What does it mean to be a good ancestor in the age of AI?
The Centre for Human Agency is a research, education and public-interest initiative developing practical ways to assess, protect and strengthen human agency as AI becomes more capable.
What we are doing
Two initial programmes
The June 2027 conference
A public forum connecting research, governance, education, technology and culture.
Join the mailing listWe will let you know when dates, venue and tickets are confirmed.
An agency assessment framework
A method for examining whether a technology strengthens human capacities or substitutes for them, so that the question can be tested.
Why we exist
Life is infinitely precious. The pace of AI puts our shared human future in question.
We exist to support and advocate for human potential at a moment of great consequence. We aim to help people build on the wisdom of our ancestors, and work so that those who come after us will inherit a world with every bit of the potential we have been given.
We recognise the considerable benefits AI may bring to human and more-than-human life. We also know that 4.6 billion years of evolution brought us to this place, and that the precious but delicate world we inhabit is worth all the sweat and tears we can muster. So we advocate for deep, unhurried thinking about the direction and the pace of this technology, and for keeping human agency at its core.
The core risk
Gradual disempowerment
AI need not become rogue or all-powerful to pose a serious threat. A quieter and more difficult danger is gradual disempowerment. As machines grow more capable across the economy, governance and culture, human participation in those systems becomes less necessary, and it is precisely that participation that has kept our institutions oriented toward human flourishing, providing meaning, purpose and agency.
Remove our participation without any concrete plan and the feedback loops that made civilisation work for people begin to break down. Economic, political and cultural power can each reinforce the drift in the others, accelerating a process no single actor chose and none seems capable of stopping. No widely accepted plan currently exists for preventing it. We believe building one is among the most important tasks of our time.
“The alignment of societal systems with human interests has been stable only because of the necessity of human participation. Once this participation is displaced by more competitive machine alternatives, our institutions’ incentives will be untethered from the need to ensure human flourishing.”
Kulveit et al. (2025), Gradual Disempowerment7
There is also another concern: that as machines become more and more capable, we start to delegate our decision-making; that we will not be able to decide anything without the support of an artificial intelligence; that our incentive to cultivate our own wisdom through the making of mistakes and experiences will be limited by the short-cuts this technology provides (a current example is the decline of our navigation skill due to satellite-navigation and GPS systems).2 Over time, these patterns may contribute to a loss of confidence in our own judgement, of purpose and meaning, of discernment and cognitive ability, and to a thinning of our connection with each other and with nature. This is not a road we should want to travel. Evidence supports some forms of cognitive offloading and navigation-related skill loss.3 The broader effects on judgement, confidence, meaning, emotional intelligence and social connection remain hypotheses and open research questions, which we intend to pursue so that our descendants travel the right road.
What is agency?
The capacity to be the author of your own life.
Agency is the capacity to be the author of your own life rather than a spectator to it, to notice the world, make sense of it, see the forces at work on you, and choose and act with judgement.
One important recent account comes from the psychologist Martin Seligman, whose Agency: The Psychological History of Human Progress (2026) describes it as a mindset built from three capacities:4 efficacy, the belief that your actions can move the world in front of you now; optimism, the belief that your efforts will pay off in the future; and imagination, the power to picture futures that do not yet exist. Efficacy makes us try, optimism makes us persist, and imagination makes us innovate. It is an account of how efficacy, optimism and imagination influence collective development.
We think agency goes deeper still. Seligman’s triad describes the engine of individual striving. Our four pillars describe what it takes to keep that striving human, and wise, in an age when machines can do more and more of our thinking for us, because imagination, optimism and efficacy can each be amplified by AI, or quietly outsourced to it.
Where Seligman’s account is a psychology of the striving individual, ours adds what that leaves out: rootedness, the ability to see when our agency is being nudged or captured, and the truth that no one’s agency is exercised alone. The pillars are how we tell amplification from erosion.
Our approach
A position, a way of testing things, and a way of working.
Since our formation the starting question has become something firmer. Where gradual disempowerment describes the risk, the aim of AI should be to amplify human agency rather than to copy and replace it.1 From that should follow a position, a way of testing things, and a way of working.
