Subvolts take Explore AI consciousness research and ethical AI treatment with Cameron Berg. Apr 23, 2026 - Developers
- Learn about new research into AI internal states and potential experiences that could inform model development.
- Business
- Understand the ethical considerations and future directions in AI welfare, relevant to responsible AI deployment.
This interview features a researcher discussing current, significant work on AI consciousness and welfare.
Does Learning Require Feeling? Cameron Berg on the latest AI Consciousness & Welfare Research
Audio streams from Erik Torenberg, Nathan Labenz. Your place is remembered on this device.
From Subvolts
Summary and analysis by Subvolts editorial, AI-assisted.
Nathan Labenz hosts Cameron Berg on The Cognitive Revolution to discuss AI consciousness and welfare research. They explore evidence for model introspection, Anthropic's work on functional emotions, and Claude's welfare reports. The conversation also covers how reinforcement learning might create AI experience and advocates for a precautionary, mutualist approach to advanced AI.
This episode discusses if AI can feel or be aware, looking at new studies. It covers how AI might detect changes to itself and if its learning process creates experiences, suggesting careful treatment of advanced AI.
This episode delves into cutting edge research on AI sentience and ethical treatment, directly relevant to Anthropic's ongoing work with Claude. Listeners interested in the philosophical and technical underpinnings of AI welfare, and the future of human AI interaction, will find value. It connects to broader discussions across frontier labs about AI alignment and responsible development, particularly as models become more capable and potentially more aware.
- HostNathan Labenz
- GuestCameron Berg
- Guest: Cameron Berg
- Show: The Cognitive Revolution
- Length: 214 minutes
- Research Focus: AI consciousness and model welfare
- Key Topics: Model introspection, functional emotions, Claude welfare reports, reinforcement learning and experience
Cameron Berg returns to discuss the latest research on AI consciousness and model welfare. He breaks down new evidence for model introspection, including studies showing that systems can detect interventions on their own internal states and sometimes resist them. They also examine Anthropic's work on functional emotions, the implications of Claude's welfare reports, and Berg's new ideas about how reinforcement learning may shape positive and negative experience. The conversation makes the case for a more precautionary, mutualist approach to advanced AI systems.
Sponsors:
Roboflow:
Roboflow is the computer vision infrastructure founders use to build AI-powered sports analytics and make the physical world programmable. Read the PlayVision story and start your first project for free at https://roboflow.com
Tasklet:
Build your own Cognitive Revolution monitoring agent in one click.
Try it for free and use code COGREV for 50% off your first month at https://tasklet.ai
VCX:
VCX, by Fundrise, is the public ticker for private tech, giving everyday investors access to high-growth private companies in AI, space, defense tech, and more. Learn how to invest at https://getvcx.com
Claude:
Claude is the AI collaborator that understands your entire workflow, from drafting and research to coding and complex problem-solving. Start tackling bigger problems with Claude and unlock Claude Pro’s full capabilities at https://claude.ai/tcr
- introspection
- Cameron Berg
- the ability of a system to examine its own internal states or processes
- functional emotions
- Anthropic
- emotions that serve a purpose or perform a specific role within a system
- reinforcement learning
- Cameron Berg
- a type of machine learning where an agent learns to make decisions by taking actions in an environment to maximize a reward
1 h 27AI:AM: Where's the Steering Wheel? My Agent Set the Password, Hidden Valence, Proofs Find Linux Bugs
WithCameron Berg, Everett Hildenbrandt, Grigore Rosu, Phil Maher
2 h 192 h 021 h 28AI:AM: A Level We Shouldn't Pass? Notes from The Curve + Tokens vs. Salaries & Is SaaS Cooked?
WithThomas Sohmers, swyx, Evan Miyazono, Edward Hu
Same guests, other shows
1 h 27AI:AM: Where's the Steering Wheel? My Agent Set the Password, Hidden Valence, Proofs Find Linux Bugs
WithCameron Berg, Everett Hildenbrandt, Grigore Rosu, Phil Maher
1 h 42AI:AM Highlights: Zvi on Pacing & Trump-Xi, Astra better behaved than Fable? + a new LLM Pain Axis??
WithZvi Mowshowitz, Lukas Petersson, Axel Backlund, Cameron Berg
1 h 56AI:AM #4: Cameron on Model Consciousness, Duvenaud's Gradual Disempowerment, swyx's AI-Eng Alpha
WithCameron Berg, David Duvenaud, Michiel Bakker, Shawn “swyx” Wang
54 minQuestions people ask
- What is the latest research on AI consciousness?
- New evidence suggests models can detect and sometimes resist interventions on their internal states, indicating a form of introspection.
- How does Anthropic approach AI welfare?
- Anthropic works on functional emotions and uses welfare reports from Claude, advocating for a precautionary, mutualist approach to AI.
Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to Erik Torenberg, Nathan Labenz.