New Decentralized AI Project and Philosophy x Entrepreneurship Series (Sept 2025)

Plus: Apply to attend our AI x Free Speech Symposium with FIRE

Dear Cosmos community,

This month, our first Cosmos Incubation Fellow launched the Institute for Decentralized AI with 5 fully funded positions at Oxford and Stanford—apply by October 8th. We also premiered a new Philosophy x Entrepreneurship video series, featuring top founders and investors. And Brendan delivered the Adam Smith Lecture in Edinburgh on Adam Smith, AI, and the Outsourcing of Moral Sentiments.

🎤 AI x Free Speech Symposium: Limited Spots Available

We’re co-hosting the AI x Free Speech Symposium with FIRE on December 5-6th in Austin, Texas, moderated by Jonathan Rauch, author of The Constitution of Knowledge.

This intimate gathering brings together top AI researchers, policy experts, and select grantees working on AI for truth-seeking. We have limited spots available for exceptional applicants.

🔬 Research

Cosmos Incubation Fellow Samuele Marro launched The Institute for Decentralized AI, combining research and infrastructure to support AI that operates without centralized gatekeepers. The Institute has 5 fully funded academic visitor slots at Oxford and Stanford for researchers working on agent security, anomaly detection in multi-agent systems, and decentralized safety. Apply by October 8th.

The Human-Centered AI Lab at the University of Oxford moved into new premises at the Schwarzman Centre for the Humanities. The lab, which Cosmos helped launch last year, marked its first on-site day with a visit from Nvidia’s Startup Ecosystem Director, a presentation on AI-enabled Capybara toys from an Alpha School student, and deployment of a local cluster for full-size DeepSeek V3.1 inference.

✨ Fresh Ideas from the Cosmos Ecosystem

We launched a Philosophy x Entrepreneurship video series, where we explore the connections between deep thinking and practical building with world-class founders. Episodes released so far:

• Colin Moran, whose top hedge fund hires humanities scholars, on faith and investing

• Francis Pedraza, from broke to $2B AI founder, on heroism and the Great Books

• Marcus Ryu, who built a $16B startup, on using Marx’s diagnosis, not his cure

Future episodes include Peter Thiel on his philosophical influences, MacKenzie Price on reimagining education, and Joe Lonsdale on fighting bureaucracy.

Brendan did a 2hr interview on AI and human autonomy, deep-diving into the arguments for preserving our capacity for deliberation in a world where machines can do the thinking for us:

Cosmos essays and guest essays:

• Brendan McCord on Western Dynamism

• Harry Law on Is Algorithmic Mediation Always Bad For Autonomy? and AI Doesn’t Seek Truth, People Do

• Séb Krier on Coasean Bargaining at Scale

• Zachary Gartenberg on Learning and Authentic Learning in the Age of AI

Articles and talks from our network

• Seth Lazar on AI Agents and Democratic Resilience

• B. Scot Rousse on The Limits of Autonomy (and Brendan’s response)

• Alex Komoroske on Why Centralized AI is Not Our Inevitable Future

• Michael Dean on The Architecture of Essays

• Luke Drago on Breaking the Intelligence Curse

🎙️ Talks and Events

Brendan delivered the Adam Smith Lecture at Panmure House in Edinburgh, where Smith completed his masterworks. He argued that outsourcing moral judgment to AI hinders our ability to develop the independent moral reasoning Smith saw as essential for flourishing. Congratulations to Ashley Zhang and Alaka Halder, who won last month’s essay contest and attended the lecture and dinner in Edinburgh.

Brendan also gave talks to recent high school graduates at Palantir on Western Dynamism and with FIRE on the future of free speech.

🏗️ Grants

In August, we welcomed 40+ new projects to our grant program. In September, we had a kick-off call with our general track and HAI lab winners. Projects ranged from studying how AI impacts critical thinking in universities to building more transparent and truth-oriented AI systems.

We’ll be announcing our next batch of winners soon, as well as a new program to support AI use in advancing research in the humanities and social sciences.

📚 What We’re Reading

• The Blank Page Revolution | Steven Johnson

• AI Rule and a Fundamental Objection to Epistocracy | Sean Donahue

• A School of Strength and Character | Tanner Greer

• Defeating Nondeterminism in LLM Inference | Thinking Machines

• Real AI Agents and Real Work | Ethan Mollick

• Why Language Models Hallucinate | Kalai et al

• Unlocking a Million Times More Data for AI | Andrew Trask, Lacey Strahm

• The Personality Illusion | Han et al

• The Algorithm Will See You Now | Deena Mousa

• Devon Zuegel is Building a New Town | Colossus

🤝 Get Involved

• Write: Pitch an article for how AI can support human flourishing. Submit pitch→

• Build: Apply for a $1k-10k+ fast grant for AI prototypes. Express interest→

• Research: Join our fellowship at Oxford’s HAI lab or partner institutions. Express interest→

• Learn: Attend seminars with Oxford, St John’s College, and Liberty Fund. Sign up for updates→

• Join our team: World-class and mission-aligned? Pitch us a role→

Other opportunities:

• The Institute for Decentralized AI has 5 fully funded visitor positions at Oxford and Stanford. Apply by October 8th→

• Workshop Labs is hiring ML research and infrastructure engineers to build personal AI agents that support human autonomy.

• Edge City is hosting a d/acc Residency for 15 builders working on AI, robotics, governance, cybersecurity, and information integrity.

• Cosmos Fellow Ryan Kearns presents “Agentic AI in Practice” on October 17th at Oxford’s Generative AI Summit. Register→

It's time to (philosophically) build,

Brendan

---

For readers joining us this month:

Cosmos Institute is the Academy for Philosopher-Builders. Our mission is to ensure AI becomes civilization's greatest force for human flourishing by expanding truth-seeking, preserving autonomy, and enabling decentralized progress.

---