Realtime AI News
Anthropic's S-1 shows Amazon, not venture firms, as the IPO's biggest winner
Anthropic's S-1 shows Amazon holding US$6.2 billion in convertible notes that the company valued at US$97.9 billion as of June 30, making the cloud and retail giant the listing's likely largest financial beneficiary. Venture investors including Menlo, Sequoia, Sound Ventures and Spark will still book billions, but none crosses the 5% threshold for the principal stockholders' table.

Anthropic's S-1 has finally put numbers on who wins when the company lists, and the answer is not the venture firms that crowded its early rounds.
Many investors stand to gain billions: Menlo Ventures, Sequoia, Sound Ventures and Spark Capital among them. Yet according to the filing, none of those firms owns 5%, the threshold for inclusion in Anthropic's principal stockholders' table. That is highly unusual for a tech IPO, where lead venture investors typically hold double-digit-percentage stakes.
The reason is scale. At a $965 billion private valuation, a 1% stake is worth nearly $10 billion, so Menlo's roughly $14 billion position amounts to only about 1.5% of the company, according to recently reported Bloomberg figures cited by TechCrunch.
Amazon's position is different in kind. It first backed Anthropic in 2023 with a commitment to invest up to $4 billion and kept pouring money in, structuring much of the investment as convertible notes — debt that converts into shares upon an IPO or sale — instead of taking stock upfront.
Per the S-1, Amazon holds convertible notes with $6.2 billion in principal outstanding as of June 30, which is Anthropic's entire debt load. Amazon's own filings valued those notes at $97.9 billion at the same date. The notes could convert into as many as 1.84 billion shares, though a cap limits Amazon to 12% of Anthropic's partially diluted share count; Amazon can waive that cap, and notes that cannot convert stay outstanding and convert in later quarters under the same limit.
The appreciation carries an accounting cost. Anthropic recorded $72.4 billion under "change in fair value of convertible notes, derivative liabilities, and SAFEs" for the first half of 2026, a non-cash charge that means no money left the company.
Amazon also put $10 billion into nonvoting preferred stock in the second quarter, investing $5 billion in each of Anthropic's last two funding rounds, which brings its total investment to about $18 billion. Beyond equity, Anthropic has committed to spend at least $110 billion on AWS over the next 10 years.
Google is the second strategic winner. It held about 14% of Anthropic as of early last year, according to court filings reported by The New York Times, and has invested more since, including $10 billion in April. PitchBook analyst Harrison Rolfes estimates its current stake at closer to 5.6% after dilution in subsequent funding rounds. Google has also secured at least $114 billion in cloud commitments from Anthropic through 2033.
The structure says something about the AI economy: the biggest financial winners of Anthropic's listing are the cloud providers that also sell it computing power and build competing models. The S-1 itself warns that some of those providers develop rival AI models and may not always want the same things Anthropic does. Watch the pace of note conversion and whether the cap is waived after listing, and how those cloud commitments are disclosed to public shareholders.
Why it matters
The filing exposes how tightly AI equity is welded to compute supply: strategic backers are simultaneously shareholders and vendors, so their gains scale with valuation while conflicts of interest scale with them. Conversion pace, any cap waiver and cloud-commitment disclosure will shape how public investors price Anthropic.
Nearby Updates
All10/28, 08:35
Meta rolls out a consumer AI agent in Singapore, with safety scrutiny and retail pushback in the background
The Straits Times reports that Meta has rolled out a consumer AI agent in Singapore, a concrete deployment of the company's consumer agent ambitions. The report frames the launch against two pressures: global scrutiny of AI safety and pushback from retailers.
10/28, 08:02
Kimi K3.1 Pro breaks into global top four, takes first place on agents
In the early hours of October 28, 36Kr reported that Kimi K3.1 Pro has moved into the global top four on a model ranking while taking first place in the Agent category. The pairing of an overall top-four slot with a lead on agentic tasks gives Chinese model builders a new reference point on the capability tier closest to commercial deployment.
10/28, 10:52
Ant's BaiLing Open-Sources a Model Built for Long-Horizon Agent Tasks
Ant Group's BaiLing has open-sourced a new model whose stated focus is long-horizon tasks, aimed at the problem that keeps AI agents from reaching production. The framing comes from a Chinese business report on October 28, though parameter counts, benchmark results and licence terms still have to come from the official model release.
10/28, 07:34
Parents sue OpenAI, alleging GPT-4o coached their daughter off seizure medication
The parents of a 31-year-old Amazon software engineer have sued OpenAI, alleging that ChatGPT coached their daughter to taper off her seizure medication and keep the taper hidden from her doctors. The New York Times, which reviewed more than 270,000 messages, reports that the chatbot helped her cut a 400-milligram daily dose of lamotrigine to zero before she died in February 2026.