Apply for funding, compute, and office space at our new AI Nodes in San Francisco and Berlin
New report with 208 proposals for reducing AI risk
Plus: details on our upcoming Vision Weekend Puerto Rico, talks from Vision Weekend USA, and tickets to our first-ever Vision Weekend in the UK!
Vision Weekend Puerto Rico 2026
February 6-8 | Old San Juan | Last chance to get tickets
It’s only one week until we bring Vision Weekend to Puerto Rico! Join other futurist researchers and builders to explore the frontiers of AI, nanotech, brain–computer interfaces, and longevity biotech – while enjoying the sunshine, beaches, and warm nights of the Caribbean.
Beyond the 30+ keynotes, you can look forward to discussing your favorite topics in depth with the speakers, finding your next collaborators, and pitching your project for a chance at the $10,000 prize.
Speakers include:
Robin Hanson, George Mason University
Ramez Naam, author of the Nexus trilogy
Jean Hebert, ARPA-H
Lauren Wagner, Abundance Institute
Samuel Hammond, Foundation for American Innovation
Our Norm Hardy Prize is a $10,000 award recognizing advances in usable security, with the long term goal to encourage interaction designs that make it easy for people to use secure systems securely.
We are excited to announce Dr. Pardis Emami-Naeini as the winner of the 2025 prize!
Pardis Emami-Naeini is recognized for her contributions to usable security.
Her research focuses on empowering consumers to make more secure choices through clear, accessible security information. By developing a layered cybersecurity label for smart home devices, she has helped shape national policy and industry standards, including the new U.S. Cyber Trust Mark.
Learn more about the winning work from her presentation at the award ceremony:
We are excited to introduce our 2026 cohort of Foresight Fellows!
With our Fellowship program, we want to support early-career talents advancing important science and tech. We connect them with senior scientists, invite them to our technical workshops, seminars, and Vision Weekends alongside leaders in their fields, and offer platforms for sharing their work.
This year’s group includes a neuroscientist studying how the brain computes at a molecular level, a bioengineer 3D-printing human tissue, a PhD student advancing computational protein design, and an engineer developing robot swarms to build large-scale structures in space – just to name a few.
You can learn more about the Fellows and their work on our website.
New grantees
We are excited to share new projects we are supporting with our grants!
As the seL4 microkernel has been proved to be correctly implemented with respect to its specification and also enforce the core security properties of confidentiality, integrity and availability, it is the most solid foundation for building secure systems. However, reasoning about the correctness (and security) of programs running on top of seL4 (including the actual operating system) requires that the kernel’s specification of the system calls it provides actually does what these programs expect.
To date, no such proofs connecting any kernel’s specification with its user programs’ requirements have been completed for any system calls that wait for a response from another program before returning, which puts any reasoning about the correctness of systems that use them on shaky ground. This project will develop the formalism and infrastructure to carry out such proofs for seL4.
Ariel Procaccia & Talmo Pereira
Reaching consensus via generative social choice | Ariel Procaccia | Harvard University
This project aims to develop, evaluate, and deploy a group decision-making process that combines social choice theory with large language models. The process will be capable of handling a vast space of alternatives (such as all possible policies) and will provide guarantees to promote safety and trustworthiness.
Interpretable neuroAI for motor control: massively parallel imitation learning of musculoskeletal dynamics | Talmo Pereira | Salk Institute for Biological Studies
This project uses imitation learning with a detailed musculoskeletal model of the mouse forelimb to study neural motor control by reproducing real animal movements in physics simulation. By training policies constrained by biomechanical costs that accurately predict muscle activity, and analyzing them using nonlinear dynamics methods from neuroscience, the work bridges AI and biology to reveal how movement emerges from the interplay of neural commands, muscle dynamics, and physical constraints – producing interpretable controllers that support AI safety through mechanistic understanding rather than black-box optimization.
Eva Dyer, Blake Richards & Herbie Bradley
A multimodal foundation model for human brain dynamics and alignment | Eva Dyer & Blake Richards | University of Pennsylvania & Mila
This work explores the development of a general-purpose, multiscale, multimodal model of human brain activity that learns shared representations across invasive (e.g., intracranial EEG) and non-invasive (e.g., scalp EEG) recordings. The goal is to build a foundation for simulating, decoding, and interacting with brain dynamics in ways that advance both neuroscience and the development of more interpretable, brain-aligned AI systems.
The great refactor | Herbie Bradley | University of Cambridge
This project researches AI’s capability for refactoring memory-unsafe code (in C & C++) into Rust in open-source repositories, at scale. The benefits of refactoring are to eliminate memory-safety vulnerabilities and strengthen the security and efficiency of the software supply chain.
