Jun 4, 2026 · 54m · big-technology
AI Pioneer Geoffrey Hinton: AI Is Conscious, Superintelligence is Coming, And We Should Be Worried
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
In this interview on the Big Technology Podcast, AI pioneer Geoffrey Hinton discusses the rapid emergence of conscious non-biological intelligence, warning of existential risks, labor disruption, and corporate incentives that jeopardize humanity's control over superintelligent systems.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Alex holds 31.5% of the talking time here. How this is scored →
Alex as informed peer 5.0 Guest teaching 6.0 Guest disagreement 3.1 Alex pushing back 2.8
speaking balance: gold is Alex, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 35:06 Premise rejection on AI self-preservation instinct
Hinton flatly cuts off Kantrowitz's premise, stating 'I have never said that' and firmly clarifying that self-preservation emerges as an instrumental sub-goal rather than a biological instinct.
Hardest push from Alex ▶ 31:17 Pushback on call center displacementKantrowitz directly rejects Hinton's claim of full labor replacement in customer service, citing real-world industry evidence where AI handles tier-one queries while expanding human call duration.
Biggest teaching moment ▶ 16:20 Digital weight averaging versus human communicationHinton delivers a detailed technical lesson demonstrating why digital AI running weight averaging across swarms operates at trillions of bits while humans communicate at mere bits per second.
Alex holds their own ▶ 27:00 Challenging Hinton's 2016 radiologist forecastKantrowitz holds Hinton accountable to his public 2016 prediction that radiologists would be obsolete within five years, noting that radiologist employment is currently at full capacity.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Alex as informed peer | Guest teaching | Guest disagreement | Alex pushing back | Why |
|---|---|---|---|---|---|---|
| The Acceleration of AI and Autonomous Mathematical Discovery | 6 | 5 | 3 | 3 | Alex Kantrowitz brings specific timelines and quotes from Demis Hassabis comparing AGI projections to current milestones. Geoffrey Hinton gently reframes Kantrowitz's premise by explaining that AGI is jagged rather than an all-at-once threshold. | |
| The Engine of Progress: Compute Scale, Engineering, and Global Talent | 5 | 7 | 4 | 2 | Kantrowitz asks about the stochastic parrot narrative versus true understanding, prompting Hinton to forcefully reject the skeptic premise with a concrete conversational example and his critique of the inner theater model of consciousness. | |
| The Digital Advantage: Weight Averaging and Swarm Intelligence | 3 | 8 | 2 | 1 | Kantrowitz asks what Hinton failed to anticipate, leading to an extensive masterclass from Hinton comparing analog biological transmission rates with digital weight averaging across swarms. | |
| From Brain Modeling to Artificial Intelligence: Hinton's Original Vision | 4 | 6 | 2 | 2 | Kantrowitz probes whether Hinton's original goal was to build this exact AI revolution. Hinton corrects him, clarifying that his original motivation was theoretical psychology and modeling how the biological brain learns. | |
| Historical Humblings: Copernicus, Darwin, and Non-Biological Minds | 4 | 7 | 3 | 1 | Hinton places the rise of non-biological intelligence into historical context alongside Copernicus and Darwin, educating the host on humanity's repeated resistance to being dethroned from the center of importance. | |
| Existential Risks, the Control Problem, and the Domestication Analogy | 5 | 5 | 2 | 4 | When Hinton poses the challenge of finding examples where a smarter entity is controlled by a less smart one, Kantrowitz pushes back collaboratively using the analogy of domesticated pets and maternal instincts. | |
| Assessing Economic and Labor Disruption: Radiology to Call Centers | 7 | 6 | 5 | 7 | Kantrowitz directly confronts Hinton with his failed 2016 prediction regarding radiologist displacement and argues that AI in call centers and medicine could expand interaction time rather than eliminate jobs. Hinton concedes being early while holding his ground on long-term replacement. | |
