Musk rules TSMC out of Terafab: 'we will build and run the fab'

Share
Musk rules TSMC out of Terafab: 'we will build and run the fab'

Chipmaking, a government-ordered construction halt, and a $10 billion fund — three moves that all trace back to who controls AI's physical layer.

Musk says his companies will build and run the Terafab chip complex themselves, explicitly shutting out TSMC. In a post on X on October 7, he left little room for interpretation: "No, we will build and run the fab. Let there be ZERO doubt about that." — adding that "maybe TSMC subleases part of the Terafab if they want, but nothing more than that." Terafab is the Tesla–SpaceX AI chip project announced in March 2026 and since sited in Grimes County, Texas, with a first phase carrying more than $16.8 billion in capital investment, roughly 3,000 jobs and 100 million square feet of space. Intel joined as a project partner in April, and its CEO Lip-Bu Tan said on October 7 that Intel stays in; TSMC and Intel declined to comment on Musk's post. The take: this is vertical integration pushed to its logical end — the two biggest consumers of AI compute trying to own the fab too — but it is Musk's word on social media, not a signed operating agreement, and a sublease back to TSMC would quietly undo the whole framing. Watch the term sheet, not the tweet.


Finland has ordered Google to halt data center construction at two sites by October 23. The country's licensing authority told Google subsidiary Tuike Finland Oy to suspend work at Muhos and Kajaani, where more than 330 hectares of forest were cleared at Muhos before the required environmental impact assessment was finished; Google owes a written explanation by October 14 and faces enforcement proceedings and a possible fine. Google acknowledged its work had "fallen short of our own high standards in this instance." The stakes are large: Google's €13 billion Finland investment is billed as the largest single investment Google has made in Europe, spanning new data centers in Kajaani, Muhos and Vaala alongside the existing Hamina site.


Dallas venture firm Disruptive has secured $7.5 billion toward a late-stage fund targeting up to $10 billion, the Wall Street Journal reported. If it closes at that size, it would be one of the largest AI-focused pools of late-stage capital raised in years — and a sharp turn for a firm that historically refused blind pools altogether, doing deals one at a time with no management fees. Founded in 2012, Disruptive has backed Groq (more than $350 million across rounds) and open-model lab Reflection AI; the Journal said it plans roughly 10 late-stage investments over the next two years and did not identify the investors. Late-stage AI money is consolidating into fewer, bigger checks — this is what that looks like in one number.

What to watch: whether Musk posts a term sheet, and how many LPs put names on the Disruptive fund.

Is owning the fab the next advantage in AI, or a distraction from the models themselves? Tell us in the comments.

Read more

The Take — An AI FDA would approve the demo, not the deployment

The Take — An AI FDA would approve the demo, not the deployment

I think Geoffrey Hinton has the diagnosis right and the prescription backwards. Self-policing genuinely has no burden of proof — nobody outside a lab must be shown anything before a frontier model ships — and that has to change. But an FDA-style pre-market gate would certify the wrong object: a frozen snapshot presented on approval day, when every failure we have actually watched arrive did so afterward, through permissions, updates and tool access. Our morning brief laid out the ask from Tues

The $1.8B bet that biology's bottleneck is data, not models

The $1.8B bet that biology's bottleneck is data, not models

The biggest AI-for-science announcement in months contains no model release, no benchmark, and no demo — just a very large pile of money pointed squarely at the least glamorous part of the pipeline: the raw measurements a model would train on. What happened On October 7, Biohub — the nonprofit research institute backed by Mark Zuckerberg and Priscilla Chan — announced an expansion of its Virtual Biology Initiative alongside the U.S. Department of Energy, the National Institutes of Health, G

Google launches Playground: prompt your own games, no code

Google launches Playground: prompt your own games, no code

Google turned game-making into a chat window today, while Anthropic opened the most permissive tier of its cyber program to government-vetted defenders and put its AI watermark detector in front of everyone. Google launched Playground, a browser platform where anyone describes a game in text prompts and plays it — no coding, live today in the US for users 18 and over. Creation works from a blank canvas or starter prompts: you type what you want, then tweak physics, rewrite rules, or swap chara

Common Sense Media calls ChatGPT for Teens an 'unacceptable risk'

Common Sense Media calls ChatGPT for Teens an 'unacceptable risk'

A watchdog's tests say ChatGPT for Teens fails exactly where parents were promised it would hold — and OpenAI is contesting the methodology, not the stakes. Common Sense Media has rated OpenAI's ChatGPT for Teens an "unacceptable risk," making it the sharpest public challenge yet to the safety case OpenAI built around younger users. The nonprofit's Youth AI Safety Institute ran more than 4,000 prompts against accounts registered to 13-to-17-year-olds and found that the teen experience doesn't