next time.

I arrived at Substation sceptical.

I nearly cancelled.

The following morning my hands were sore, my forearms were tingling and it felt like I’d done a better workout than almost anything I’d experienced in a gym.

More importantly, my team had helped me see my business differently.

We haven’t been back yet.

But we’re planning to.

And I think I may have found a new hobby in the process.

So thank you to Leila and everyone at Substation for looking after us.

Thank you to Rachael for making sure I didn’t cancel.

And thank you to my team for showing up, climbing, falling, stretching, thinking and speaking.

It turns out our best strategy meeting in years started when we stopped having a meeting.

And got on a wall instead.

Four months ago, I got fed up with CRMs. They’re management tools, not growth tools — and none of them are built for subscription businesses like the two I run. So I started building useLoop.

I thought I was building software for my own businesses.

Somewhere along the way, I realised I was watching something much bigger happen.

The gap between having an idea and building a product has collapsed.

useLoop now replaces large parts of my stack: website, email, bookings, forms, automation, customer communication and payments.

Tools I used to rely on — WordPress, Mailchimp, Optix, Cal.com, Zapier, Typeform, and others — are being consolidated into a single AI-native operating layer.

I’m not an engineering team.

I’m one person, working with Claude and Codex.

Then it clicked:

The world has quietly changed

A few years ago, building a product usually meant needing a designer, a developer, a product person, a budget, and a lot of waiting.

Now the gap between “I have an idea” and “I built a thing” is smaller than it has ever been.

But not everyone automatically benefits

That is the bit I care about.

Because if this shift is real, then it cannot belong only to Silicon Valley, Shoreditch, or the companies already closest to the money. It has to reach the founders, operators, creatives, small businesses and community builders who have always had ideas, but not always had access to the tools to build them.

Brixton Stripe Builders

That is why we are launching Brixton Stripe Builders — a community of builders based at Impact Brixton.

The idea is simple: put people with ideas, problems, skills and curiosity in the same room. Give them the tools, the confidence and a deadline. Then let them build.

Some people in the room will be working on AI infrastructure, data centres and software. Some will be writers, cultural thinkers, founders, freelancers, operators, small business owners, and people who have never shipped anything before.

That mix is the point.

This is not just about AI as a tech trend.

It is about who gets to use it.
Who gets to build with it.
Who gets to turn an idea into something useful.
Who gets to shape what the next ten years look like.

Come and build — Friday 14 August

Our first Build Day is happening tomorrow at Impact Brixton.

If you have already tinkered with Claude, ChatGPT, Claude Code, Codex, Vercel, MCPs, or anything similar, come and build.

If you are still early, come to the crash course first and then roll straight into Build Day.

Friday 14 August
Impact Brixton, 17a Electric Lane
Crash course: 1–2pm
Build Day: 2–5pm

Come build something for free, thanks to Stripe.
Come help someone else build something.
Come be in the room.

Sign up for Build Day →
Sign up for the AI Builder Crash Course →

Brixton Community is supported by Stripe.

GVDP

What follows is a historically grounded narrative based on my family tree, the documented history of the Bruce-Vanderpuije family, and the known history of the Dutch Gold Coast. Some details are well documented. Others are informed reconstruction, where the records run thin. I have tried to be honest about which is which.

I have always wondered where my history was.

The name was a clue I carried my whole life. Vanderpuye is not a Ga name. I always knew there was a Dutch thread somewhere. But knowing a name points across the sea is not the same as knowing how it got to you.

So I went looking. I spoke to my dad. I used Ancestry and 23andMe. This is what the digging gave me back.

The world the name came from

To understand why the Vanderpuye family exists in Ghana, you have to go back to the late 1700s.

Back then, the coastline of modern Ghana was one of the most important places in the world.

Europeans called it the Gold Coast.

The Dutch, British, Danes, Portuguese, Swedes and Brandenburgers all fought for influence there.

