
Mistral says ML4 was trained using 3,800 Nvidia Grace Blackwell GPUs in its own data centers in Europe and much of its training data was multilingual
THE SO WHAT
Running 3,800 Grace Blackwell GPUs in owned European data centers with multilingual training data is a clear play for regional sovereignty — on both infra and language coverage. If you operate in or sell into Europe, this strengthens the case for a “local-first” model strategy on latency, data residency, and cultural fit.
READ THE SOURCE
MORE FROM THE WIRE
Applied AIGoogle launches Playground to turn prompts into games, with Unity on board
Prompt-to-game with Unity in the loop is a shot at compressing early prototyping cycles from weeks to hours. Studios and tool makers should treat this as a testbed for where AI-native workflows can sit in the game stack — narrative, level grayboxing, or full mechanic exploration.
Applied AIAnthropic is offering startups a free year of Claude Team and $1,000 in credits to kickstart their AI journey
A free year of Claude Team plus credits is Anthropic buying default status with early-stage companies. Founders should treat this as a low-cost way to standardize on one model stack for 6–12 months — but keep abstraction layers in place so you can switch if pricing or capabilities shift.
The funniest, most bizarre, and just plain weird moments from Kevin Roose's new book about OpenAI and Anthropic
The “weird moments” history of OpenAI and Anthropic will shape how non-technical stakeholders perceive AI labs — as volatile, personality-driven institutions. If you’re selling into enterprises, assume your buyers are reading this and adjust your narrative to emphasize governance, stability, and clear lines of responsibility.
Applied AIAleph Alpha and Mistral releases raise hopes for Europe’s sovereign AI ambitions
European labs shipping frontier-class models is less about catching US labs and more about giving EU governments and corporates a politically acceptable stack. If you operate in Europe, assume procurement and data residency pressure will tilt toward these “sovereign” vendors and plan for a dual-vendor model strategy.