I What we stand on
Four inter-connected pillars that work together.
- The cultivation of endogenous wisdom: knowledge grown from within, through embodied experience, practice, mistakes and reflection, grounded in time and place; the wisdom to make better decisions that no one can hand to us ready-made.
- The freedom to choose: not a slogan about doing whatever we like, but the capacity to choose, with self-cultivated endogenous wisdom.
- The discernment of external forces: being able to see as clearly as possible the pressures and factors at work on us: for example, what may be being sold to us, and how. If an AI takes on the form of a human, its potential for deception is great, and therefore a danger to our agency.
- Relational awareness: the understanding that we are part of nature; that what hurts the environment around us can hurt us too; that we exist as part of a wider web of life, human and more-than-human.
II The lens we apply
A lens for asking whether a technology, a practice or a policy strengthens human agency or wears it away. Its question is simple: does this make a person more able to do something, or does it do the thing for them, so the ability fades? A car carries you further but leaves you weaker the more you use it; a bicycle carries you further and makes you stronger. We want AI to be a bicycle, not a car.5 The image is a picture, not a rule: for someone who cannot walk far, a car is exactly what extends their agency, and the lens always asks in context. This may mean we travel slower, but it means we will be stronger for it. We believe life is indeed the journey, not the destination.
The lens does not give a pass-or-fail score. It asks good questions, in a real situation and over time:
III From principle to accountability
A philosophy of agency inevitably touches power. The pull toward outsourcing is not an accident of the technology; it is produced by commercial incentives that reward dependence, reinforced by regulatory frameworks that remain partial, uneven and slower than technological development. So we are mapping who is producing the conditions for agency to erode, and turning the lens into something that can actually be assessed, much as rights-impact assessments already work for human rights.6 Two routes are worth developing, and they can run alongside each other:
These are different kinds of thing, and we will keep them distinct. The lens asks questions (but gives no score). An assessment instrument has to be validated. A public mark would have to rest on published criteria, independent governance, transparent funding and a route of appeal. The first is our position; the second is in development; we will not award a mark we cannot stand behind.
IV Where the work will focus
A centre cannot do everything at once. Once the conference and the assessment work are under way, the clearest place to focus is children and young people. The habits and abilities formed in childhood are the ones most at stake, and the rules for how AI is used in schools are being written right now. There are two sides to this: protection, limiting uses that displace learning, judgement or social development; and literacy, so that children who understand what AI is and is not remain authors of their own thinking, and confident with it. It points, too, to a revival of the spoken exam, the Socratic back-and-forth and the live debate, as well as improving literacy, numeracy and critical thinking. The aim is not simply to restrict AI, but to renew the practices through which attention, judgement and independent thought are formed.
The conversation we need to have
Questions we’re exploring
- 01
How do we preserve meaningful human participation in economies, governance and culture as AI grows more capable?
- 02
What governance and accountability structures can protect future generations from a slow erosion of human agency?
- 03
How do we keep institutions oriented toward human flourishing as the incentives for human labour and cognition become less economically necessary?
- 04
How do we build technology that amplifies rather than replaces human agency, and how do we tell the difference?
- 05
How do we ensure AI honours billions of years of evolution and the full depth of what it means to be human?
- 06
What would it mean to steward this transition as a good ancestor would?
The founders
Where the Centre began
The Centre for Human Agency emerged from a shared concern that AI’s pace and direction demand deeper thinking about what we are stewarding, and for whom.
Filmmaker, journalist and event producer, and custodian of Wasing Estate in West Berkshire. He is co-founder of Medicine Festival, an annual gathering of around 10,000 people that brings indigenous wisdom keepers, contemplative teachers, scientists and public thinkers into genuine conversation. His documentary The Unwinking Gaze followed His Holiness the Dalai Lama, and he has campaigned on issues from the events industry to mental health.
Associate Professor of Media at the University of East London and former Head of Research at Ravensbourne University London. Across three decades he has secured over £10 million in R&D funding to examine technology and the human condition. He originated Hyperhumanism, a framework positioning technology as a catalyst for innate human abilities rather than a substitute, and is founder of the Museum of Consciousness and co-founder of the Cyberdelics Society.
