Anthropic is having a month.

The AI lab finished May by surpassing OpenAI in market share of business spending for the first time, Ramp just revealed. It raised $65 billion at a $965 billion valuation (also besting OpenAI) at the end of May, then waltzed into June by filing confidential paperwork for an IPO, reportedly on the strength of its first-ever profitable quarter.

Then on Friday, the Trump administration renewed its war on the model maker by sending a letter demanding it ban non-Americans, including Anthropic’s employees, from accessing its state-of-the-art models: the limited-release Mythos 5 and the more guarded version of Mythos released to the public three days earlier, called Fable 5.

This essentially forced Anthropic to pull its latest all-powerful model from the market altogether.

Although the White House invoked an obscure export control directive when ordering the ban, the exact cause remains unclear. The chatter was that hackers easily bypassed Fable 5’s guardrails, which were intended to prevent access to Mythos’ capabilities. That model is so good at finding security flaws in software code that Anthropic itself marketed it as dangerous and restricted its public release.

This new drama comes after Anthropic famously refused to allow the government to use its models for mass surveillance of Americans and fully autonomous weapons. As a result, in March, the Trump administration declared the company a supply-chain risk.

That didn’t deter Anthropic’s sales to businesses. Quite the opposite, Ramp’s data shows. Ironically, this latest feud with the Trump administration, which also appears to validate the hubbub over Mythos’ mythological power, may help rather than hurt Anthropic, according to Ramp’s lead economist, Ara Kharazian. Kharazian is the person who compiled the business-spending AI data.

“If anything, it’ll probably boost them,” Kharazian told TechCrunch. “Anthropic’s best month on record, as far as business adoption, was the month that the Department of Defense labeled them a supply-chain risk. There’s a lot of aura that comes with your model specifically being named too dangerous to use.”

Ramp’s data isn’t granular enough for us to see how much of a financial hit the company will take by pulling Mythos and Fable 5 off the market.

Still the data, from more than 70,000 businesses that use its platform, shows that customers heavily use Anthropic’s Opus models and that business use has been growing.

For instance, Ramp reported that Anthropic’s share of AI subscriptions paid for by businesses rose 2.5 percentage points in May to 41%. This compares to OpenAI, which commanded 39.5% of AI subscriptions by its customers, essentially flat from the prior month. (OpenAI still greatly leads Anthropic in overall consumer usage, according to new data from Sensor Tower.)

Beyond subscriptions, the vast majority of what companies spend money on is API calls to the model, which cover token use for activities like coding. Anthropic’s Claude Code has a strong reputation as a powerful AI coding tool.

Ramp can’t always see from the spending data which models most businesses are using. When it can see the model details — in about one-third of transactions — businesses are mostly spending on various flavors of Claude Opus, particularly the later versions. Opus is the model that preceded Mythos and is still openly available.

In fact, in late May, Anthropic released a new version, Opus 4.8.

Mythos had not been on the market for that long, having been released to limited users as of April. And Fable 5 was shut down after a few days.

While we can’t predict how this latest drama with the White House will impact Anthropic’s ability to go public as it hoped to (public-market investors tend to be wary of companies embroiled in controversies with the government), the numbers indicate that Anthropic’s available models are more popular with businesses than ever before.

Martin Scorsese has signed on as a partner and adviser to AI image-generation startup Black Forest Labs, The New York Times reported on Tuesday.

The caveat is that one of the world’s most famous living directors is using the tech solely for storyboarding.

“For 70 years, I’ve been creating my own storyboards,” he said in a statement to the Times. The tool, he said, helps him communicate his vision to cinematographers and production designers far faster and more efficiently.

Black Forest Labs is a 70-person outfit headquartered not in San Francisco, but in Freiburg, Germany, the closest major city to the actual Black Forest. Despite its unlikely address, the startup powers image features inside Adobe, Canva, Microsoft, and Meta, and was last valued at $3.25 billion by its investors, which include BroadLight Capital, co-founded by Scorsese’s talent manager, Rick Yorn.

Black Forest Labs was founded by the team behind Stable Diffusion and according to Wired, declined to partner with Elon Musk’s xAI in recent months after an earlier collaboration on Grok’s image generator ended amid concerns about the platform’s content safeguards.

You can imagine that some in the entertainment industry will be concerned about the development, even given its limited scope. Still, it’s just the newest sign that Hollywood’s once-fierce resistance to AI is softening.

Pacific Fusion took the wraps off its latest pulser module prototype on Tuesday, a piece of equipment that allows the company to move ahead with its demonstration fusion power plant. Construction on the fusion power plant is expected to begin this summer.

Results from the shipping container-sized prototype were good enough to unlock another tranche of Pacific Fusion’s Series A round, which exceeds $1 billion, the company exclusively told TechCrunch. The company did not disclose the size of the tranche. Pacific Fusion is among the best-funded fusion startups.

The tranche-based model is more widely used in biotech, where it saves startups time on fundraising, allowing them to remain focused on hitting technical milestones. 

The funding arrangement has allowed the company “to keep our heads down,” Pacific Fusion CTO Keith LeChien told TechCrunch. “It means that we can lean into the future without spending 20% to 50% time constantly looking for the next piece of capital.”

Pacific Fusion is pursuing a form of fusion power known as inertial confinement. Its reactor will use 156 pulser modules to deliver an enormous jolt of electricity to a small fuel target in the fusion chamber. That electrical pulse will create a magnetic field around the eraser-sized fuel pellet, squeezing it until atoms inside fuse and release enormous amounts of energy.

The startup’s next challenge will be to scale up from the sub-scale prototype to a full-size pulser module, the core component of the demonstration power plant. The company hopes that power plant will be able to produce more electricity than the facility requires to operate, a feat no one has achieved yet.

But because the race for fusion power is heating up, the company isn’t waiting for results from a full-scale pulser module before starting work on the demonstration power plant. “The shovels go in the ground for that facility this summer,” LeChien said.

To date, inertial confinement is the only way humans have been able to produce a controlled fusion reaction that releases more energy than was required to start it, a milestone known as scientific breakeven. And so far, only one experiment, at the National Ignition Facility (NIF) at Lawrence Livermore National Laboratory, has been able to produce and replicate those results.

But where NIF relies on large, expensive lasers, Pacific Fusion is hoping it only needs thousands of less costly electrical switches and capacitors. Those capacitors and switches will be coordinated to generate enormous, precisely timed pulses of electricity — each about 100 nanoseconds long. 

Pacific Fusion’s challenge is ensuring the capacitors can release their energy at precisely the right moment. If they don’t, the fuel pellet won’t be hit with sufficient energy to compress quickly enough to spark a fusion reaction.

The company’s demonstration device will have 156 full-size pulser modules. Each module will contain 32 circular stages, and each stage will be ringed with 10 bricks. One brick contains two capacitors to store power and one switch to release it. 

The prototype pulser module that the company recently tested is about a third the size of the full module. It contains nine stages and 90 bricks, and it released 440 gigawatts of peak power in just 80 nanoseconds.

“It meets all our requirements for scaling up to build our big demonstration system,” LeChien said. 

Once Pacific Fusion has its demonstration power plant ready, it intends to skip past scientific breakeven and go straight for facility breakeven, in which its demonstration device generates enough energy to power the entire facility. 

“Any fusion approach, regardless of your specific technology, has to traverse through that,” LeChien said. “It’s the next tectonic milestone in fusion.”