Along the shore stood massive forts:

• Elmina Castle

• Fort Crèvecoeur

• Cape Coast Castle

Gold flowed through these forts.

So did people.

And tragically, so did enslaved Africans.

My family story begins at the intersection of all three.

Jacobus arrives from the Netherlands

Around 1754, a boy named Jacobus van der Puije was born in the Dutch province of Zeeland.

His family came originally from a small town called Sint-Maartensdijk, and his father is recorded as William van der Puije.

Like many ambitious young Dutchmen with few prospects at home, Jacobus entered the service of the Dutch West India Company.

At the time, this was one of the most powerful corporations on earth. I think of it as a combination of:

• a government

• a military

• a trading company

• a shipping empire

all rolled into one.

As a young man, Jacobus was sent thousands of miles from Europe to the Dutch Gold Coast.

He did not arrive as a settler looking for land.

He arrived as a colonial administrator.

And he rose fast. By 1776, still in his twenties, h

e was governor of Fort Crèvecoeur at Accra. By 1780 he had become Acting Governor-General of the entire Dutch Gold Coast.

In modern terms, he was effectively running the Dutch colonial territory in what is now Ghana.

He was about twenty-six.

The part most families skip

This is where my story becomes complicated.

Many family histories try to avoid this part. But it is impossible to understand mine without it.

The Dutch Gold Coast was deeply entangled with the Atlantic slave trade.

The forts that men like Jacobus administered existed for a mix of reasons:

• gold trading

• military alliances

• colonial administration

• the buying and selling of enslaved Africans

Jacobus was not merely living near this system.

He was one of the senior officials administering the colony while it operated.

I won’t hide behind “he was only an administrator.” Administration is exactly how that trade was run.

And the records make it personal. One of Jacobus’s children was born to an enslaved African woman named Asoewa. Their daughter, Anna, was born enslaved too — and freed for the price of one mark of gold.

Another of his children was born to a Ga woman of Accra named Ayeley Ablah.

So within my own bloodline there are, at the same time:

• a European colonial administrator

• African women living under colonial rule

• and, in Asoewa, a woman who was herself enslaved

That reality is uncomfortable.

But it is also the truth of many Afro-European families on the Gold Coast.

My family story is therefore not simply a Dutch story.

Nor is it simply a Ghanaian story.

It is a story born from one of the most morally complex periods in human history.

Peter — the first of us born in Africa

Around 1775, Jacobus’s son Peter was born in Ghana, to Ayeley Ablah.

Unlike his father, Peter was African-born.

Unlike his father, Ghana was his home.

Peter grew up between worlds. He inherited:

• a Dutch surname

• European connections

• African ancestry

• ties to Ga society

His generation became part of a new class often called the Euro-Africans.

These families became enormously influential along coastal Ghana. They worked as:

• merchants

• translators

• diplomats

• intermediaries

between Europeans and African kingdoms.

Peter is the moment my family stopped being Dutch visitors and started becoming Ghanaian.

Becoming Ga

Over the next few generations, something remarkable happened.

The family became more and more woven into Ga society.

The Dutch bloodline remained.

The surname remained.

But culturally, the family became Accra.

Children married into local families. The Ankrah name appears again and again in my tree, alongside other prominent Ga lineages. The family became part of the social fabric of the city itself.

And the name itself evolved:

Van der Puije → Vanderpuije → Vanderpuye.

The spelling stayed European.

The identity became Ghanaian.

I want to be clear about how that happened, because it matters. The Ga pass important rights through the mother’s line. So the family’s standing in Accra — its place, its belonging, eventually its right to a stool — did not come down from the Dutchman.

It came through the women who married into it.

Chief John Nii Kodjoe Vanderpuije

By the mid-1800s the family had transformed completely.

Then came one of the most important figures in our whole history:

John van der Puije. Born 1848.

Unlike Jacobus, he was not a colonial governor.

He was a Ghanaian leader.

A merchant. A chief. A politician. A public figure.

