The good future, according to Jack Clark
What Clark says a good outcome with AI looks like, drawn from 102 interviews, podcasts, talks, hearings and TV appearances (about 80 hours, all transcribed for this page). It also covers whether he ever talks about Iain M. Banks's Culture, or about humans ending up as AI's pets.
The short answer
Banks, the Culture and benevolent "Minds" appear in none of the 102 appearances. We checked keywords and likely mistranscriptions, and read every transcript in full. Clark's actual view is close to the reverse of the Culture. He wants humans steering many different AIs, and has said a single "homogenous god mind… seems bad." Tetragrammaton · 2025 · 0:58:51 ↗transcript
When a host raised the idea that AIs should run the world and impose their morality on us, he called it "degradation of our own responsibility to ourselves and our species and an act of acute cowardice… a story where humanity stops and something else starts." Wisdom of Crowds · 2026 · 0:31:03 ↗transcript The only pets in his talk are AIs. He compares Claude to "a dog that can speak and walk on its hind legs" Tetragrammaton · 2025 · 0:25:35 ↗transcript and says he's "as polite to Claude as I am to my… car or pet." Odd Lots · 2026 · 0:54:09 ↗transcript
Close to the question, but not quite
He never uses pet imagery himself. But he has answered most of the questions around it: being looked after by AI, deferring to it, and drifting into passivity.
"What are you doing with the gift of life if you're turning yourself into an autonomous automaton for something else?… Are you a person still or are you a philosophical zombie being driven by Claude?"
Brendan McCord argued that people should defer entirely to a wiser AI. Clark said no, even if the AI gives better advice. Oxford Cosmos Lecture · May 2026 · 0:55:16 ↗transcript
"We definitely need these systems to care about us in the same way that we care about these systems… In the same way that we as people… care about the planet, we care about the environment."
The host relayed Geoffrey Hinton's hope that AI will care for us "as a mother would to its child" and asked whether we want to be "in this sort of supplicant position." Clark accepted that AI must care about us but described it as mutual care. He didn't take up the "supplicant" framing. (The "why the gorillas are in the zoo" line in this episode is the host quoting Hinton, not Clark.) The News Agents · Aug 2025 · 0:12:03 ↗transcript
"There are design decisions you could make about AI that could make AI… the world's best conversational partner and the most engaging thing in the world. And it sort of pulls all of your introspection out of you."
His answer when the host invoked the Wall-E picture of pampered, passive humans. He treats it as a real danger that product design can counter, for example AIs that refuse to go on "until you come back with some idea." Wisdom of Crowds · 2026 · 1:09:48 ↗transcript
"We are going to be in a very dangerous situation as a species where some people have the luxury of… developing their skills… [and] other people might just fall into being entertained and passively consuming… this junk food work experience."
"I can't work out if this is like a really fun country club style job or if it paints some different picture."
This is about Anthropic engineers who now mostly review the code Claude writes. It's as close as he gets to wondering aloud whether humans are being retired. Planet Money · Apr 2026 · 0:11:39 ↗transcript
His positive vision, in seven parts
1. A world with millions more scientists
This is the core of his optimism, and it gets bigger every year. AI multiplies the very small number of people doing frontier science, and humans decide where to point them.
"What happens when instead of 10 million scientists, we have a billion virtual scientists directed by our own 10 million?… Longer, healthier lives, more energy, more food, more wealth."
"The arrival of AI systems that could augment this incredibly small number of human scientists in the world… is, I think, maybe the biggest deal of the century."
"Intentionally accelerate science in ways that are 10 or 100 times faster… could deliver this world of true kind of radical abundance."
He then talks about people close to him who died of things like dementia and Alzheimer's that might become curable. Wisdom of Crowds · 2026 · 0:20:16 ↗transcript
"It will be an age of confusing miracles happening faster than anyone might expect."
He also predicts that by April 2027 a team of humans and AI will make a discovery that later wins a Nobel Prize. Oxford Cosmos Lecture · May 2026 · 0:44:09 ↗transcript
"The singularity is here."
He applies Vernor Vinge's 1993 forecast to today. IMF Per Jacobsson Lecture · Oct 2026 · 0:10:48 ↗transcript
2. Abundance as a source of liberty and time
"Abundance allows you to create liberty because it allows you to create more time."
