Jensen Huang Declares AGI Has Arrived
In a wide-ranging conversation on the Lex Fridman podcast, Nvidia CEO Jensen Huang made waves by declaring that artificial general intelligence is no longer a future milestone — it's here now. Huang defined AGI as systems capable of autonomously building and running billion-dollar companies, tying the claim to the rapid progress in agentic AI and multimodal models.
The statement reframes industry timelines considerably. While some researchers push back on the definition, Huang's confidence reflects Nvidia's unrivaled vantage point as the company supplying the compute infrastructure powering virtually every frontier AI lab. His comments immediately boosted investor sentiment across the AI chip sector.
OpenClaw: The Open-Source Agent Framework Taking the AI World by Storm
OpenClaw, an open-source AI agent framework developed by independent Austrian developer Peter Steinberger, has gone viral this week — with Nvidia CEO Jensen Huang himself calling it "the next ChatGPT." The framework demonstrates that fully autonomous AI agents can run locally on personal computers without relying on expensive cloud-based APIs.
OpenClaw's appeal lies in its accessibility: developers can build and deploy sophisticated agentic systems on consumer hardware, dramatically lowering the barrier to entry for AI agent development. The project has sparked intense discussion about whether the future of AI belongs to massive cloud providers or to a decentralized ecosystem of local, open-source tools.
Apple Gets Green Light to Distill Google's Gemini for Siri
In a significant development for the Apple-Google AI partnership, Google has granted Apple full permission to "distill" its Gemini model — meaning Apple can now create smaller, customized versions of Gemini optimized specifically for Siri and other Apple Intelligence features. Some of these distilled models could even run entirely on-device without an internet connection.
This moves the companies closer to shipping a fundamentally rebuilt Siri, though the timeline has slipped. While iOS 26.4 was originally targeted for the first Gemini-powered Siri features, 9to5Mac reports the rollout has been pushed to iOS 26.5 or even iOS 27. Bloomberg previously reported that Apple is paying Google roughly $1 billion per year for custom Gemini access, with all processing running on Apple's Private Cloud Compute to maintain its privacy guarantees.
Google Releases Gemini 3.1 Flash-Lite
Google quietly released Gemini 3.1 Flash-Lite, a new model offering 2.5x faster time to first token compared to its predecessors. Priced at just $0.25 per million input tokens, it's one of the most affordable frontier-adjacent models on the market.
The standout feature is "adjustable thinking levels," which let developers modulate how much reasoning effort the model applies to each query — a practical tool for balancing cost, speed, and quality in production applications. This positions Flash-Lite as a strong contender for high-volume, latency-sensitive workloads where full frontier model power isn't necessary.
CFOs Signal AI-Driven Workforce Reductions Ahead
A new Wall Street Journal survey found that chief financial officers across major companies expect AI to reduce headcount in 2026, with administrative and back-office roles particularly exposed. While the projected overall reductions are modest, the signal from finance leadership is clear: AI-driven workforce restructuring has moved from theoretical discussion to concrete budget planning.
The findings come in the wake of high-profile moves like Block's 40% staff reduction (attributed by CEO Jack Dorsey to AI capabilities), Atlassian's 1,600-person layoff to redirect resources toward AI, and reports that Meta's Mark Zuckerberg has been telegraphing similar logic internally. Three distinct corporate strategies are emerging: slash-and-rebuild (Block), augment-and-redeploy (Salesforce), and invest-massively-and-concentrate-talent (Meta).
In Other News
Anthropic launched auto mode for Claude Code, enabling AI-assisted coding with fewer manual approvals while maintaining built-in safety reviews. ARM unveiled its AGI CPU, marking the company's first move from licensing chip designs to manufacturing its own silicon for AI data centers. The U.S. Treasury Department launched its AI Innovation Series, a public-private initiative to support financial system resilience in the age of AI. And Cursor's Composer 2 sparked controversy after it emerged the product was built on top of Moonshot AI's Kimi K2.5, with Cursor executives clarifying that 75% of compute came from proprietary reinforcement learning.