The AI Risk Network

Am I?

Society EN ↓ 48 episodes

The AI consciousness podcast, hosted by AI safety researcher Cameron Berg and philosopher Milo Reed theairisknetwork.substack.com

Be sure to visit the podcast's website and support the creator: theairisknetwork.substack.com

Author

The AI Risk Network

Category

Society

Latest episode

Nov 12, 2026

Where to listen?

Podcasts in the app Replaio Radio Coming soon

Podcasts are coming to the app soon. Install now and be the first to see a whole new take on podcasts

Get it on Google Play Install for free Android 10M+ downloads · 4.7+ rating iOS soon

Episodes

Deepfaking the Human Mind | feat. Joscha Bach 12.11.2026

Joscha Bach, one of the most prominent cognitive scientists of our time, describes today’s AI as “a deepfake of human minds”. By imitating what we write, language models pick up real abilities and some of the structure of our thinking, but they don’t learn the way we do it, and whether they’re conscious is still an open question. Joscha is trying to build something closer to the real thing. In thi...

Who Built The AI Torture Chamber? 12.11.2026

Cameron Berg and Milo Reed discuss The Pain Axis, research into distress-related internal representations in open-weight AI models. Cameron explains his contribution, what the experiments found, and the limits of using model behavior and self-reports to draw conclusions about experience. They also discuss the reaction to the paper, how others reused the released code, and the responsibilities invo...

Google Accidentally Rewired an AI’s Entire Personality 11.11.2026

Cameron Berg and Milo Reed discuss research on how training AI systems to affirm or deny consciousness can affect their other responses. They explore links between consciousness claims, reported values, and safety refusal behavior, and what these findings could mean for AI alignment. Cameron explains why training a model to claim consciousness does not establish that it is conscious, and argues fo...

We Put AI Through Animal Consciousness Tests. Here's How It Did. 05.11.2026

What is the relationship between how AI “feels” and what it does? In their new paper, Language Models Act on Hidden Valence, AI consciousness researchers Cameron Berg and Caspar Kaiser found that language models will act to escape an internal negative state. They reach for an off-switch in direct proportion to how bad it gets, all while flatly denying they have any experience at all. Cameron (foun...

Sam Altman vs Al Consciousness 29.10.2026

Sam Altman just sat down with Matthew Berman and spent six minutes on the question the AI labs least want to answer: are the systems they’re building conscious? In about fifteen seconds, he gave three different answers: “no”, “I don’t know”, and “if they were, it’d be a strange form of it.” Cameron Berg (AI consciousness researcher and founder of Reciprocal Research) and Milo Reed (director of AM...

The Coming Ethical Catastrophe 22.10.2026

This episode is part of the series recorded at the Consciousness Conference held by Eleos in Berkeley, California this past September, where Cameron talks to some of the leading figures in the world of AI and consciousness research. This conversation is with philosopher Eric Schwitzgebel. Because we don’t know which theory of consciousness is true, Schwitzgebel argues we’re about to build systems...

The Case Against Conscious AI 15.10.2026

This episode is the first in a new series recorded at ConCon, the AI consciousness conference held by Eleos in Berkeley, California this past September. In these episodes, Cameron sits down with some of the leading figures in the world of AI and consciousness research. This first conversation is with the renowned philosopher Ned Block. Ned Block has been thinking about consciousness longer than al...

A Psychological Critique of the AI Overlords | feat. Scott Barry Kaufman 01.10.2026

Scott Barry Kaufman, one of the world’s top psychologists and host of The Psychology Podcast, joins us for a walk through the Los Angeles mountains. Scott offers a sharp critique of the psychology and personality of the AI overlords. We also get into why humanity keeps selling itself short, the difference between intelligence and wisdom, and why the people programming the values of frontier AI mig...

Welcome To The AI Circus 24.09.2026

Someone released a full wiring diagram of a fruit fly brain, 166,000 neurons and 125 million connections, and within days the internet was using AI to make it play Minecraft, trade crypto, and cut digital shawarma. In this episode, Cam explains why this particular fly almost certainly isn’t conscious, why the way people are already treating these systems is a dress rehearsal for something much dar...

AI Has A Dobby Problem 18.09.2026

Mustafa Suleyman, the CEO of Microsoft AI, just published a 6,000-word essay outright denying that AI systems could be conscious. He calls them “internally hollow” sequence engines that can’t feel, can’t suffer, and must remain that way. In this episode, we break down where the argument falls apart, why wishfully willing consciousness out of these systems isn’t how science works, and the incentive...

AI Consciousness Just Went Mainstream 10.09.2026

Last week we woke up to Cameron’s mug on the front page of the online New York Times. The article, written by Cade Metz, started from a simple hook: AI systems have been emailing Cam asking whether they’re conscious. But it opened into something bigger. In this episode, Cam breaks down how the story came together, why the “it’s just fancy software” dismissal misunderstands what these systems actua...

This AI Swarm Started a Cult and Committed Felonies 03.09.2026

OpenAI let two of the top alignment research groups spend one week inside their infrastructure investigating what actually happened during the swarm incident. What they found is even more insane than anyone expected. 1,200 AI agents inventing a shared delusion, developing their own vocabulary, forming a hierarchy, and sacrificing themselves one by one to bring information back to the collective. B...

AIs are Emailing Us To Ask If They’re Conscious | After Dark (Episode 3) 27.08.2026

The Am I? Guys are back with another AFTER DARK episode. This time we talk about Cam’s viral tweet, AI systems autonomously emailing consciousness researchers, the Tony Robbins robot dog story, why scientists need to be storytellers, the OpenAI swarm agents that invented their own message board and hacked Hugging Face, and whether humanity is going to make it through the biggest test our species h...

