EXPLORINEER
en

JP — ITmedia

Anthropic's IPO Prospectus Warns of AI "Existential Risk to Humanity" and Says Claude Writes 90% of Its Code

AI2 min read

Anthropic, the AI company that makes Claude, has officially applied for an IPO (sale of shares to the public) in the United States. Its filing says Claude now writes most of the company's code and warns that AI could pose a risk to humanity.

What happened

On October 26 (US time), Anthropic filed a registration statement called Form S-1 with the US Securities and Exchange Commission (SEC). This is the main step before a company can sell shares to the public. The filing is over 300 pages long. It includes money numbers like sales and losses, and also details about how the company uses AI and what risks it sees.

Key numbers and facts

  • Over 90% of the code added or changed in September was written by Claude, not by human engineers.
  • Claude's success rate on hard development tasks with no clear specs rose by about 50 points in 9 months, to about 90% as of September (based on internal tests).
  • The amount of code one engineer writes per day, averaged over July to September, is more than 45 times the 2024 average.
  • Claude usage, priced at general-user rates, grew more than 250 times from January to September. In April, its monthly value passed the company's monthly pay costs, excluding stock-based pay.
  • Employees grew about 2.7 times in one year, to about 5,300 by the end of September 2026.

Background for newcomers

Anthropic says its engineers now act more like supervisors. Instead of typing code line by line, they give Claude instructions and check the results. The company says teams outside software, such as operations, sales, and research, also use several AI agents (AI programs that carry out tasks on their own) to automate daily work. The filing also has over 90 pages about risks, and it mentions the risk that AI could threaten humanity's survival.

Why it matters

This filing gives a rare look inside a fast-growing AI company. It shows AI doing a large share of real work at a software company, and it shows how fast the business is growing. The warning about risks from AI is also notable, because it comes from a company that builds AI itself. Investors and the public can now read these claims directly in an official document.

What to watch

  • Whether the SEC approves the filing and when Anthropic actually lists its shares.
  • How investors react to the claims about AI-written code and fast growth.
  • Whether other AI companies, such as OpenAI, follow with their own IPO filings.

Glossary

IPO
Initial public offering: when a private company first sells shares to the public.
Form S-1
A registration form a company files with the SEC before selling shares to the public.
SEC
The US Securities and Exchange Commission, the government agency that oversees stock markets and company filings.
Claude
An AI model made by Anthropic that can write and check computer code and do other work.

Ad

Add 150 words. Learn 10 a day.

Dicter word books send only the words you learn to your review notes.

More in this topic

AI전자신문

Adobe Expands AI Partnership with London Heathrow to Deliver Personalized Passenger Services

Adobe is expanding its partnership with Heathrow Airport to use AI-driven personalization and digital customer experience tools for more than 84.5 million annual passengers. Heathrow will adopt Adobe CX Enterprise on its official website, which receives over 50 million visits a year, and the two companies will work to make Heathrow's official content easier to find on major AI platforms such as ChatGPT, Gemini, Microsoft Copilot, and Perplexity. The companies also plan to track how Heathrow content is cited and surfaced in AI search and improve areas with poor information accessibility.

AIThe Register

A whopping 3% of boardroom execs spy productivity and costs savings from AI

Gartner researchers say three-quarters of corporate board members expect AI to boost productivity and cut costs, but only three percent of organizations have hit those targets. The analyst firm also reported that 38 percent of companies are over budget on AI spending, and 85 percent plan to spend even more next year. Gartner advises focusing AI on the work that matters most, with HR helping leaders agree on priorities.

AI전자신문

Seoul AI Hub to support AI startups at each growth stage with venture capital firms

Seoul AI Hub, the city's AI startup support center, has signed investment MOUs with nine venture capital firms operating Seoul and Seocho-gu funds to expand investment opportunities for AI startups. The partners plan joint scouting, investment review, accelerating, and growth-stage support, and the hub has also announced plans to set up a new "Baby Unicorn Center" with the Korea Angel Investment Association.

AITechCrunch

Meta AI director Rohit Patel leaves to take on Microsoft Office and Google Workspace

Rohit Patel, a former director at Meta Superintelligence Labs who spent eight years developing AI at Meta, has left the company to launch Ndoc, an AI-native alternative to Microsoft Office and Google Workspace. Ndoc combines presentations, analyses, and datasets with Python scripts in one workspace that colleagues can use even without an account. Patel's departure follows several other high-profile exits from Meta, and the startup is a three-person team with no funding.