Top AI experts at OpenAI, Anthropic and other companies warn of rising dangers of their technology, with some quitting in protest or going public with grave concerns.

Why it matters: Leading AI models, including Anthropic's Claude and OpenAI's ChatGPT, are getting a lot better, a lot faster, and even building new products themselves.

• That excites AI optimists — and scares the hell out of several people tasked with policing their safety for society.

Driving the news: On Monday, an Anthropic researcher announced his departure, in part to write poetry about "the place we find ourselves."

• An OpenAI researcher also left this week citing ethical concerns. Another OpenAI employee, Hieu Pham, wrote on X: "I finally feel the existential threat that AI is posing."
• Jason Calacanis, tech investor and co-host of the "All-In" podcast, wrote on X: "I've never seen so many technologists state their concerns so strongly, frequently and with such concern as I have with AI."
• The biggest talk among the AI crowd yesterday was entrepreneur Matt Shumer's post comparing this moment to the eve of the pandemic. It went mega-viral, gathering 56 million views in 36 hours, as he laid out the risks of AI fundamentally reshaping our jobs and lives.

Reality check: Most people at these companies remain bullish that they'll be able to steer the technology smartly without societal damage or big job loss. But the companies themselves admit the risk.

• Anthropic published a report showing that, while low risk, AI can be used in heinous crimes, including the creation of chemical weapons. This so-called "sabotage report" looked at the risks of AI without human intervention, purely on its own.
• At the same time, OpenAI dismantled its mission alignment team, which was created to ensure AGI (artificial general intelligence) benefits all of humanity, tech columnist Casey Newton reported yesterday.

Threat level: The latest round of warnings follow evidence that these new models can build complex products themselves and then improve their work without human intervention.

• OpenAI's last model helped train itself. Anthropic's viral Cowork tool built itself, too.
• These revelations — in addition to signs that AI threatens big categories of the economy, including software or legal services — prompted a lot of real-time soul-searching.

The bottom line: The AI disruption is here, and its impact is happening faster and more broadly than many anticipated.

The share of U.S. companies paying for Anthropic's AI tools and services jumped in January, per new data.

Why it matters: It was Anthropic's breakthrough month, writes Ara Kharazian, an economist at Ramp, which has been tracking business spending on AI.

By the numbers: The share of companies paying for Anthropic increased to 20% from 17%, per Ramp, a company that offers corporate credit cards and expense-management tools to roughly 50,000 companies nationwide.

• OpenAI dropped slightly from 37% to 36%.
• 1 in 5 businesses that use Ramp now pay for Anthropic, up from 1 in 25 last year.

The big picture: The competition between Anthropic and OpenAI is shaping up to be the Kendrick vs. Drake tech battle of our time.

• But you can enjoy listening to "Not Like Us" and "God's Plan." Similarly, companies appear willing to pay for both companies' tools.
• Anthropic isn't gaining users at OpenAI's expense — at least so far, per the report. According to Ramp, about 79% of Anthropic users also pay for OpenAI.

Between the lines: Anthropic's new software coding product, which went viral earlier this year, helped drive adoption.

Yes, but: The Ramp data doesn't take into account workers inside companies who are using free AI tools — which would skew the numbers more in OpenAI's favor, as ChatGPT remains the leader for consumers overall.

• And Ramp's data skews toward more tech-forward early adopter type companies; not the full breadth of the business sector.
• A December 2025 report from Menlo Ventures found that Anthropic captures 40% of enterprise LLM spend — up from 24%, while OpenAI's share fell to 27%, down from 50%. But the data looked only at API usage (not chatbot sessions or consumer subscriptions).
• Zoom out: Adoption is happening crazy fast: Nearly 47% of businesses paid for AI in January, a new high. In 2023, that number was less than 7%.

The bottom line: "The race isn't zero-sum," says Ramp's Kharazian in a release yesterday. "At least not yet."

Anthropic is donating $20 million to Public First Action, a bipartisan advocacy group focused on AI transparency and safeguards, the company announced today.

Why it matters: AI policy is becoming a campaign flashpoint, and super PACs are raising millions to sway voters.

Context: Public First Action is tied to two super PACs, one Democratic and one Republican, that plan to back candidates who support AI safeguards.

• The priorities include giving the public more visibility into AI companies, opposing preempting state-level AI regulation without a strong federal standard, export controls on AI chips and regulation on high risks like AI-enabled biological weapons.

What they're saying: "At present, there are few organized efforts to help mobilize people and politicians who understand what's at stake in AI development," Anthropic said in a press release.

• "Instead, vast resources have flowed to political organizations that oppose these efforts. Public First Action is working to fill that gap."

The big picture: Anthropic is an outlier in the industry, positioning itself as the poster child for regulating AI, from limiting exports of sensitive technology to proposing an AI transparency framework.

• Anthropic's peers have thrown their weight behind PACs that are focused on innovation and preventing a patchwork of state-level regulation.
• AI super PAC Leading the Future has raised more than $125 million from OpenAI's Greg Brockman, a16Z and others in the industry.

• Anthropic added additional tools to its free plan, including connections with Google workspace apps. (CNET)
• A Democratic New York assembly member running for Congress released a new policy platform to support AI guardrails. (Axios)
• Elon Musk announced a reorganization of xAI after two co-founders announced they were leaving. (Wall Street Journal)
• Meta began construction of a $10 billion data center in Indiana. (Reuters)

Ina here, back from watching Jordan Stolz win the 1,000-meter race in Olympic record time, sitting just a couple rows behind Snoop Dogg.

• While I have plenty of good photos of Stolz and Snoop, and even one of Stolz holding the flag while skating past Snoop, I think this photo of a Canadian skater warming up is my absolute favorite.