1. Google's Gemini 3.5 Pro Hits General Availability
Google's Gemini 3.5 Pro entered its general-availability window this week, with analysts placing wide release between June 23 and June 30. The model ships with a 2-million-token context window — double Gemini 3.5 Flash's 1M and the largest in any production frontier model — plus a "Deep Think" reasoning mode gated to the $250/month Ultra tier. It lands alongside Google's biggest Search overhaul in 25 years, with AI Mode now running on Gemini 3.5 Flash.
2. Claude Fable 5 Free Window Closes After Export-Control Drama
Anthropic's 13-day complimentary access to Claude Fable 5 ended June 23; the model now requires paid credits at $10 per million input tokens and $50 per million output — double Claude Opus 4.8. The rollout was rocky: a June 12 U.S. export-control directive forced Anthropic to pull Fable 5 and Mythos 5 offline globally, and the API reportedly stayed down until roughly June 18.
3. Claude Opus 4.8 Breaks 60 on the Intelligence Index
Claude Opus 4.8 took the #1 spot on the Artificial Analysis Intelligence Index with a score of 61.4 — the first model to clear 60 by a clean margin. The milestone caps a year in which SWE-bench Verified coding scores climbed from roughly 60% to near 100%, and frontier models began matching or beating human baselines on PhD-level science, multimodal reasoning, and competition math.
4. Microsoft Unveils a Full In-House MAI Model Family
Microsoft introduced a suite of models trained in-house — not by OpenAI — for Azure AI Foundry: MAI-Thinking-1 (reasoning flagship), MAI-Code-1-Flash (coding), MAI-Image-2.5 (image generation), MAI-Transcribe-1.5 (described as state-of-the-art transcription), MAI-Voice-2 (speech), and two further reasoning-specialized models. The lineup signals Microsoft's continued push to reduce its dependence on OpenAI.
5. AI Infrastructure Debt Set to Near $570B in 2026
Morgan Stanley projects global AI-linked debt will nearly double to $570 billion this year. Marquee deals include a $36 billion Apollo–Blackstone financing to fund Google TPU purchases for Anthropic, Microsoft's $190 billion capex commitment, and OpenAI's Stargate venture targeting 10 gigawatts of data-center capacity — underscoring how the buildout is increasingly debt-financed.
6. FERC Orders Grid Operators to Speed Up Data-Center Hookups
On June 18, the Federal Energy Regulatory Commission issued "show cause" orders under Section 206 of the Federal Power Act to all six U.S. regional grid operators. Each must defend its existing interconnection framework or propose reforms that let large-load customers — explicitly AI data centers — connect to the grid faster, a direct regulatory response to surging compute power demand.
7. Google DeepMind Loses a Second Top AI Exec in a Week
DeepMind VP and Nobel laureate John Jumper — co-creator of AlphaFold — departed, becoming the second senior AI leader to exit the lab within a week. Alphabet shares fell on the news, highlighting how sensitive markets have become to talent stability at the frontier labs amid an intensifying hiring war.
8. White House AI Framework Pushes Federal Preemption of State Laws
The Trump administration's National Policy Framework for AI continues to shape 2026 policy, recommending against any new federal AI regulator in favor of existing agencies and industry-led standards. It also urges Congress to preempt state AI laws deemed "unduly burdensome," while carving out exemptions for child safety, compute and data-center infrastructure, and state-government procurement.
// KEY TAKEAWAYS
Three forces dominated this week: a frontier-model arms race (Gemini 3.5 Pro, Opus 4.8 clearing 60, Microsoft's in-house MAI family), the hard physical and financial constraints behind it (≈$570B in AI debt and FERC scrambling to power data centers), and a maturing policy fight over who regulates AI. With Fable 5's export-control turbulence and DeepMind's executive churn, capability gains are now bumping up against capital, energy, talent, and law.