Vision Weekend USA 2025: talks now live
Talks from our Vision Weekend in San Francisco last month are now available on our YouTube channel. Explore presentations from leading thinkers like Ed Boyden, Steve Jurvetson, Ariel Ekblaw, Haleh Fotowat, and Anders Sandberg!
Apply to our AI Nodes: funding, compute, and office space
This year we are launching two new hubs for AI-powered science and safety in San Francisco and Berlin. Our AI Nodes combine grant funding, compute, workspace, and programming to support teams advancing open, secure, AI-driven progress.
We invite applications for:
Grant funding for your project
Office and event spaces
Compute accessto support your research
We support work in the following areas:
AI for security
Private AI
Decentralized and cooperative AI
AI for science and epistemics
AI for neuro, brain-computer interfaces and whole brain emulation
Online seminar with Nobel laureate Ben Feringa | Feb 24
Ben Feringa won our 2025 Feynman Prize in the Experiment category. In a few weeks, we have the honor of welcoming him for an online award ceremony to celebrate his achievements and learn more about his pioneering work on the control of molecular motion!
This year marks our 40th anniversary and we are bringing Vision Weekend to the UK for the first time to celebrate! Join us in London as we gather at the forefront of neurotech, AI, longevity, and nanotech to explore which scientific and technological breakthroughs will shape the coming decades, and how to make them reality.
Confirmed keynotes include:
João Pedro de Magalhães, University of Birmingham, on hacking aging biology
Sergey Stavisky, UC Davis, on high-bandwidth BCIs for motor and communication recovery in people with paralysis
Anders Sandberg, Institute for Future Studies
Ronit Kanwar, Renaissance Philanthropy, on igniting a 21st century renaissance
Adam Shai, Simplex AI Safety, on the neural basis of intelligent behavior
Marius Hobbhahn, Apollo Research, on AI scheming
Jerzy Szablowski, Rice University, on precision neuromodulation through molecular engineering and focused ultrasound
What risk-reduction efforts would still make sense if transformative AI arrives within just a few years?
This is the question asked by the newly launched 2025 Peregrine Report – a compilation of insights from 48 in-depth interviews with researchers, policymakers, and practitioners working across the AI safety ecosystem. The report is co-authored by our Fellow Alumnus Gavin Leech and previous grantee Maximilian Schons.
The result is a portfolio of 208 initiatives, alongside insights on what the field most urgently needs: faster execution, better coordination, shared standards, and stronger operational capacity.
Authors: Maximilian Schons, Samuel Härgestam, Gavin Leech, and Raymund Bermejo, in collaboration with and supported by Halcyon Futures.
Last week, we had the great honor of celebrating the accomplishments of our 2025 Feynman Prize winner in the Theory category: Nicola Marzari. He received the prize for developing the theoretical and computational foundations that underpin modern atomically precise materials discovery. Learn more about his work in his presentation from the awards ceremony:
Jason Crawford, founder of the Roots of Progress Institute, recently joined the Existential Hope Podcast to discuss how technology expands human choice and control. He dives into the core arguments of his book, The Techno-Humanist Manifesto, and explores his philosophy of progress centered around human life and wellbeing.
In another episode, we hosted Elle Griffin, author of The Elysian – a newsletter imagining a better future through essays and fiction. Against a cultural backdrop dominated by dystopian narratives, Elle shares how utopian literature shaped her thinking and what it can teach us about designing societies that are not just functional, but genuinely desirable.
Tune into the Existential Hope Podcast on YouTube, or listen in your favorite podcast app.
Seminars
Watch the latest talks from our technical seminar groups. To join future conversations live, you can find all upcoming seminars here.
Community updates
Chronicle Biois hiringan AI Data Scientist. They area company building a multi-omic biobank and data platform for chronic, neuroimmune-mediated diseases, conditions like Long COVID, ME/CFS, POTS, and complex disorders that have been historically under-researched and poorly understood. In this role, you’ll get hands-on access to some of the most unique multimodal datasets in the world, build models that directly impact druggable targets and patient care, and collaborate with leading computational biologists, clinicians, and industry partners. Interested candidates can reach out directly with their CV to rohit@chroniclebio.com
New Grantees, 2026 Fellows & Norm Hardy Prize Winner
In this newsletter:
The winner of our 2025 Norm Hardy Prize
New work we have funded with our grants
Upcoming seminar with Nobel laureate Ben Feringa
Meet our 2026 cohort of Fellows
Apply for funding, compute, and office space at our new AI Nodes in San Francisco and Berlin
New report with 208 proposals for reducing AI risk
Plus: details on our upcoming Vision Weekend Puerto Rico, talks from Vision Weekend USA, and tickets to our first-ever Vision Weekend in the UK!
