Subvolts take See Anthropic's Fable model in action, from safety to coding and beyond. Jun 13, 2026 - Developers
- Learn about Fable's real-world applications in coding and safety, and hear field reports on agents and interpretability.
- Business
- Understand how advanced AI models like Fable are being integrated into workflows and discuss potential impacts on legal reasoning.
This episode provides a solid roundup of recent AI developments and research discussions from a leading lab.
AI in the AM — Week 2 Highlights (June 2026)
Audio streams from Erik Torenberg, Nathan Labenz. Your place is remembered on this device.
From Subvolts
Summary and analysis by Subvolts editorial, AI-assisted.
This episode of The Cognitive Revolution features Nathan Labenz discussing Anthropic's Fable launch and its real-world applications. Topics include safety, autonomous coding, 3D world-building, and a Twitter experiment. Guests Geoffrey Irving and Daniel Murfet discuss alignment theory, while others share field reports on data agents, interpretability, and token economics. Mercury, a financial tool, is also featured.
This episode covers how Anthropic's new AI, Fable, is being used in real jobs, like making sure AI is safe and helping write computer code. It also explores new ideas for making AI smarter and more reliable.
This episode offers a practical look at Anthropic's Fable model in action, moving beyond theoretical discussions to real workflows. It's for AI researchers, developers, and business leaders interested in the latest advancements in model safety, autonomous systems, and specialized AI applications. The discussions on alignment theory and field reports connect directly to current industry efforts in building more reliable and capable AI systems, mirroring the broader AI race.
- HostNathan Labenz
- GuestGeoffrey Irving
- GuestDaniel Murfet
- GuestRahul Sonwalkar
- GuestShlok Khemani
- GuestTom McGrath
- GuestAndrew Moore
- GuestNate Jones
- GuestKarina Nguyen
- Episode Title: AI in the AM — Week 2 Highlights (June 2026)
- Show: The Cognitive Revolution
- Length: 105 minutes
- Featured Model: Claude Fable 5
- Key Discussion Point: Alignment theory and guarantees before recursive self-improvement
Week 2 highlights follows Anthropic’s Fable launch in real workflows, from safety gates and API refusals to autonomous coding, 3D world-building, and a Claude-run Twitter experiment. Geoffrey Irving and Daniel Murfet argue for alignment theory and guarantees before recursive self-improvement, while prinz tests Fable on legal reasoning and monitoring. Rahul Sonwalkar, Shlok Khemani, Tom McGrath, and Andrew Moore add field reports on data agents, hybrid authorship, interpretability, context systems, token economics, and power concentration.
Mercury: Run your finances with virtual cards, spending limits, merchant/category locks, and AI-friendly tools like API keys, MCP, and CLI. Check out Mercury at https://mercury.com
LINKS:
- Claude Fable 5 announcement
- Julius AI platform
- Rahul Sonwalkar homepage
- Nate Jones homepage
- Shlok Khemani homepage
- FrontierCode benchmark blog
- Lovelace AI company
- Andrew Moore Wikipedia profile
- Geoffrey Irving homepage
- Daniel Murfet LessWrong profile
- Sequent Research announcement
- Timaeus research organization
- Automated Alignment paper
- Goodfire AI company
- Tom McGrath homepage
- Predictive data debugging tool
- prinzbench legal benchmark
- Unit distance conjecture disproof
- Dario Amodei policy essay
- Vending-Bench 2 benchmark
- Andon Labs site
- Recursive Superintelligence startup
- Sakana AI company
- PostTrainBench ben
- recursive self-improvement
- AI researchers
- AI systems that can improve themselves over time
- alignment theory
- AI safety researchers
- Ensuring AI goals match human values
- token economics
- AI developers
- How the cost and use of AI processing units (tokens) affect AI development and deployment
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 40Designing How AI Grows — Tom McGrath
WithTom McGrath
1 h 47Don't Fight Backprop: Goodfire's Vision for Intentional Design, w/ Dan Balsam & Tom McGrath
WithDan Balsam, Tom McGrath
2 h 19
38 minSpeaking of intelligence
WithGeoffrey Irving, Chris Dyer, Angeliki Lazaridou, Lisa-Anne Hendriks
59 minJoelle Pineau on Why Sovereign AI Just Got Real
WithJoelle Pineau
Questions people ask
- What are the real-world applications of Anthropic's Fable model discussed?
- The episode covers Fable's use in safety gates, API refusals, autonomous coding, 3D world-building, and a Twitter experiment.
- What is the focus of the discussion on alignment theory?
- Geoffrey Irving and Daniel Murfet argue for alignment theory and guarantees before recursive self-improvement.
Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to Erik Torenberg, Nathan Labenz.