Apple Bets Big on Google Gemini to Power a Rebuilt Siri
Today's WWDC 2026 keynote in Cupertino delivered what may be the most significant Siri overhaul in the assistant's 15-year history. Apple unveiled a completely rebuilt Siri powered by a custom 1.2-trillion-parameter Google Gemini model, licensed for approximately $1 billion per year — a deal that marks Apple's clearest admission yet that it needs outside help to compete in the AI race.
The new Siri gains a chat mode for extended conversations, personal-context access spanning emails, photos, and files, on-screen awareness, and deeper cross-app actions across iOS 27, macOS 27, watchOS 27, and visionOS 27. Perhaps most notably, Apple is also introducing an Extensions system that lets users choose between ChatGPT, Gemini, and Anthropic's Claude as the AI model powering Apple Intelligence features — an unusually open move for a company known for tightly controlling its ecosystem.
This is also Tim Cook's final WWDC as CEO, adding a symbolic weight to the announcement. After two years of playing catch-up in AI, Apple appears ready to leapfrog competitors by partnering rather than building everything in-house.
Sources: MacRumors, Gizmodo, Let's Data Science
Anthropic Calls for Global AI Development Pause
In a sweeping proposal published June 4 through its Anthropic Institute, the AI lab warned that frontier models are approaching recursive self-improvement — the point where an AI system can design and develop its own successor without human intervention. The company is urging the world's top AI labs to develop a coordinated pause mechanism, essentially a "brake pedal" for the industry.
The numbers backing the warning are striking: more than 80% of code merged into Anthropic's own codebase is now written by Claude, engineers are shipping roughly eight times more code per quarter than before 2025, and AI task-completion capability has been doubling every four months.
The timing has drawn scrutiny. Anthropic confidentially filed its S-1 with the SEC on June 1, just days before the safety announcement, and is eyeing a potential October IPO at a $965 billion valuation. Critics note the company also walked back a key safety pledge in February, saying it would no longer withhold potentially dangerous AI if competitors were close to matching its capabilities. Whether this is genuine concern or strategic positioning — or both — remains the subject of heated debate.
Sources: Al Jazeera, Fortune, CNN
Trump and Sanders Find Common Ground on AI Ownership
In one of the more unexpected political developments of the year, President Trump and Senator Bernie Sanders are converging on the idea that the American public should own a piece of the AI industry. Trump said on June 5 that the U.S. government may take direct equity stakes in AI giants like OpenAI, Anthropic, and xAI, calling the concept "a beautiful thing."
Sanders has gone further with his American AI Sovereign Wealth Fund Act, which would impose a one-time 50% tax — paid in stock, not cash — on major AI companies, granting the federal government voting shares and equal board representation. Sanders met with OpenAI CEO Sam Altman on June 4 to discuss the model; Altman reportedly said he couldn't support the 50% threshold but endorsed the general concept.
The Trump administration is already directing the Treasury and Commerce departments to develop a broader U.S. sovereign wealth fund plan. Whether any legislation can bridge the gap between Sanders' aggressive proposal and the administration's more business-friendly approach remains unclear, but the bipartisan momentum is real.
Sources: Fortune, CNBC, Washington Times
The AI IPO Race Heats Up
The race to be the first major pure-play AI lab on public markets is fully underway. Anthropic confidentially filed its S-1 with the SEC on June 1 after closing a staggering $65 billion Series H that valued the company at $965 billion — surpassing OpenAI for the first time. With a revenue run-rate near $47 billion (up from roughly $9 billion at end of 2025), Anthropic is targeting an October listing.
OpenAI is not far behind, working with Goldman Sachs and Morgan Stanley on its own IPO targeted for September 2026 at a $730–850 billion valuation. The two companies are now in a direct sprint to Wall Street, and the outcome could reshape how investors value the entire AI sector.
Sources: Fortune, NPR, Yahoo Finance
Quick Hits
xAI lands federal contract: The GSA awarded xAI an 18-month OneGov agreement giving all federal agencies access to Grok 4 at $0.42 per agency through March 2027.
Colorado AI Act countdown: Colorado's comprehensive AI legislation takes effect June 30, requiring security risk management programs, impact assessments, and measures to prevent algorithmic discrimination — making it one of the strictest state-level AI regulations in the U.S.
Meta doubles down on infrastructure: Meta raised its planned investment in an El Paso, Texas AI data center from $1.5 billion to over $10 billion, expanding the facility to one gigawatt of capacity by 2028.