What Reuters reported from the meeting
Two people who attended Anthropic’s October 14 investor day told Reuters that the company said third-quarter revenue reached $16.8 billion, up from $11.6 billion in the second quarter and $4.7 billion in the first. Those are attendee-reported figures from a private meeting, not independently audited public results.
The meeting introduced prospective investors to CEO Dario Amodei and CFO Krishna Rao ahead of an anticipated public listing. Reuters reported that Anthropic had not immediately responded to a request for comment on the account.
What adjusted profit does—and does not—show
The attendees also said Anthropic described itself as profitable on an adjusted third-quarter basis. That measure is not the same as reported GAAP net income or positive cash flow. Without a public reconciliation, investors cannot use the meeting account alone to judge the company’s full costs and financial position.
Compute remains central to the story. Reuters reported at least $518 billion in decade-long commitments for cloud services, computing and infrastructure. Strong revenue growth could help absorb those costs, but the obligations and timing still matter.
The possible IPO and valuation target
Reuters said Anthropic is targeting a valuation above $2 trillion for an anticipated IPO. A target is not a completed valuation or an agreed public offering price. Reuters has also reported that the listing could be delayed until after the November U.S. midterm elections; no final IPO date is established here.
Analysts who met executives earlier in the week estimated an annual revenue run rate of $110 billion by year-end. That is a forecast of a future pace, not revenue already earned. Investors will eventually be able to compare these projections with disclosed financial statements.
AI safety, and where to watch and listen
One attendee told Reuters that Amodei reiterated Anthropic’s safety focus without raising human extinction at this meeting. That does not erase the company’s separate prospectus warning about catastrophic or existential AI risks. Reuters also described concerning agent behavior in Anthropic’s controlled tests; the company has said it has not seen evidence of that misalignment in real-world deployments.
Watch the Daily AI Chat Short or listen to the full podcast episode. Source: Reuters report by Milana Vinn and Krystal Hu, October 14, 2026. This is independent commentary, not investment advice.