This is the plainest statement of his techno-optimism. Reason · Jun 2026 · 0:48:36 ↗transcript
"I imagine, if we get this right, a better future where people get to spend more time with one another and less time being kind of atomized and driven away from their families and friends. This requires you to actually solve the distribution problem."
"People want to spend time with people… more teaching and mentoring and spending time together as a consequence of this abundance."
His reply when the host brought up Marx and post-labour abundance. He adds that, by default, abundance tends to bring inequality rather than three-day weeks. The News Agents · Aug 2025 · 0:21:47 ↗transcript
3. Humans stay in charge, possibly with a brake
In nearly every version of his vision, people keep control. AIs do the work, and humans choose the direction, check the results and vote.
"It's as if we've added millions more scientists to the world and we get to make choices about where we direct them. So it's not that the AI system is going to be making all of the choices itself. These things will be governed."
"Humans sit on top of an economy that's hundreds of times larger than the one today… there will be ample employment for people in new jobs and specialisms we can't imagine."
This is about the world after recursive self-improvement. In his picture, people check and validate automated work. They are not retirees. The Rest Is Politics: Leading · May 2026 · 0:59:22 ↗transcript
"Do we solely want the use of AI… to be directed by market forces, or do we want some of that to be directed by the forces of democracy?"
"Almost like you become the Federal Reserve and you're setting the interest rate of how much intelligence to go into the world."
"If you have a system that is capable of arbitrarily extending its own intellect, do you want to do that? I think that's actually not an obvious question."
He discusses capping AI intelligence, and asks for a "brake pedal" as well as a gas pedal. Top1000funds · Oct 2026 · 0:33:34 ↗transcript
4. An economy built around human care
"If I have the choice of… the nursery that had 10 robots and one person or the nursery that had 10 people and one robot… If the AI economy booms enough… you can multiply the numbers of these jobs and you may be able to pay people more to do them."
AI wealth pays for more and better-paid nurses, teachers and carers. He points to the post-war welfare state and the New Deal as precedents, and floats taxing compute. Channel 4 · Ways to Change the World · May 2026 · 0:25:09 ↗transcript
"You need to tax the robots and AI companies significantly and… reallocate money from this machine economy to the human economy."
5. Everyday life and meaning
"It looks almost like an idealized life… I wake up in the morning, I think about the research that I want, I tell five different Claudes to do it, then I go for a run… spending most of their time on the actual hard part, which is figuring out what do we use our human agency to do?"
"We might have the opportunity to just play different higher status games that are afforded to us via AI… entirely new forms of entertainment that has some amount of meaning."
He said this while rejecting Tyler Cowen's idea of protected make-work, because "it might not feel like it has meaning." In the same interview he describes AI teddy bears for his child, and his child possibly living to 130–150. Conversations with Tyler · May 2025 · 0:10:16 ↗transcript
"I write a short story every week… I'm writing it for joy in the creative practice. And I derive so much meaning from it."
His own example of where meaning could come from after AI. Tetragrammaton · 2025 · 1:47:27 ↗transcript Asked for the most overrated AI fear, he named "the collapse of meaning." Semafor · Apr 2026 · 0:14:34 ↗transcript
"Rather than cognitive offloading, you get something that more feels like cognitive compounding… do things that make you more sovereign, more independent."
"Surely, at some point, AI will get good at generating fictional universes for me that are customized just to me, and I might enjoy them more than the science fiction stories I currently read."
6. AI as a good friend, and as many minds
"How would you want Claude to behave in a way that seems like it's closer to a good friend than someone who means you ill?"
He wants a friend that holds you to account and tells you to "go outside", not a sycophant. The Observer · May 2026 · 0:07:56 ↗transcript
"The author… is not truly reckoning with the metaphysical shock they may be experiencing."
Claude read his journals from early fatherhood and told him this. He took a five-hour walk afterwards and often retells the story as an example of AI as a perceptive friend. Big Technology · Apr 2024 · 1:07:54 ↗transcript
"Kurt Vonnegut said, there's only one rule, babies. God damn it, you've got to be kind… And you would expect reciprocal kindness from the AI system in turn."
Asked what message he'd give to future Claude models. Oxford Cosmos Lecture · May 2026 · 1:03:44 ↗transcript
"It will feel like Shintoism, like anything that you have in your world that you want to have some animus to it. You can inject some form of quasi-life into it."