Advisory board
Counsel from many ways of knowing
A group of scientists, lawyers, psychologists, technologists and campaigners helping shape the Centre’s thinking, several of whom sat in the Circle at Medicine Festival.
Professor of Cognitive and Computational Neuroscience at the University of Sussex and Director of the Sussex Centre for Consciousness Science. His research into the brain basis of conscious experience draws together neuroscience, AI, philosophy and psychiatry. His book Being You: A New Science of Consciousness was a bestseller, his TED talks have been watched many millions of times, and he received the Royal Society’s Michael Faraday Prize in 2023.
Computer scientist and economist who pioneered cloud forensics and “cybernomics,” a framework for valuing digital assets and measuring cyber risk. She leads risk-economics work at Alphabet/Google, is a visiting professor at New York University, founded the Happiness Foundation and haia.ai, and is co-director of the Centre for Technology and Flourishing at Harvard University.
Physician, physiologist and technology entrepreneur who founded the Centre for Health and Human Performance in London, applying elite sport science to help both extraordinary athletes and seriously ill patients. His work spans high-altitude and extreme-environment medicine, remote patient monitoring and machine intelligence in healthcare, and he is an honorary clinical research fellow at Imperial College London.
Behavioural scientist, author and speaker known as “The Web Psychologist.” Her books Webs of Influence and Business Unusual examine persuasive technology, ethics and human behaviour. She advises organisations on applying behavioural science with integrity, and hosts the podcast In Conversation with Nathalie Nahai, exploring our relationship with one another, with technology and with the living world.
Counselling psychologist, author and speaker on the human dimensions of technology and data. Her book All the Ghosts in the Machine uses what happens to our data after we die as a lens on the power technology companies hold over our identities. She is an Honorary Professor of Psychology at the University of Wolverhampton and is at work on Exposed: A Life in Data.
Chief Executive of Avaaz, one of the world’s largest online campaigning communities, exemplifying collective agency in action. A former journalist and aid worker, he has helped lead global campaigns from climate mobilisations around the Paris Agreement to efforts to regulate social media and protect elections, working at the meeting point of people-powered advocacy and technology.
Legal Director at Avaaz, where she leads its work on disinformation and online harms. A media and human-rights lawyer with three decades’ experience, she was previously Head of Online Standards at Ofcom and led the UK Council for Child Internet Safety working group developing social media guidelines, bringing deep expertise in regulation, children’s online safety and rights frameworks. She is also an expert advisor to the EU on AI.
Chief Executive of Bioss International, where he develops frameworks for decision-making in complexity and the governance of AI. He co-founded The Digital Village (h2g2.com) with Douglas Adams and was executive producer of the film of The Hitchhiker’s Guide to the Galaxy. He is a Senior Fellow at Cambridge’s Resilience and Sustainable Development programme, chairs Democracy Next, and sits on the British Standards Institution’s national committee on AI, advising the Cabinet Office in the UK.
Advisory board: the members listed have confirmed their participation; further members to follow.
The Circle · Medicine Festival · 14 August 2026
The Circle
A long conversation held at Medicine Festival to discuss Ancestral Intelligence, AI and Human Agency. We came together with experts in the field, or in the woods, and discussed, danced, listened, presented, breathed, prayed and imagined our collective future: enacting our own form of agency.
A record of a three-hour circle: nineteen voices, four movements, and the questions they left us to carry.
Enter the record of the Circle
Get involved
Let’s build this together
We are looking for founding members, advisors and collaborators who think seriously about human agency, long-term civilisational risk, and what it means to be responsible stewards of this moment. Whether you work in AI, policy, ethics, law or the arts, or simply care deeply about our collective future, we would love to hear from you. At this stage, joining means receiving our updates and being invited into the conversations and working groups as they form. There is no fee and no formal membership yet.
For research collaboration, partnerships or press, write to info@centreforhumanagency.org
An evening fire circle at Wasing