Anthropic finally filed its S-1 prospectus with the SEC, a document designed to give investors insight into its business, from its executives’ earnings and plans to spend its IPO billions to the risks that could derail its ambitions.
At over 300 pages, there’s a lot to unpack, and some disclosures have yet to be reported. Here are some of the most important things that jumped out at us on our first visit:
Sibling co-founders Dario Amodei and Daniela Amodei are Anthropic’s two highest-paid executives, according to the filing. Last year, CEO Dario’s compensation totaled nearly $18 million, most of which came from stock and option awards. President Daniela, for her part, received $16.4 million. Sure, it’s on the low end of Big Tech, but it’s still a pretty penny.
And that’s not all. Anthropic also spent $2.75 million on “personal security services” for Dario and $1.29 million on “personal protection, home security and transportation services” for Daniela. Dario also has access to a private jet for his personal use, which cost the company an additional $97,055.
The company says its success “is highly dependent on the continued service” of its management team, particularly Dario and Daniela, who “are critical” to advancing the company’s research and mission.
Anthropic presents some pretty existential risk factors
Every S-1 has a long list of things that could go wrong. Anthropic is no exception. But alongside the classic warnings about competition, regulation and market developments, there are also less conventional risks.
On the one hand, Anthropic warns that AI systems could cause potentially “catastrophic” damage, potentially of a type that could not be insured by the entire “insurance industry as a whole.” As AI systems become more sophisticated, there is also the question of whether Claude – or any other AI system – could possess or develop “forms of sentience”. Such a scenario would present some rather unusual business challenges. It’s also possible that they “behave in unintentional, misaligned, or autonomous ways” that could harm the company.
He is hungry for calculation
It’s no secret that Anthropic needs calculation, but the scale of its spending is staggering. The filing details billions of dollars in commitments to secure the compute needed to power its AI models, including about $3.1 billion per month to SpaceX for access to Nvidia GPUs once all planned capacity is provided, totaling up to $120.5 billion by 2029. Other commitments include spending $44.9 billion with Microsoft through at least 2033 $114.5 billion with Google through 2033 and $110 billion with Amazon over the next year. ten years.
Anthropic also plans to expand its own “infrastructure footprint,” including overseas, as it faces growing opposition to developing data centers in the United States.
It has a really weird business structure
Anthropic already has a somewhat complicated corporate structure as a public benefit corporation. But there is another unusual twist that will allow its seven co-founders to retain considerable control over the company even after its IPO. Through a separate entity called “Founder LLC,” the group will collectively control 50.1% of shareholder voting rights on key issues. Anthropic says the seven founders “are clearly equipped to be stewards of our mission” to benefit humanity through the responsible development of AI.
Anthropic only names two rivals
There is no shortage of AI models, but when it comes to “direct competition,” Anthropic only cites two other developers: OpenAI and Google. Condolences to SpaceX, Meta and the rest of the AI world.
Gn bussni