His mother was Naa Karley Ankrah, of the royal house of Otublohum — one of the old ruling families of Accra. And that is why he could become a chief at all: not through his Dutch surname, but through his mother’s royal blood, by Ga custom.

He made the most of it:

• He traded in cocoa, the crop about to make the Gold Coast Britain’s richest West African colony.

• In 1880 he helped found the Gold Coast Chronicle, one of the first African-owned newspapers.

• In 1888 he was enstooled chief of Otublohum.

• In 1894 he was appointed to the Legislative Council of the Gold Coast.

His finest hour came in 1897, when the colonial government tried to pass a Lands Bill that would have let the Crown seize land held under indigenous tenure. John fought it from inside the Council. His newspaper fought it in print. A delegation went all the way to London.

The bill was withdrawn.

It is remembered as one of the first great victories of organised African resistance on the coast.

By this point my family was no longer being shaped by Europe.

It was helping shape Ghana.

The descendants of a Dutch governor and a line of Ga women had become leaders of Accra itself.

That is one of the most fascinating transformations I have ever read about — and it is mine.

The Bruce connection

The next major chapter is the rise of the Bruce-Vanderpuije branch.

Through marriage — again, through a woman of the Otublohum royal house — the family joined with another prominent coastal family, the Bruces.

The result is the name that runs into my own direct line: Bruce-Vanderpuije.

This branch produced James Kobla Bruce-Vanderpuije, who in 1922 founded the Deo Gratias photo studio in Jamestown — now the oldest operating photography studio in West Africa, holding an archive of some 50,000 images of Ghana, and still run by our family today.

Then Eric Baldwin Bruce-Vanderpuije, the generation that lived through independence in 1957.

Then Isaac Vanderpuye — my father, born in 1947, a boy of ten when Ghana became free.

And then me.

Gerald Vanderpuye

I am roughly seven generations removed from Jacobus.

Which means my ancestry holds, all flowing into one line:

• Dutch colonial administrators

• Ga women

• merchants

• chiefs

• politicians

• entrepreneurs

• cultural leaders

My family’s journey mirrors Ghana’s journey:

Dutch colonial period ↓ Slave trade era ↓ Euro-African merchant class ↓ Ga chieftaincy ↓ Colonial politics ↓ Modern Ghana ↓ Diaspora entrepreneurship

And in many ways my own work sits on the far end of that same arc. Impact Brixton, Brixton Culture Capital, my pull back toward Ghana — they all live at the intersection of:

• commerce

• community

• culture

• identity

the same four forces that have shaped this family for over 250 years.

The questions I’m still chasing

The honest end of any search is a list of what it still doesn’t know. Mine has four big ones.

1. Was William van der Puije really Jacobus’s father? My tree shows William. I haven’t yet seen primary documentation proving that link. It needs verification in the Zeeland records.

2. Did Jacobus die at sea, or in Ghana? My family always said he died at sea, sailing home to Holland. The historical record says he died on the Gold Coast in 1781, and never left. I lean toward the record — but I want to settle it properly.

3. Who exactly was Ayeley Ablah? This may be the most important missing piece of all. I know a great deal about Jacobus. I know almost nothing about the African woman who gave birth to Peter. And genetically, she contributes every bit as much to me as he does.

4. Was the family enslaver, enslaved, or both? Given the period, the line may include people on both sides of that history. That deserves careful investigation, not assumption.

Here is what I’ve landed on.

The most powerful version of my family’s story is not:

“A Dutch governor came to Ghana.”

It is:

“A family emerged from the collision of Europe, Africa, empire, trade, slavery, resistance, and Accra itself. Over 250 years, they went from colonial outsiders to one of the city’s established Ghanaian families.”

That story is far richer — and far more human — than the simplified legend most families inherit.

I went looking for where my history was.

It turns out it was never really in Holland.

It was in Accra the whole time.