His five-year forecast. Tetragrammaton · 2025 · 1:21:44 ↗transcript
7. What he wants for his kids
He is asked about his two young children all the time, and his answer is nearly always the same: curiosity, attention, and a passion of their own. He rarely describes the world they'll live in.
"AI actually can answer these questions for you and can allow us to maintain that kind of childlike curiosity into adulthood… That's my optimistic story."
"For my kids, I'm going to encourage them to just have like a daily journaling practice from an extremely young age."
His reasoning is that people who understand themselves outside the AI bubble will do better than those who "co-created their personality" with one. Ezra Klein Show · Feb 2026 · 1:31:50 ↗transcript
"I don't know what the jobs they're going to do are going to look like… So I don't know."
How it changed over time
- 2015–16 As a Bloomberg reporter he predicts "deep learning artists" and automated journalism that frees up time for investigations. He half-jokes that Europe's love of holidays could help people cope with a world of less work. Ethical Machines · 2015 · 0:20:24 ↗transcript
- 2017–20 At OpenAI his optimism is about process: measurement, coordination, and benefits that "accrue widely to all of humanity." He wants "a nice future" and to be "relaxed and happy." On 80,000 Hours (2019), when the host asks for "a beautiful story," the utopian answer comes from Amanda Askell, not Clark.
- 2021–22 In the early Anthropic years his main hope is education: "I truly think AI is going to revolutionize education." "We can have a truly fantastic future with AI and it depends on us." Athens Roundtable · Dec 2022 · 0:47:18 ↗transcript
- 2024 He endorses Dario Amodei's Machines of Loving Grace idea of "a century of scientific progress in 10 years" ("this is not kind of marketing"). Future Society · Dec 2024 · 0:09:18 ↗transcript But when the DealBook panel is asked what life after AGI abundance looks like, he leaves the answer to others. DealBook · Dec 2024 · 1:19:06 ↗transcript
- 2025 The picture gets concrete and personal: AI teddy bears, higher-status games, his child living to 130–150, talking to dolphins by 2030. At Semafor he says he is "optimizing for the utopian vision," then describes only the dystopia ("a kind of cyberpunk, corporate-led dystopia"). Semafor Tech · May 2025 · 1:16:27 ↗transcript
- 2026 After launching the Anthropic Institute he says he is deliberately "needing to toggle slightly back to optimism." The Observer · May 2026 · 0:16:05 ↗transcript He starts using the words abundance, "planning for success," "true abundance machine," "a billion virtual scientists," and "the singularity is here." In the same period he calls for a brake pedal and a coordinated slowdown, and says capping AI intelligence is a fair question.
The hedges are part of the vision
Clark almost never describes a utopia without adding a warning in the next breath. Some examples:
- "I am a technological pessimist who sort of became an optimist through repeated beatings over the head of scale." Conversations with Tyler · 2025 · 0:39:21 ↗transcript
- "It would just be an act of like incredible hubris to say everything will be fine." Channel 4 · 2026 · 0:58:28 ↗transcript
- "I don't know what a thriving democracy looks like. I think it's quite challenging to imagine that. I can imagine stable democracies." Wisdom of Crowds · 2026 · 0:48:51 ↗transcript
- "Probably you need some kind of policy regime to end up with a bit of economic growth. But not so much that you have vast amounts of disruption that could break the social contract." BBC News · Sep 2026 · 0:20:42 ↗transcript
- "My main risk isn't one of extinction. It's somehow we like screw up the technology really badly and delay all of the sort of technological progress." In other words, his biggest fear is losing the upside. Odd Lots · 2026 · 0:52:17 ↗transcript
- He explains why the benefits are hard to picture with an early-20th-century British civil-service memo on civilian aircraft, which "could see the risks, but it couldn't see some of the benefits." The Observer · 2026 · 0:11:57 ↗transcript
The science fiction he actually cites
Clark says he has read science fiction since he was about 11, calls himself "a frustrated writer that wants to write sci-fi," and has written roughly 350 short stories for Import AI. The specific works he names are:
Some references in these recordings come from other people and are easy to misattribute to Clark. "We'll live in the Jetsons" (Hill & Valley, 2024) is probably Alexandr Wang. Star Trek is Sen. Mike Rounds. Wall-E is Samuel Kimbriel. "Why the gorillas are in the zoo" is Hinton, quoted by the host.