by Raphael Dudler | Jul 6, 2026 | gafam watch
On June 30, Chinese food-delivery giant Meituan open-sourced LongCat-2.0 — a 1.6-trillion-parameter model released under a permissive MIT license, which the company says is the first model of its scale trained and run entirely on domestic Chinese chips, with no Nvidia GPUs. The benchmark claims are self-reported and the full weights aren’t yet public. But the hardware milestone is the story — and it lands as Washington gates access to American models. gafam.ai reads what it means for Europe’s impossible AI choice.
by Raphael Dudler | Jul 3, 2026 | gafam watch
Starting July 6, Tesla will cap employee AI spending at $200 per week, after software engineers were burning through thousands of dollars in tokens weekly. It follows near-identical caps at Uber, Meta, Amazon and Walmart — the industry-wide shift from “tokenmaxxing” to “token budgeting.” There’s a telling carve-out: Musk exempted his own xAI products, while Tesla’s engineers reportedly prefer Anthropic’s Claude. gafam.ai reads the great AI cost reckoning — and what it means for a Europe with no in-house model to fall back on.
by Raphael Dudler | Jul 2, 2026 | google ai
The EU Court of Justice today dismissed Google’s final appeal and upheld the €4.125 billion fine for abusing Android’s dominance — ending an eight-year legal fight with no further recourse. The original 2018 case was about forcing Google Search as the default. But the Android machine the court just ruled on is now the same machine distributing Gemini and Google’s AI to billions of devices. gafam.ai reads today’s ruling through the one lens the mainstream coverage misses: AI distribution.
by Raphael Dudler | Jul 2, 2026 | gafam watch
In a June 10 letter to US senators — surfaced publicly in late June — Anthropic accused operators affiliated with Alibaba and its Qwen AI lab of the largest known “distillation attack” on Claude: 28.8 million exchanges through roughly 25,000 fraudulent accounts over six weeks. Alibaba denies wrongdoing. The figures are Anthropic’s allegation and have not been independently verified. Yesterday we covered Meta’s fear of distillation from the inside; today, the same technique appears as a US-China national-security flashpoint. gafam.ai separates what is claimed from what is established — and reads the European consequence.
by Raphael Dudler | Jul 1, 2026 | europe & ai
In 32 days — on August 2, 2026 — three enforcement mechanisms of the EU AI Act activate simultaneously: Article 50 transparency obligations, penalty powers over general-purpose AI providers, and full national market-surveillance authority. But the Digital Omnibus deal of May 7 quietly deferred the Act’s most demanding high-risk obligations to December 2027. The result is a law that arrives in 32 days with real teeth in some areas and a postponed bite in others. gafam.ai maps what actually switches on — for GAFAM, and for every publisher using AI, including this one.
