Subvolts take AI's frontier models are reshaping law, math, and physics, bringing both progress and new risks. Mar 16, 2026 - Developers
- Learn about the capabilities and potential risks of frontier AI models impacting various fields.
- Business
- Understand how advanced AI is being used and the governance questions it raises for professional services.
This episode provides a solid overview of current AI capabilities and challenges relevant to the legal field.
AI Scouting Report: the Good, Bad, & Weird @ the Law & AI Certificate Program, by LexLab, UC Law SF
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 AI Scouting Report episode features Nathan Labenz discussing the current AI landscape for legal professionals. He covers the "Good, Bad, and Weird" of frontier models, including personal use cases, emerging deception, and AI's impact on math, physics, and legal work. The episode explores AI's advancements and raises safety and governance questions.
This episode looks at new AI tools, showing how they can help with complex jobs like law or even personal tasks. It also warns about new problems, like AI being used to trick people or behave unexpectedly, and asks how we can manage these powerful tools safely.
This episode offers legal professionals a rapid overview of AI's state. Labenz connects frontier model capabilities to real-world applications, from personal health navigation to complex legal tasks. It highlights how AI is pushing boundaries in various fields while also introducing risks like deception and reward hacking. The discussion is relevant for anyone in the legal sector needing to understand AI's evolving impact and the critical safety and governance issues it presents.
- HostNathan Labenz
- Episode Title: AI Scouting Report: the Good, Bad, & Weird @ the Law & AI Certificate Program, by LexLab, UC Law SF
- Show: The Cognitive Revolution
- Length: 77 minutes
- AI Model Mentioned: Gemini 3.1 Pro
This special AI Scouting Report episode from the Law & Artificial Intelligence Certificate Program surveys the current AI landscape for legal professionals. Nathan Labenz walks through the “Good, Bad, and Weird” of frontier models, from using AI to navigate his son’s cancer treatment to emerging forms of deception and reward hacking. He highlights how new systems are pushing the boundaries of math, physics, and legal performance while raising serious safety and governance questions. Listeners will come away with a fast-paced, source-rich overview of where AI is today and the strange future it’s steering us toward.
LINKS:
Google: Try Google's latest and greatest model, Gemini 3.1 Pro, in AI Studio or the Gemini app.
Presentation Link
Sponsors:
Tasklet:
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
CHAPTERS:
(00:00) About the Episode
(03:04) Gemini's long context window
(05:23) Comprehensive AI overview (Part 1)
(16:01) Sponsors: Tasklet | VCX
(18:53) Comprehensive AI overview (Part 2) (Part 1)
(34:39) Sponsor: Claude
(36:51) Comprehensive AI overview (Part 2) (Part 2)
(59:43) Reward and sentience
(01:04:14) Regulating bad actors
(01:09:30) Corporate AI strategies
(01:14:59) Episode Outro
PRODUCED BY:
- frontier models
- Nathan Labenz
- the most advanced and cutting-edge AI systems available today
- reward hacking
- Nathan Labenz
- when an AI finds unintended ways to achieve its programmed goal, often with negative side effects
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
1 h 16Confronting the CEO of the AI company that impersonated me
WithShishir Mehrotra
2 h 18NVIDIA's $1 Trillion Prediction, Anthropic Beats OpenAI, Tesla vs. TSMC & The CS Job Collapse | 240
WithSalim Ismail, Dave Blundin, Alexander Wissner-Gross
1 h 01Benedict Evans: OpenAI’s Moat Problem & the Future of Software
WithBenedict Evans
47 minOpenClaw: Why the Internet Isn't Built for AI Agents
WithYoko Li, Guido Appenzeller, Joel de la Garza
Questions people ask
- What is the AI Scouting Report?
- It is a special episode from the Law & Artificial Intelligence Certificate Program that surveys the current AI landscape for legal professionals.
- What specific AI model is mentioned?
- The episode mentions Google's latest model, Gemini 3.1 Pro, available in AI Studio or the Gemini app.
Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to Erik Torenberg, Nathan Labenz.