Five Things AI: Cursor, Safety, Coders, Europe, Costs
There is so much AI out there! Find out what it does!
Heya and welcome back to Five Things AI!
SpaceX is buying Cursor for $60 billion in stock it printed on Friday, which is less a strategy than a confession: Grok cannot win enterprise on merit, so SpaceX bought the editor enterprises already had open. Anthropic, meanwhile, turned safety into a moat and is quietly winning the talent it needs while OpenAI spends its energy fighting itself. The public has feelings about all of this too, and they are not the ones the industry was hoping for: Americans now narrowly back banning AI from software engineering, of all jobs, the one that spent a decade as the ladder out. Europe is still thinking its nineteen thoughts and committing to none of them. And the labs are moving everyone onto metered pricing, because somebody has to pay for the data centers, and it is going to be you.
Five Things AI. Let’s go.
SpaceX’s $60 Billion Deal to Buy Cursor Gives It More AI Coding Power
Cursor makes a tool that allows developers to toggle between different AI models, from OpenAI and Anthropic to xAI and Google and others. The company competes with Anthropic’s Claude Code and OpenAI’s Codex, which can write code, debug software and automate tasks.
The Cursor deal could help SpaceX win over lucrative enterprise clients, who have largely shied away from using its in-house chat assistant Grok, which competes with Claude and ChatGPT. Elon Musk has positioned Grok as an anti-woke, truth-seeking chatbot, and it remains behind the leading models.
I didn’t see this coming, but it shows just how desperate Musk is to get some users for Grok. Now that SpaceX is publicly listed, people will soon find out just how much he overpromised.
Anthropic’s Safety Superpower
Here’s the thing about these safety justifications: I think they work because, to Anthropic, they aren’t justifications. The company really believes that they are the only ones who believe in super intelligence, and thus are the only ones who are sufficiently concerned about the dangers. That excuses decision after decision, policy after policy, and confrontation after confrontation that, to people on the outside, look like a bizarre combination of cynicism and naiveté.
The contrast to OpenAI is massive: I think that one way to understand how and why OpenAI lost its lead is that, in the years following the release of ChatGPT, the company has been at war with itself internally as what used to be a research lab was suddenly seized with the burden of being the accidental consumer tech company; to the extent OpenAI solved that conflict, it was by bleeding huge amounts of talent to Anthropic in particular.
It’ll be interesting to see how the two IPO will look like.
Americans want artisanal code
Overall, Americans supported the ban on AI in software engineering by a split of 44 to 41. It’s a small margin but a notable one given that bans for many professions were underwater by double digits.
One obvious explanation for this is that the public responds to a drumbeat of news stories about whatever industry is being automated next. The industry with the second-most support for a ban was manufacturing, which was sitting at break-even support (44 to 44). For a long time, manufacturing has been the stereotypical job facing automation, but now that software engineering is in that spotlight, Americans are willing to protect it.
That explanation is probably true, but it’s also a big step from hearing that an industry could be automated to supporting a ban to protect it.
The surprising support for a ban on the automation of software engineering likely also stems from the fact that it has spent over a decade as the field that represented social mobility, including for those who didn’t go to an elite school.
I doubt that this sympathy for the coders will last long.
Nineteen thoughts on AI and Europe
It may seem implausible for a democracy to commit economic self-harm on that scale. But, through the energy transition, Europe has signaled that governments will tolerate destroying industries as important as the German car and chemical industries. It is not clear how you rebuild this trust.
If it does come down to a transatlantic stand-off for model access, the upside of combining forces in the European Union may be lower than the downside of increased coordination costs. There are many members of the EU with conflicted interests who might yield to pressure: Eastern Europe needs the American military presence in Europe to deter Russia. Having countries like these involved in decisions may diminish total European leverage.
These are some really thought-provoking points about Europe and AI, well worth a read. But Germany will not get back to nuclear power and why whould we? It’s too expensive, takes too long and we have plenty of wind and solar.
How Meter Pricing Is Testing the Economics of AI
A growing number of tech firms have begun to introduce usage-based pricing options for their AI services rather than simply charging a flat subscription fee. As a result, the heaviest users are set to incur additional costs each time they ask an AI chatbot or agent to produce a slide deck, draft an email or debug a complex piece of code.
Leading AI labs have been spending tens if not hundreds of billions of dollars on chips, data centers and talent to develop and run their models. The shift to a pricing model more commonly associated with metered-power or pay-as-you-go phone contracts may lead to more selective use of a technology that consumes huge amounts of electricity and other resources.
This was bound to happen and it make total economic sense.
