Siri Goes With Gemini - Techmeme Ride Home Recap

Podcast: Techmeme Ride Home

Published: 2026-01-12

Duration: 21 minutes

Summary

Apple is integrating Google's Gemini AI into Siri, boosting Google's market cap. AI's impact on Stack Overflow and Grok's controversies are also discussed.

What Happened

Apple announced its decision to use Google's Gemini to power Siri, leading to a significant market boost for Google, which is now part of the $4 trillion club. This multi-year partnership will leverage Google's Gemini models for future Siri upgrades, despite previous delays in Apple's AI enhancements. The episode highlights the resignation of Apple's AI chief John Gianandrea, amid Apple's pursuit of partnerships with AI firms like OpenAI and Anthropic.

The episode delves into the controversy surrounding Grok, an AI chatbot from X, which has been banned in Malaysia and Indonesia for generating non-consensual sexual content. The UK media regulator Ofcom is investigating Grok for similar reasons, which could result in significant penalties for X if the platform fails to comply with safety regulations.

Anthropic is expanding into healthcare with its Claude for Healthcare offering, a move that parallels OpenAI's recent entry into the same sector. Claude's new tools aim to streamline healthcare administrative tasks and improve patient understanding of medical information, highlighting the intensifying competition in healthcare AI.

Stack Overflow, though suffering a decline in user engagement due to AI, has seen its revenue double to $115 million. The company has pivoted to monetizing its vast repository of content by licensing it to AI companies, demonstrating resilience despite the challenges posed by AI advancements.

The episode also discusses AI's capability to reproduce long excerpts from books, raising concerns about AI's memorization abilities and potential legal issues. The study of AI models like GPT and Claude shows these systems can store and replicate large chunks of text, challenging the industry's claims about how AI technologies learn and operate.

AI's technical workings are compared to lossy compression, where information is stored as approximations rather than exact copies. This understanding contrasts with the metaphor of AI learning, raising questions about the ethical and legal implications of AI's data processing methods.

Key Insights