Inside Anthropic’s IPO + How AI Agents Spy on Your Messages
Oct 28, 2026 · 53m
More Technology podcasts
Summary
Kevin and Casey analyze Anthropic’s S-1 filing, highlighting its massive revenue growth, flat cap table, and unique risk disclosures regarding AI safety and existential threats. They debate the implications of the company’s public benefit structure and its potential impact on San Francisco’s economy and inequality. The episode concludes with a discussion with Meredith Whitaker, president of the Signal Foundation, who addresses how new AI agents and screen-recording features threaten the integrity of encrypted messaging.
Topics discussed
Sponsor: 80,000 Hours and episode preview
Anthropic files for IPO and host disclosure
Explaining the S-1 filing process
Anthropic's valuation and revenue growth
Cash flow and the AI bubble debate
Cap table, control, and safety risks
Public benefit structure and model risks
Market response and Sam Altman's absence
IPO timing and bubble concerns
Impact on San Francisco housing and inequality
Sponsors: 80,000 Hours, Mint Mobile, Aviator
Introduction to Meredith Whittaker
The importance of private communication
AI agents and the erosion of privacy
Meta Muse and screen recording risks
Microsoft Recall and OS-level surveillance
Sponsors: Mercury, 80,000 Hours, LinkedIn
Right to privacy with AI systems
Government threats to encryption globally
AI ethics, marketing terms, and real risks
Supporting Signal and final privacy advice
Sponsors: Doppel, 80,000 Hours, LinkedIn
Correction: The talking skeleton story
Credits and final sponsors
More from Machine Gods
Listen ad-free on Castria