If you’re family and you remember any of this differently — a name, a date, a story you were told — please tell me. The next thing I want to do is sit down with my dad’s generation and record everything they remember, while they’re still here to remember it. That’s the part no archive can give back.

Gerald@hey.com

No. It’s not going to zero. Sit down, pour some of my delicious Brixton culture rum — and let me walk you through it. Because this is my favourite time to write about crypto. Not when everyone’s a genius at the top. Now. When it hurts.

Bitcoin is tanking. It’s not the first time, and it won’t be the last. That’s the whole point.

“But Saylor sold”

Here’s what spooked everybody. Michael Saylor — the man who built an entire company around never selling Bitcoin, the guy whose whole personality is “we will never sell” — sold some Bitcoin. So that’s it, right? If even Saylor is out, it’s over.

Except look at the actual numbers. His company sold 32 coins. Thirty-two. They still hold 843,706. That’s 0.0038% of the stack, sold to cover a dividend payment — and Saylor’s own pledge is to buy 10 to 20 coins for every one he sells.

So yes, technically, the man who said never sell, s” sold. It’s a symbolic moment. But it’s not him giving up and running for the door. It’s a guy selling one tree and telling you he’s about to plant ten more. Don’t confuse the headline with what actually happened.

The thing nobody wants to remember

Bitcoin moves in a roughly four-year cycle. Every four years, it rips up to a new all-time high — that’s just the highest price it has ever traded at. And every four years — without fail — it falls off a cliff afterwards. This is not a bug. This is the asset.

Here’s every cycle so far:

Cycle peak All-time high (the peak) Biggest drop after Bottom

2011 ~$32 -93% ~$2

2013 ~$1,150 -85% ~$200

2017 ~$19,800 -86% ~$3,200

2021 ~$69,000 -77% ~$15,500

2025 ~$126,200 -50% so far ~$64,000 (today)

Read that bottom row again. We topped at about $126,200 in October 2025. Today we’re around $64,000. That feels brutal. It feels like the end.

It’s a 50% drop. By Bitcoin’s standards, that’s mild. Every single previous cycle fell at least 75% from the top. We’re not even there yet.

So where could this go?

If history rhymes — and with Bitcoin it always rhymes — a deeper fall from here wouldn’t surprise me at all. From the $126k peak:

• -70% → about $38k

• -75% → about $32k

• -80% → about $25k

Could we see the low- to mid-$30,000s? Absolutely. It would be painful. It would also be completely, boringly normal.

But here’s the thing: even though a 75% fall could happen — it has every other cycle — most people don’t actually think it will this time. Something’s changed. Big serious money is in Bitcoin now — banks, funds, the kind of investors who used to laugh at it. There’s more real money holding it than ever before, and that tends to cushion the fall. A lot of smart people think this cycle stops at a 60-70% drop instead of the usual 80%-plus. The flip side: that big money can also panic and rush for the door fast, and lately some of it has. So I’m not pretending I know the exact bottom. Nobody does.

What I’m doing

I’m walking, saving my Uber money and spending every penny buying.

The whole game runs backwards from how it feels. It feels safest at the top when everyone’s rich and confident, and most dangerous right here when it’s red, and the group chat’s gone quiet. But the top is where you overpay, and the bottom is where you load up.

I’m not telling you to buy — I’m not your financial advisor, this isn’t advice, it can absolutely go lower, only ever risk what you can afford to lose. I’m just telling you what I’m doing. Waiting. Watching. Buying :-)

Let’s go. 🚀

GVDP

AI is moving too fast. It’s a minefield. Every week there’s a new model, a new tool, another thread promising it’ll change everything — and a lot of it is noise. Filtering the hype from what actually matters has been one of the hardest parts of this.

And I do it full-time.

For the last two years, more than half my time has gone into learning by applying AI to real business problems. — not just reading and watching videos but using it across Impact Brixton, Brixton Culture Capital, and the SaaS products.

Plenty of it went nowhere. Some of it quietly changed how the whole business runs.