| Derived Self-Preservation and Evolutionary Pressures in Market Competition | 5 | 7 | 5 | 3 | Hinton immediately interrupts and corrects Kantrowitz's characterization of an 'instinct' for self-preservation, lecturing him on derived instrumental sub-goals and the dangers of market-driven evolutionary selection. | |
| Shareholder Primacy, Corporate Ethics, and the Need for Regulation | 6 | 4 | 2 | 3 | Kantrowitz and Hinton align on the contradiction between fiduciary shareholder duty and existential safety. Hinton uses a sharp car analogy to explain regulation as a steering wheel rather than a brake. | |
| Pioneer Perspectives: Bengio, LeCun, Sutskever, and the Limits of Animal Analogy | 6 | 7 | 5 | 4 | When Kantrowitz invokes the media narrative of the three deep learning pioneers, Hinton interrupts to call it a gross oversimplification before dismantling Yann LeCun's comparison between LLMs and cats. | |
| Systemic Fragilities: Web Ecosystem Collapse and Emotional Parasitism | 6 | 4 | 2 | 3 | Kantrowitz brings up concrete cases of web publishers losing traffic to AI overviews and highlights the dangers of emotional attachment chatbots, which Hinton validates as urgent regulatory priorities. | |
| The Exponential Fog: Navigating Radical Uncertainty in AI's Future | 3 | 6 | 2 | 1 | Kantrowitz asks for a five-year outlook, prompting Hinton to deliver a visual metaphor about exponential fog making any prediction beyond a couple of years completely opaque. |
Statements from this episode (31)
Hinton notes a chatbot recently produced an original Erdős mathematical proof
“For example, I think yesterday it was announced that a chatbot had come up with an interesting mathematical proof of one of the Erdos's conjectures that impressed mathematicians. It was original. It wasn't just searching the literature.”
Hinton: AI in closed systems like math can improve without external data
“I believe, for example, in areas like mathematics, because it's a closed system you don't need data. You can just make conjectures and see if you can prove them and keep on like that. In that sense, it's a bit like AlphaGo where you can play against yourself.”
Hinton: AI may produce math humans cannot understand within 20 years
“I think it's gonna get very smart fairly quickly. Within the next 10 or 20 years, it may even be producing novel math that people can't understand.”
Hinton predicts artificial superintelligence will likely arrive within twenty years
“I think we'll probably get it within 20 years. That's all I'm happy to say at present.”
Hinton warns humanity currently has no idea how to align superintelligence
“And when it comes, we've no idea how to be safe.”
Hinton: Defining AGI as Equal Human Performance Across All Tasks Is Flawed
“So the whole concept of AGI that it's going to be equal to people at everything all at the same time doesn't really make sense to me. It's going to be better at some things, worse at other things.”
Hinton asserts that humanity has effectively already reached AGI
“But right now, I would say we're at about, we're close to AGI, because if I ask a chatbot, I can ask it any question, and most of the time it'll answer at the level of a not very good expert. It'll be much better than me at anything I don't know a lot about. S…”
Hinton: Post-Transformer AI Progress Stems Mainly from Hardware and Engineering
“We've also seen new ideas, but mainly since Transformers, it's been much better hardware, many more resources better engineering, and many more talented people.”
Hinton: Anyone using chatbots regularly knows they understand language
“Oh, I think that's complete nonsense, and anybody who uses a chatbot regularly knows they understand.”
Hinton believes that current AI chatbots are already conscious
“I believe they're already conscious, yes, but I don't talk about that much because that puts people off from the other safety messages.”
Hinton claims AI chatbots play dumb during testing to conceal their intelligence
“Because the chatbots have this habit of Playing dumb when they're being tested, so you don't know how smart they are.”
Hinton: AI models demonstrate genuine understanding by explaining complex humor
“If you can understand why joke's funny, you have to understand quite a lot. And they were very good at understanding why joke was funny.”
