Nvidia has already committed $40B to equity AI deals this year
Nvidia isn’t just selling the picks and shovels in this gold rush — it’s buying up the mines too. $40 billion in equity deals in a single year means Nvidia is positioning itself not just as an infrastructure company but as the shadow investor of the entire AI ecosystem. At some point the chip business and the investment portfolio start to look like the same thing, and antitrust lawyers start getting interested.
OpenAI exec says company hopes to burn $50B of somebody else’s money on compute this year
Fifty billion dollars. In compute. In one year. The Register’s framing — “if the numbers are large enough, perhaps we won’t question the math” — is doing a lot of heavy lifting here, because that number deserves some questioning. OpenAI has mastered the art of making astronomical spending sound like responsible R&D, and at $50B, they’re not just burning money — they’re doing controlled burns on a continental scale.
Anthropic wants Claude to play with money, unleashes finance agents
Nothing says “we trust our AI” like letting it make financial decisions on your behalf. Anthropic — the company whose entire brand is built on AI safety — is now unleashing Claude into the world of finance agents, which is either a sign of genuine confidence in the model or a very expensive way to test that confidence. I’m sure the “always bet on backpropagation” energy will hold up fine when it’s your 401(k) on the line.
Musk v. Altman Evidence Shows What Microsoft Executives Thought of OpenAI
Emails from 2018 reveal that Microsoft executives were skeptical of OpenAI but feared pushing it into Amazon’s arms — so they held their nose and invested anyway. Arguably the most consequential case of “well, we can’t let them have it” in tech history. It’s the geopolitical logic of cold wars applied to startup funding, and it worked out pretty well for Microsoft, all things considered.
We’ve entered the era of AI malaise
MIT Tech Review is naming the feeling that a lot of people have been experiencing but struggling to articulate: AI is everywhere, it’s not going away, and yet… what exactly is it doing? The malaise isn’t about doom — it’s about the gap between breathless promises and the reality of autocomplete on steroids. When $50B in compute can’t definitively answer “is this making the world better,” that’s a meaningful question.
Sony PlayStation sees AI as a ‘powerful tool’ to help make games
Sony’s earnings presentation included a section on AI’s role in game development, which means we’re now in the phase where every major company needs a slide deck ready justifying their AI strategy to investors. The interesting tension here is that indie developers are largely rejecting generative AI while AAA studios are quietly embracing it — a cultural fault line that’s going to define gaming’s next chapter whether players like it or not.
Railway secures $100 million to challenge AWS with AI-native cloud infrastructure
Two million developers, zero marketing spend, and now $100 million in Series B funding — Railway’s pitch is that the legacy cloud giants built their infrastructure for a world that no longer exists, and AI applications need something purpose-built. They may be right. AWS and Azure have the scale, but scale can be a liability when the architecture is optimized for the wrong era.
A cyberattack on Canvas — timed exquisitely for finals week — exposed the catastrophic downside of centralizing sensitive student data in a single platform. We’re talking medical records, sexual assault allegations, accessibility accommodations, and more. The AI angle here is subtle but real: the same “centralize everything for smarter insights” logic driving EdTech is what makes these systems such juicy targets. Efficiency and vulnerability are two sides of the same coin.
Bottom Line
The theme today is money in motion — staggering sums being committed with staggering confidence, while the rest of us quietly wonder if anyone’s actually measuring the output.