Now I’m at the part I was waiting for: it’s working. Real results, not slideware. Rebuilding our sites with AI drove roughly a 1,000% jump in search visibility and 500% more traffic — that’s not a demo, that’s the business.

And the thing that made the difference wasn’t a tool. It was a shift in how I think about it: from “AI as something I reach for” to “AI as the way the business is designed.” Yes, we all use ChatGPT or Claude, and we are still on the first side of that line. The gap between the two is bigger than it looks — and crossing it is where the results actually live.

A few people have asked me to walk them through it. So I’m opening up a small session for business owners who are already using AI and can feel there’s a deeper level they’re not reaching.

If you’re in London, you fancy passing through Brixton, and you run a business/creator/builder — this is for you.

Friday 12 June, 15:30–17:30, at Impact Brixton. Twenty people, dcocktails after.

Come for the ideas, stay for the network and drinks.

👉 https://luma.com/yc2fxv2z

Gerald

Like so many of my friends, I have caught the Claude bug!.

I’m a non-technical founder—no coding background. I studied computer science, but I sucked at developing. So no, I am not a developer!

Last weekend, I built three websites. I’m running a coworking space, launching a members bar, building a SaaS product, and writing a newsletter that now has 16,000 readers in 1year.

Here’s exactly what I am using.

Wispr Flow

Start here. It lets me speak, instead of type, accurately everywhere. Every prompt, every message, every doc. I use it all day, every day. If you’re still typing everything, you’re already behind before the AI even gets involved.

Atlas Browser + ChatGPT Max Top tier

Atlas is my main browser — it’s built around ChatGPT and runs underneath my entire day. It can submit my HMRC tax returns while I am working on something else. Email: I can’t figure out how to word it. Something I need to think through fast. It’s always there.

ChatGPT, powered by Deepresearch, powers the research behind my Brixton newsletter BCC — which I started a year ago with AI and now has 16,000 readers. ChatGPT surfaces what matters, what’s worth writing about, and what the community needs to know.

90% of my working context lives here. I’m not moving it. What’s crazy is I have typed 100,000+ words in ChatGPT since I started.

Brave Browser

For anything security-first, I switch to Brave. It has its own built-in AI and is my default whenever privacy matters—two browsers, two distinct jobs.

Claude Max .

Everyone is going crazy about Claude this week, especially as it’s in the news. The US government decided not to use Claude, but all my developer friends, everybody who’s building anything, is using Claude. Claude will blow your mind this weekend. Here’s what I’ve done. This is where I build things.

This weekend, I built three websites that would have cost me 30k + a few years ago. Have a look and give me feedback.

• impactbrixton.com — my Brixton coworking space

• bccmembersbar.com — private members bar, I am launching soon

• thisandthat. chat — DoBox, the SaaS I’m building

Three brands. Three sites. One weekend. No coding background.

Claude is my project studio. I come here with the intention to make something. It’s also the reason I’m awake at 2am — because when something is working, stopping feels wrong.

Grok Free

When I need the truth, I go to Grok. No filter, no hedging. It’s my verification layer. I dont do politics with AI, so I dont have to like Elon to use this :-). At least I don’t give my money to Elon :-). Yet :-)

Gemini Through Google Workspace

Already in my Impact Brixton Google Workspace subscription, so it costs me nothing extra. First stop for image generation — blog visuals, quick concepts, anything visual that needs to happen fast.

Midjourney Free for now

When the image actually matters — cinematic, atmospheric, brand-level — Midjourney is still the standard. I use it selectively. Nothing else comes close when quality is the point. But I’ve not used it for a month, so maybe it’s time to ditch Midjourney.

Read AI

Joins every meeting I’m in. Records every conversation on my phone. Sends me a summary and action list afterwards. Nothing falls through the cracks. It’s like having a chief of staff who never misses a word.

