Key Highlights

  • Reports published Saturday, October 24, 2026 say OpenAI displaced Anthropic for most of Replit's artificial-intelligence model spending.
  • The reported agreement exceeds USD 100 million annually, with OpenAI reducing price by more than 30% as Anthropic withdrew its own discount.
  • Rippling is among other customers reportedly moving workloads, and some individual accounts have cited reductions reaching 90%.
  • Pricing competition arrives while Anthropic plans an IPO and OpenAI approaches funding of up to USD 30 billion at a USD 1.25 trillion value.

Why Does Replit's Reported Switch Create a Pricing Test for the Fund?

The AIQ fund (NASDAQ:AIQ) owns public businesses that develop or apply artificial intelligence, making model prices relevant across its portfolio. Saturday reports dated October 24, 2026 said coding platform Replit shifted most model expenditure away from Anthropic and toward OpenAI. Annual value reportedly exceeds USD 100 million.

OpenAI is said to have reduced price by more than 30%, while Anthropic withdrew its previous concession. Both model developers remain private. Anthropic may publish an IPO prospectus on Monday, October 26 at the earliest and target a mid-November debut. Its reported third-quarter revenue reached USD 16.8 billion after 45% sequential growth.

Why Are High-Volume Coding Workloads So Sensitive to Price?

Coding software uses large quantities of tokens, the pieces of text processed and produced by a model. Applications such as Replit reach providers through an interface, allowing a change to take weeks instead of years. When quality is similar, that low switching barrier makes price a powerful competitive tool.

When price drops by more than 30%, some economics move from the model supplier toward the application and its customers. Anthropic investors will therefore examine gross margin, or revenue remaining after compute expense, in the prospectus. Volume provides the offset: if lower prices prompt enough extra use, chip and data-centre demand may increase even while revenue per token declines.

Could Wider Price Cutting Challenge AI Valuations and Model Margins?

Reports also name payroll-software provider Rippling among customers moving workloads, with reductions reaching 90% in isolated accounts. If competition extends beyond coding into general enterprise work, profitability across the model layer could compress. Neither provider has published the contract terms, so final agreements may differ from reported details.

Valuation increases the sensitivity. Anthropic discussion has declined from USD 2 trillion toward approximately USD 1.5 trillion, while OpenAI is reportedly raising capital at USD 1.25 trillion. Pricing pressure tests whether either level can be supported. Results remain unknowable in advance, especially while reported contract information lacks direct company confirmation.

What to Watch Next: Will October 26, 28 and 29 Reveal the Broader Effect?

The first possible catalyst is an Anthropic prospectus, potentially on Monday, October 26. Wednesday, October 28, 2026 brings results from Microsoft (NASDAQ:MSFT) and then Alphabet (NASDAQ:GOOGL). Amazon.com (NASDAQ:AMZN) follows on October 29, a Thursday. Their cloud commentary may show whether lower model prices are increasing usage or simply reducing revenue per request.

The Replit report points to more forceful competition while both laboratories seek large public-market valuations. Lower pricing can squeeze model margins yet broaden demand for software, chips and data-centre services held by the fund. The Global X Artificial Intelligence & Technology ETF therefore has exposure to both sides of that trade-off across different holdings. The balance ultimately depends on whether incremental usage offsets lower revenue earned for each token processed. For portfolio companies, the crucial distinction is whether lower unit prices primarily reduce supplier revenue or instead stimulate enough additional usage to expand total spending across the wider artificial-intelligence ecosystem. Reported provisions remain unconfirmed, and nothing here constitutes investment advice.