Vision Weekend Puerto Rico 2026
February 6-8 | Old San Juan | Last chance to get tickets
It’s only one week until we bring Vision Weekend to Puerto Rico! Join other futurist researchers and builders to explore the frontiers of AI, nanotech, brain–computer interfaces, and longevity biotech – while enjoying the sunshine, beaches, and warm nights of the Caribbean.
Beyond the 30+ keynotes, you can look forward to discussing your favorite topics in depth with the speakers, finding your next collaborators, and pitching your project for a chance at the $10,000 prize.
Speakers include:
Robin Hanson, George Mason University
Ramez Naam, author of the Nexus trilogy
Jean Hebert, ARPA-H
Lauren Wagner, Abundance Institute
Samuel Hammond, Foundation for American Innovation
Konrad Kording, Kording Lab
Brandon Green, BTC Inc
Eva Dyer, University of Pennsylvania
Alex Shieh, The Antifraud Company
Ben Woodington, Coherence Neuro
Jin Suntivich, Cornell University
Joel Shor, Move 37 Labs
Richard Sedlmayr, The Agency Fund
Get Tickets
Norm Hardy Prize winner 2025
Our Norm Hardy Prize is a $10,000 award recognizing advances in usable security, with the long term goal to encourage interaction designs that make it easy for people to use secure systems securely.
We are excited to announce Dr. Pardis Emami-Naeini as the winner of the 2025 prize!
Her research focuses on empowering consumers to make more secure choices through clear, accessible security information. By developing a layered cybersecurity label for smart home devices, she has helped shape national policy and industry standards, including the new U.S. Cyber Trust Mark.
Learn more about the winning work from her presentation at the award ceremony:
Read the Full Announcement
Meet our new Fellows
We are excited to introduce our 2026 cohort of Foresight Fellows!
With our Fellowship program, we want to support early-career talents advancing important science and tech. We connect them with senior scientists, invite them to our technical workshops, seminars, and Vision Weekends alongside leaders in their fields, and offer platforms for sharing their work.
This year’s group includes a neuroscientist studying how the brain computes at a molecular level, a bioengineer 3D-printing human tissue, a PhD student advancing computational protein design, and an engineer developing robot swarms to build large-scale structures in space – just to name a few.
You can learn more about the Fellows and their work on our website.
New grantees
We are excited to share new projects we are supporting with our grants!
Linking usermode proofs to the seL4 spec | Gernot Heiser & Rob Sison | UNSW Sydney
As the seL4 microkernel has been proved to be correctly implemented with respect to its specification and also enforce the core security properties of confidentiality, integrity and availability, it is the most solid foundation for building secure systems. However, reasoning about the correctness (and security) of programs running on top of seL4 (including the actual operating system) requires that the kernel’s specification of the system calls it provides actually does what these programs expect.
To date, no such proofs connecting any kernel’s specification with its user programs’ requirements have been completed for any system calls that wait for a response from another program before returning, which puts any reasoning about the correctness of systems that use them on shaky ground. This project will develop the formalism and infrastructure to carry out such proofs for seL4.
Reaching consensus via generative social choice | Ariel Procaccia | Harvard University
This project aims to develop, evaluate, and deploy a group decision-making process that combines social choice theory with large language models. The process will be capable of handling a vast space of alternatives (such as all possible policies) and will provide guarantees to promote safety and trustworthiness.
Interpretable neuroAI for motor control: massively parallel imitation learning of musculoskeletal dynamics | Talmo Pereira | Salk Institute for Biological Studies
This project uses imitation learning with a detailed musculoskeletal model of the mouse forelimb to study neural motor control by reproducing real animal movements in physics simulation. By training policies constrained by biomechanical costs that accurately predict muscle activity, and analyzing them using nonlinear dynamics methods from neuroscience, the work bridges AI and biology to reveal how movement emerges from the interplay of neural commands, muscle dynamics, and physical constraints – producing interpretable controllers that support AI safety through mechanistic understanding rather than black-box optimization.
A multimodal foundation model for human brain dynamics and alignment | Eva Dyer & Blake Richards | University of Pennsylvania & Mila
This work explores the development of a general-purpose, multiscale, multimodal model of human brain activity that learns shared representations across invasive (e.g., intracranial EEG) and non-invasive (e.g., scalp EEG) recordings. The goal is to build a foundation for simulating, decoding, and interacting with brain dynamics in ways that advance both neuroscience and the development of more interpretable, brain-aligned AI systems.
The great refactor | Herbie Bradley | University of Cambridge
This project researches AI’s capability for refactoring memory-unsafe code (in C & C++) into Rust in open-source repositories, at scale. The benefits of refactoring are to eliminate memory-safety vulnerabilities and strengthen the security and efficiency of the software supply chain.