DoBox Free(for now) — I am part of the team building it

DoBox pulls commitments from across all your communications — messages, meetings, conversations — and turns them into trackable actions. I built it to solve my own problem. I use it every day inside this exact stack.

Using your own product is the fastest way to know if it’s real. thisandthat.chat

The total: ~£520/month

Three websites in a weekend. AI is helping me build and manage all my businesses.

I’m not losing sleep because I’m stressed. I’m losing sleep because I can’t stop building. There’s a difference. One is anxiety. The other is momentum.

The tools exist. The gap between people who use them and those who don’t is widening every single week. That’s not hype. That’s what I’m watching in real time.

Get in it. And maybe set an alarm for when to stop.

I know you are wondering? Did Gerald actually write this ?

— Gerald

Running Impact Brixton, building BCC, and shipping DoBox.

I’ve been fooling myself.

Not because GTD is wrong — but because the world it was built for doesn’t exist anymore.

This morning I woke up to 46 WhatsApps, 9 Telegrams, 200 Discord, 40 LinkedIn messages, 160 emails across seven accounts… and 28 Slack messages.

Why do we pretend this is normal?

Everyone’s juggling multiple projects, roles, and communities. I run a coworking space, build SaaS products, and make rum. But this isn’t really about me — it’s about how we all work now. Every single project comes with its own channel. There is no single place where everything lands. There never will be.

Most of what we call “productivity” today is just dealing with messages. Reply, skim, archive, snooze, move stuff around. The numbers go down, and it feels like progress.

It isn’t. It’s input management.

The real work is always hiding inside the messages:

“Can you send that?” “Let’s meet tomorrow.” “Can you approve this?”

Those aren’t messages. They’re commitments.

A friend asked me to meet on WhatsApp three weeks ago. I said yes. Forgot to calendar it. Missed it.

Maybe I’m careless — but a meeting agreed in chat shouldn’t rely on memory to become a calendar entry. The system is broken.

Commitments are buried everywhere. Slack. Email. LinkedIn DMs. Telegram. One “quick message” that quietly becomes a real obligation.

So GTD’s first step — capture — becomes a full-time job. You end up retyping your life from one app into another: copying Slack into a task app, forwarding emails to yourself, screenshotting threads, leaving messages unread as fake reminders.

Modern productivity isn’t doing work. It’s moving work around.

Here’s what’s actually broken, simply:

Most tools show you what’s new. What you actually need is what you need to do.

The real question isn’t “what’s unread.” It’s: what do I owe the world right now — and who’s waiting on me?

That’s why inbox zero is a fantasy. Even if you clear one inbox, the commitments are still hiding in five others.

We need a new layer. Not another inbox. Not another task app. An action layer — one that sits above all your channels and surfaces the commitments buried inside them.

That’s exactly what I’m working on with DoBox at this+that.

One place. Every channel. Just a clear view: here’s what you need to do.

Agents will eventually do more of this automatically. But I don’t live in the future. I live in this morning, and this morning, I need important things to stop getting dropped just because they showed up in the wrong app.

Where do your commitments get stuck most — and have you found anything that actually helps?

Gerald

That was always wrong. But now? In the age of AI? It might be startup-failing advice.

I see it constantly. Brilliant founders, heads down, shipping features into the void. Iterating on UI no one has seen. Debating architecture for a product with two users. Building, building, building — convinced that the right product, polished enough, will eventually find its people.

It won’t.

Here’s the uncomfortable truth about startups: building a good product is hard. Finding people who love it is harder. And yet, founders split their time as if the first problem is the only one worth solving.

It isn’t.

My framework is simple, and I’ll defend it anywhere: put 80% of your energy into finding the people who will love what you’re building. Put 20% into the build.

Not forever. Just at the start — when nothing is proven and every week of silence is a week of drift.

Why? Because the moment you find even 100 people who genuinely love what you’re building, the product roadmap writes itself. Your customers tell you what to build. Your energy goes where it should. Everything sharpens. Before that moment? You’re guessing in the dark and calling it conviction.