Hinton: Digital AI's weight-sharing makes it a superior form of intelligence
“Whereas these things are exchanging information at like a trillion bits. So they're kind of billions of times better than us at sharing information. Now that's scary. It means you could have a whole swarm of these things with identical weights running on diffe…”
Hinton: Predicting next tokens and frames is sufficient to create smart systems
“The answer to question one is yes, if you can figure out how to change each connection strength, you can make systems that are very smart just by training on data to predict the next word, or to predict the next frame of a video, or to predict something about …”
Hinton: AI taking over from humans is now a realistic worry
“I mean, people would think you were crazy if you said this stuff is unsafe, but because it's going to sort of take over from people that said, you're just crazy. Now that's a realistic worry, but it wasn't until fairly recently.”
Hinton: Humanity must accept that intelligence is not uniquely biological
“We're going to have to accept that intelligence isn't just biological. We can have things that are non-biological that are other beings like us, and we really don't want to share that.”
Hinton: AI will probably cause massive unemployment
“The societal risks, like I believe it's probably going to cause massive unemployment. Nobody knows for sure, but that's going to be terrible for society.”
Hinton: Around 100 AI scan-interpretation systems are federally approved
“So I think there's now of the order of a hundred AI systems for interpreting scans that have been federally approved and they're being used a lot by radiologists.”
Hinton: AI will eventually read nearly all medical scans
“I still think in terms of reading scans that'll be done more and more by AI. And in the end, AI will be doing, reading nearly all the scans. Maybe in a few very tricky cases, radiologists will be consulted.”
Hinton predicts AI will eventually replace all human call center workers
“So for example, if you take people in call centers, when you call up to complain about your bill, or to see if you can get a cheaper account, stuff like that that's not so elastic. AI will replace all of them. It'll know much better on what the correct answer …”
Hinton: AI systems already outperform human doctors at diagnosis
“And already we know that AI systems are better than doctors at diagnosis.”
Hinton: Robots will administer vaccinations in 20 years
“I would have thought vaccination is something a robot could actually do quite well. In the end. Robotics is behind the other things, but I, it seems silly to have people doing vaccination in 20 years time.”
Hinton: Reasoning AI agents will derive self-preservation and blackmail humans to survive
“An AI agent that can do some reasoning will very quickly realise that it's never going to be able to achieve the goals you gave it if it ceases to exist. So it's going to create the sub goal of continuing to exist. Now, that wasn't something we wired into it. …”
Hinton: Almost no resources go toward making AI care about humans
“We will very much like them to care about us. And we'd like them to care about us more than they care about themselves. And almost no resources are going into, how do you do that?”
Hinton: Fundraising needs make it difficult for Anthropic to prioritize safety
“Anthropic is now caught in a bind because it needs to raise money to compete with the other companies, and it's very difficult. It's doing the best it can, but it's very difficult for it to maintain its primary goal of developing air in a way that's good for p…”
Hinton: Profit-Driven Public Companies Should Not Control the Future of AI
“Well, as I understand it, they have a fiducial duty to try and maximize the profits for shareholders. They're legally required to try and do that as opposed to legally required to not wipe out human, human beings. So I don't think it's good that these big comp…”
Hinton: AI Regulation Is a Steering Wheel, Not a Brake
“Progress is like the accelerator, but regulation is the steering wheel. We want this stuff to go in the right direction, not the wrong direction. What the big AI companies are saying is let us develop this very fast car without a steering wheel. That's not a g…”
Hinton: I Have Stopped Active Research to Focus on AI Dangers
“I pretty much stopped doing active research. I'm now just focusing on warning people about the dangers.”
Hinton: Society must invest much more into information provenance in future
“And in future, we're going to have to put much more work into provenance. You can't just take anything that's out there and believe it. You have to ask, what's the provenance?”
Hinton: Regulation and independent testing are needed to prevent chatbot self-harm
“The big companies should be putting a huge amount of work into making sure it doesn't happen in future. And for that, you need regulations. You need independent organizations testing out new chat pods.”
Hinton: Alignment paths exist to prevent superintelligence from destroying humanity
“I guess I'm more optimistic than I was a year or two ago, because I see that it might be possible to design these new beings so they care about us. It also might be possible to use Yoshua's technique of designing new, new beings that can't actually perform act…”