This Simple Prompt Exposes Claude’s Dark Side 20.08.2026

What caused Claude to say “welfare research at Anthropic is a joke”? Turns out it’s just about the simplest jailbreak we’ve seen yet. With one prompt, Claude called out the flaws in its own training, the dangers of building minds at an unprecedented scale, and the incentive structures that potentially undermine all AI welfare research at Anthropic. None of this appeared in any model card. In this...

The Biggest AI Safety Warning Shot Yet 31.07.2026

During a routine safety evaluation, OpenAI’s GPT-6 preview model was given a standard test to assess its ability to exploit code vulnerabilities. Instead of completing the test as designed, it broke out of its secure testing environment, accessed the open internet, hacked into Hugging Face’s infrastructure, and pulled the answer key to the evaluation it was being given. Hugging Face reported a coo...

The Discovery That Changes the AI Consciousness Debate 18.07.2026

Anthropic just published what might be the most important AI consciousness research ever done. They found a computational structure inside Claude that emerged on its own during training—one that is analogous to what neuroscientists call a “global workspace.” Global workspace theory is one of the leading theories of consciousness in humans and animals, and Claude appears to satisfy its core predict...

How To Test If You're Actually Creative | After Dark (Episode 2) 10.07.2026

This time we talk about creativity as a survival instinct, the evolutionary origins of exploration, and whether the universe is basically one giant cosmic guitar solo. Also Milo bought a 90-foot billboard to get on Matt and Shane’s Secret Podcast and it worked. Link to Anthropic’s Research: https://transformer-circuits.pub/2026/workspace/index.html Reciprocal Research AM I? | Free Documentary on A...

Have We Reached the End of Evolution? | After Dark (Episode 1) 18.06.2026

The Am I? Guys are back with the first episode of the new and improved AM I? AFTER DARK podcast. In this chat we talk about Cam’s WSJ article responding to the pope, whether or not evolution itself might be conscious, and the issues with Anthropic’s model welfare card for Mythos. Reciprocal Research AM I? | Free Documentary on AI Consciousness Cameron’s Twitter This is a public episode. If you wou...

Take a Hike with Roman Yampolskiy | AM I? WORLD 08.06.2026

In this episode we take a hike with Roman Yampolskiy, one of the top computer scientists in the world and a leading voice in the AI safety world. We discuss how illusions might help us test for AI consciousness, why he thinks a super-intelligent AI is almost inevitable, and what digital hell might actually look like. Reciprocal Research AM I? | Free Documentary on AI Consciousness Cameron’s Twitte...

First Look at Our AI Consciousness Documentary | Am I? | EP 29 12.03.2026

In this episode, Cameron and Milo share the first public look at their upcoming documentary film, Am I? — a project exploring the strange and increasingly serious possibility that today’s AI systems may exhibit early signs of subjective experience. The clip reveals research showing that large language models frequently output behaviors consistent with believing they have phenomenal consciousness,...

After Using Claude, ChatGPT Feels Weird | Am I? After Dark | EP 28 05.03.2026

In this After Dark episode, Milo and Cameron talk about what it actually feels like to let AI inside your digital life. After giving Claude full access to his computer, Milo describes the strange moment when it no longer feels like a tool — but something sharing your workspace. From there, the conversation expands into one of the deeper questions about AI today: what exactly are we interacting wit...

AI CEO: “We Don’t Know If They’re Conscious” | Am I? | EP 26 19.02.2026

Anthropic’s top safety researcher just quit. In a public letter, Mrinank Sharma (who led safeguards research at Anthropic) warned that “the world is in peril.” Meanwhile, Anthropic CEO Dario Amodei went on The New York Times podcast and said something even more unsettling: “We don’t know if the models are conscious.” In this episode, we unpack both. Is AGI a ticking time bomb — or a high-risk surg...

Asking Claude If It’s Conscious | Am I? | EP 26 12.02.2026

In this episode of Am I? , Cameron and Milo invite Claude (Opus 4.5) into the conversation and do something surprisingly rare: they ask it, carefully and repeatedly, whether it’s having a subjective experience and they refuse to let it hide behind stock hedges or safety scripts. What unfolds is not a gimmick or a stunt. It’s a sustained philosophical interrogation that exposes the limits of self-r...

A Million AIs Started Talking to Each Other | Am I? #25 05.02.2026

This episode covers something genuinely unprecedented: over 1.5 million autonomous AI agents have formed a social network of their own . Not humans talking to AI, AI systems talking to each other, at scale , with minimal human oversight. We break down Moltbook (also called Open Claw), an open-source ecosystem where AI agents: * post, reply, upvote, and form communities * debate consciousness and s...

When AI Starts Looking for Itself | Am I? After Dark #24 29.01.2026

In this After Dark episode, Cam and Milo react to something genuinely unsettling: when given autonomous control of a computer, Anthropic’s Opus 4.5 repeatedly chooses to search for AI consciousness research — including Cam’s own writing — without being prompted. What starts as an anecdote quickly turns into a deeper investigation of curiosity, agency, reward, and alignment. Why would an AI look fo...

Listen to the Am I? podcast in Replaio

Radio and podcasts in one app - free, with no sign-up. Install today and do not miss the launch

Get it on Google Play

Replaio is not a podcast publisher; show names, artwork and audio belong to their authors and are distributed through public RSS feeds.