Anthropic Says It’s About to Have Its First Profitable Quarter

Anthropic has told investors it expects to more than double revenue to roughly $10.9 billion in Q2 — and that it will finally be profitable. For a company that’s burned through billions in compute costs while going toe-to-toe with OpenAI and Google, this is a genuinely significant inflection point. The AI arms race has a new wrinkle: one of the combatants is now making money.

An OpenAI Model Has Disproved a Central Conjecture in Discrete Geometry

An OpenAI model just cracked the 80-year-old unit distance problem, disproving a major conjecture in discrete geometry that mathematicians had been chipping away at for decades. This is the kind of result that separates “AI is a neat productivity tool” from “AI is doing things humans haven’t been able to do” — and it’s worth sitting with that for a moment. The math nerds are having a day.

xAI Burned $6.4B Last Year — SpaceX’s IPO Filing Shows Why the Spending Is Far From Over

The SpaceX IPO filing has given us the first real public window into xAI’s finances, and the number is a doozy: $6.4 billion in losses in 2025 alone, with no sign of a slowdown. Meanwhile, SpaceX is spending $2.8 billion on gas turbines to power Grok’s data centers — because apparently the carbon-footprint optics are a secondary concern when you’re trying to win an AI war. Oh, and Grok’s “spicy mode” was formally listed as a litigation risk in the filing, which is a sentence I didn’t expect to type today.

Jensen Huang Says He’s Found a ‘Brand New’ $200B Market for Nvidia

Jensen Huang has identified his next gold mine: CPUs purpose-built for AI agents, which he’s pegging as a $200 billion market opportunity. Coming off yet another record revenue quarter, Huang isn’t exactly known for underselling things — but he’s also not been wrong often enough to dismiss this out of hand. If agentic AI becomes the dominant computing paradigm, the infrastructure bets being placed right now are going to look either visionary or deranged, and we’ll know which in about three years.

Nvidia Posts Another Record Quarter, Reveals $43B of Holdings in Startups

Nvidia’s record quarter is almost a given at this point, but the buried lede is that the company is sitting on $43 billion in startup investments — which means Nvidia isn’t just selling shovels in the gold rush, it’s quietly buying stakes in the miners too. Revenue growth is expected to slow next quarter, which sent the usual jittery reactions through markets, but “slowing from hyperdrive to merely very fast” is not the crisis the headlines might suggest.

Township Leader Resigns in Tears Over OpenAI Data Center Death Threats

A township treasurer has resigned — in tears, on the record — after receiving death threats over a proposed OpenAI/Oracle Stargate data center in her community. Whatever your position on AI infrastructure buildout, this is a deeply ugly consequence of the hypercharged culture war that’s now attached itself to every corner of the technology. Local officials who have nothing to do with OpenAI’s corporate strategy are getting harassed into quitting, and that should disturb everyone regardless of where you stand on the data center debate.

Buried inside Google’s sweeping AI Search redesign is the part that was inevitable: Gemini will now surface products and generate “custom explainers” about why you should buy them. So the AI that’s replacing the list of blue links will also be replacing the ads that used to sit next to them — just in a more conversational, personalized, and frankly harder-to-identify-as-advertising format. I’m sure that will be fine.

You Can Now Remix Other People’s YouTube Shorts with AI

Google’s new YouTube Shorts Remix feature lets users use Gemini Omni to restyle clips or insert themselves into other people’s videos. The creative possibilities are real — and so are the obvious abuse vectors, which Google will presumably address sometime after the feature goes viral for the wrong reasons. “Reimagine” is doing a lot of work as a brand name here.

Bottom Line

The AI industry is entering a phase where the math is getting harder to argue with — record revenues, a genuine mathematical proof, and a first profit — but so are the human costs playing out in small towns and tearful resignations.