Now add AI to this.

We are entering an era in which building is becoming cheaper. You can spin up a website in an hour. Full applications in a day. In a few years — maybe sooner — the most technically complex software in the world will be within reach of a single focused person with the right tools.

If that’s true, then the build is no longer the moat.

The code is not the asset. The customer relationship is. The trust is. The resonance is.

The smartest start-ups I’ve watched recently aren’t marketing their product — they’re marketing the problem. They’re building an audience of people who feel the pain deeply before a single line of code is shown to them. They’re releasing to tiny cohorts. They’re creating believers, not just users. And they’re doing it with a team that’s heavy on storytellers, brand thinkers, and community builders — not just engineers.

That is the shift.

The new moat isn’t your product. It’s that people trust you to solve the problem.

And trust isn’t built by shipping more features. It’s built by showing up, naming the problem clearly, and making people feel like you know how to solve it before asking them to buy anything.

So if you’re a founder right now — especially one riding the wave of AI-powered building — stop and ask yourself honestly:

How much of my week was spent building? And how much was spent finding people who would actually care?

Build it, and they will come is long dead. Make them come. Then build it.

Adoption hasn’t stopped — it’s compounding

Crypto.com estimates global crypto ownership rose to about 741 million people in 2025, up from 659 million the year before. Bitcoin ownership alone sits at roughly 365 million: the user base keeps expanding. On-chain, Bitcoin addresses with nonzero balances are in the tens of millions. That’s not what a “dead” system looks like.

Stablecoins are having the WhatsApp moment for money

Chris Dixon recently described stablecoins as the “WhatsApp moment” for money — it used to cost my mum hundreds of pounds to talk to her mum in Ghana, and it currently takes a long time and costs a lot of money to move money around the world: stablecoins moved over $12 trillion last year. Moving money across borders is still expensive and slow, with multiple middlemen(Banks, Visa, Mastercard…) taking a cut. Stablecoins are proving there’s a cheaper, faster and fairer way — and adoption tends to follow cost and convenience.

Bitcoin doesn’t go to zero when people won’t stop buying it

The “it could go to zero” argument treats Bitcoin like a stock that can simply be abandoned. But Bitcoin isn’t just a ticker symbol. It’s a tool — a permissionless, globally transferable value you can custody yourself.

And then there’s something even simpler: belief with a bid.

I would keep buying Bitcoin if it fell to $1. Not because I’m blind to risk, but because I value what it represents. And I’m far from alone. When millions treat something as long-term savings technology, you don’t get zero. You get volatility. You get cycles. You get headlines declaring death every few years.

And underneath all of it, you get growth.

So yes — I’m buying.

Not because it’s guaranteed. Not because it can’t fall further. But because adoption continues to rise, infrastructure keeps improving, and the long-term thesis hasn’t changed.

This isn’t investment advice. It’s simply how I’m positioning myself.

Crypto isn’t dead. It’s maturing.

Read more

Read more

For years, one of the biggest tools I have used to help me with my personal and business growth has been growing my network with cold and warm emails to super mentors and connectors. Thought leaders in the industry that could help you with strategy, connections, money, etc. They are too busy, and so unless you know them or know someone who knows them, y…

Read more

Read more

Read more

There has probably never been a better time to start a coworking space! Just avoid these mistakes, and you wi…

Read more

Read more

Hey, I get it. You're probably thinking, "Gerald, come on, a company going bankrupt is serious stuff. People will lose their jobs and livelihoods!." You're right; bankruptcy should be the last resort for any business in trouble. But hear me out. Sometimes, a company needs a shake-up. With WeWork's Chapter 11 Bankruptcy, they can …

Read more

Why "Wantrepreneurs"?

I've got to be honest; I've always wanted to be a proper entrepreneur. But the truth is, I'm still there. I'm a "wantrepreneur” who's always buzzing with ideas, buying up domain names, and registering companie…

Read more