Vision Weekend USA 2025: talks now live
Talks from our Vision Weekend in San Francisco last month are now available on our YouTube channel. Explore presentations from leading thinkers like Ed Boyden, Steve Jurvetson, Ariel Ekblaw, Haleh Fotowat, and Anders Sandberg!
Full Vision Weekend 2025 Playlist
Apply to our AI Nodes: funding, compute, and office space
This year we are launching two new hubs for AI-powered science and safety in San Francisco and Berlin. Our AI Nodes combine grant funding, compute, workspace, and programming to support teams advancing open, secure, AI-driven progress.
We invite applications for:
Grant funding for your project
Office and event spaces
Compute access to support your research
We support work in the following areas:
AI for security
Private AI
Decentralized and cooperative AI
AI for science and epistemics
AI for neuro, brain-computer interfaces and whole brain emulation
AI for longevity biotechnology
AI for molecular nanotechnology
Learn More & Apply
Upcoming events
Online seminar with Nobel laureate Ben Feringa | Feb 24
Ben Feringa won our 2025 Feynman Prize in the Experiment category. In a few weeks, we have the honor of welcoming him for an online award ceremony to celebrate his achievements and learn more about his pioneering work on the control of molecular motion!
Sign Up
Vision Weekend United Kingdom | June 5-7
This year marks our 40th anniversary and we are bringing Vision Weekend to the UK for the first time to celebrate! Join us in London as we gather at the forefront of neurotech, AI, longevity, and nanotech to explore which scientific and technological breakthroughs will shape the coming decades, and how to make them reality.
Confirmed keynotes include:
João Pedro de Magalhães, University of Birmingham, on hacking aging biology
Sergey Stavisky, UC Davis, on high-bandwidth BCIs for motor and communication recovery in people with paralysis
Anders Sandberg, Institute for Future Studies
Ronit Kanwar, Renaissance Philanthropy, on igniting a 21st century renaissance
Adam Shai, Simplex AI Safety, on the neural basis of intelligent behavior
Marius Hobbhahn, Apollo Research, on AI scheming
Jerzy Szablowski, Rice University, on precision neuromodulation through molecular engineering and focused ultrasound
Get Tickets
Find all our upcoming in-person and virtual events in Luma.
New report: 208 proposals for reducing AI risk
What risk-reduction efforts would still make sense if transformative AI arrives within just a few years?
This is the question asked by the newly launched 2025 Peregrine Report – a compilation of insights from 48 in-depth interviews with researchers, policymakers, and practitioners working across the AI safety ecosystem. The report is co-authored by our Fellow Alumnus Gavin Leech and previous grantee Maximilian Schons.
The result is a portfolio of 208 initiatives, alongside insights on what the field most urgently needs: faster execution, better coordination, shared standards, and stronger operational capacity.
Authors: Maximilian Schons, Samuel Härgestam, Gavin Leech, and Raymund Bermejo, in collaboration with and supported by Halcyon Futures.
Read the Report
Feynman Prize awards ceremony with Nicola Marzari
Last week, we had the great honor of celebrating the accomplishments of our 2025 Feynman Prize winner in the Theory category: Nicola Marzari. He received the prize for developing the theoretical and computational foundations that underpin modern atomically precise materials discovery. Learn more about his work in his presentation from the awards ceremony:
The Feynman Prizes recognize exceptional contributions to nanotechnology and have twice anticipated future Nobel laureates. Nominations for the 2026 prizes are now open.
The Existential Hope Podcast
Jason Crawford, founder of the Roots of Progress Institute, recently joined the Existential Hope Podcast to discuss how technology expands human choice and control. He dives into the core arguments of his book, The Techno-Humanist Manifesto, and explores his philosophy of progress centered around human life and wellbeing.
In another episode, we hosted Elle Griffin, author of The Elysian – a newsletter imagining a better future through essays and fiction. Against a cultural backdrop dominated by dystopian narratives, Elle shares how utopian literature shaped her thinking and what it can teach us about designing societies that are not just functional, but genuinely desirable.
Tune into the Existential Hope Podcast on YouTube, or listen in your favorite podcast app.
Seminars
Watch the latest talks from our technical seminar groups. To join future conversations live, you can find all upcoming seminars here.
Community updates
Chronicle Bio is hiring an AI Data Scientist. They are a company building a multi-omic biobank and data platform for chronic, neuroimmune-mediated diseases, conditions like Long COVID, ME/CFS, POTS, and complex disorders that have been historically under-researched and poorly understood. In this role, you’ll get hands-on access to some of the most unique multimodal datasets in the world, build models that directly impact druggable targets and patient care, and collaborate with leading computational biologists, clinicians, and industry partners. Interested candidates can reach out directly with their CV to rohit@chroniclebio.com