Startups like Perplexity and Lovable survive only on VC subsidies, so a large share of Anthropic's and OpenAI's API revenue depends on those subsidies.

Does the evidence support the cause?

Outcome: Unresolved

We checked, and the evidence so far doesn’t settle it.

Judged on a plain reading. They named no measure, so we used the reading a typical informed reader would take. It is spelled out under “How we judged it”.

Evidence up to 2026-10-28. Checked 2026-10-31.

Why

The claim has two parts. First, firms like Perplexity and Lovable stay alive only because of venture money. Second, a large share of both Anthropic's and OpenAI's API revenue therefore depends on that money. Both parts must hold for both labs. Support would need evidence that these firms rely on venture money to operate, and that a large share of each lab's API revenue comes from such customers. Contradiction would need evidence that such firms are self-sustaining, or that only a small share of either lab's API revenue comes from them. The packet has no profit, cash-flow or funding figures for Perplexity or Lovable. It has no breakdown of either lab's API revenue by customer funding source. The one item, e1 (OfficeChai, available 2026-09-29), is context only. It reports that Anthropic's annualized revenue run-rate rose past $47 billion from about $9 billion at the end of 2025. It says much of that growth came from Anthropic's own products, such as Claude Code, and from large enterprise customers. It gives no figure for the share of either lab's API revenue that comes from VC-funded customers. The evidence does not settle either part of the claim.

Caveats

• The packet has no 2026 profit, cash-flow or funding data for Perplexity or Lovable, so it does not show whether they lose money or depend on venture money.
• e1 describes Anthropic's total revenue run-rate and growth sources, not a breakdown of API revenue by customer funding source.
• The packet has no breakdown of OpenAI's API revenue by customer type or funding source.
• Whether these firms would survive without venture money cannot be observed directly.
• 'Large chunk' has no stated threshold.

Evidence (1 item)

• Context only: OfficeChai, 2026

  much of the recent growth has come from its own products such as Claude Code, along with a rapidly widening base of large enterprise customers.

  Why it matters: Suggests much of Anthropic's 2026 growth came from its own products and large enterprises rather than VC-funded API startups. But it gives no figure for the VC-dependent share of API revenue.

  Where

  section The Case That Concentration Is Fading, paragraph 1

  Publisher

  OfficeChai

  Kind of source

  News report

  About events on

  2026

  Published

  2026-09-29