LWiAI Podcast #248 - Opus 4.8, MAI, Anthropic IPO, Minimax-M3
Summary
Hosts Andrei Kerenkov and Jeremy Harris discuss Anthropic’s Claude Fable 5 release, highlighting its coding leaps and controversial safety safeguards. They analyze Apple’s Siri AI partnership with Google, Google’s price cuts, and the live translate feature. The episode also covers the competitive IPO race between OpenAI and Anthropic, plus Jeff Bezos’s $12 billion raise for physical AI startup Prometheus.
Topics discussed
Introduction and episode overview
Anthropic's Claude 5: Capabilities, alignment, and safety
User experience with Claude 5 in coding and research
Anthropic's IPO plans and market implications
Controversy over AI safety safeguards and downgrades
Apple's Siri AI update and strategic position
Google's AI translation and subscription pricing
The AI IPO race: OpenAI, Anthropic, and SpaceX
DeepSeek's funding, valuation, and hardware challenges
SpaceX's AI data centers and compute strategy
Open source models: Gemma 3 and diffusion language models
AI regulation and biological weapon risks
Dario Amodei's views on AI automation and policy
Research: AI agents, specification failures, and loopholes
US government discussions on AI compute and taxation
Music industry lawsuits against AI companies and conclusion
Listen ad-free on Castria