OpenAI Admits Its Apple Siri Deal Flopped, Court Files Show logo

OpenAI Admits Its Apple Siri Deal Flopped, Court Files Show

Court filings in OpenAI's antitrust fight with xAI reveal OpenAI's own view that its ChatGPT integration into Apple's Siri was 'persistently underperforming,' undercutting xAI's claim the deal gave OpenAI an unfair advantage.

By the Numbers

Dec 2024
Integration launched
Jan 2025
Forecast cut
'Dramatically underperforming'
OpenAI's own description
OpenAI vs. xAI
Lawsuit
TC
By the Markets Desk
Edited by Trace Cohen · Early-stage VC & angel · Founder, New York Venture Partners
2 min read
ShareXLinkedInEmail

THE RUNDOWN

1

OpenAI is using its own admission of failure as a legal defense -- arguing that if the Apple-Siri integration barely drove usage, xAI's antitrust claim that the deal harmed competition can't hold up.

2

The filings show OpenAI cut its forecast for incremental ChatGPT users from the Apple deal as early as January 2025, within weeks of the integration going live in December 2024.

3

A multi-step opt-in process inside Siri is cited as a likely cause of the weak adoption -- a design friction point that undercuts the assumption that any default-style AI placement guarantees usage.

4

The disclosure lands as Apple and OpenAI are separately locked in an escalating trade-secrets lawsuit, meaning two unrelated legal fights are now both surfacing unflattering internal details about the companies' AI partnership.

TC

The VC Read · Trace's Take

Trace Cohen

OpenAI volunteering that its own Apple deal flopped is the tell that litigation strategy can force more honest disclosure than any earnings call ever would -- nobody puts 'dramatically underperforming' in a press release. Diligence item for anyone evaluating a platform-distribution partnership pitch: ask specifically whether the integration requires active opt-in or ships as a default, because this filing is hard evidence that opt-in friction alone can gut adoption regardless of how prominent the placement looks on a slide.

Analysis

Court documents filed in OpenAI's ongoing antitrust dispute with Elon Musk's xAI reveal that OpenAI internally described its ChatGPT integration into Apple's Siri as "persistently underperforming," according to MacRumors and TechRadar.

What The Filings Actually Say

OpenAI's own internal analysis, cited in the filings, states plainly that "Apple Intelligence saw minimal usage." By January 2025 -- barely a month after Apple added ChatGPT to Siri in December 2024 -- OpenAI had already cut its forecast for the number of incremental logged-in weekly active users it expected the partnership to generate. The filings describe the integration's performance as "dramatically underperforming" relative to OpenAI's own initial expectations for the deal.

“The filings describe the integration's performance as "dramatically underperforming" relative to OpenAI's own initial expectations for the deal.”

Why OpenAI Is Volunteering This Admission

The disclosures are not a leak -- they're OpenAI's own legal strategy. xAI has argued that Apple's decision to route certain Siri queries to ChatGPT gave OpenAI an unfair, anticompetitive advantage over rival AI assistants, xAI's Grok included. OpenAI's counter is direct: if the integration drove only minimal usage, there's no meaningful competitive advantage for xAI to have been harmed by, and no exclusivity claim can produce damages against an integration this quiet. Admitting the partnership flopped is, in this specific legal context, a defense rather than an embarrassment.

Why It Flopped: The Opt-In Friction

Apple's implementation required users to go through a multi-step opt-in process before Siri would hand a query to ChatGPT rather than answering with Apple's own more limited tools -- friction that likely suppressed adoption well below what a default, no-opt-in integration would have produced. That detail matters beyond this lawsuit: it's evidence that placement alone doesn't guarantee usage in AI products, and that consumers largely didn't go out of their way to enable a third-party AI layered onto a system assistant they already had.

The Wider Apple-OpenAI Backdrop

This disclosure surfaces the same week Apple and OpenAI are separately escalating an unrelated trade-secrets lawsuit, in which Apple alleges a former engineer carried confidential material to OpenAI. Two independent legal fights are now both producing unflattering internal detail about the Apple-OpenAI relationship in the same news cycle -- a reminder that a splashy platform partnership announcement and the messy commercial reality behind it can diverge sharply, and that the gap sometimes only becomes public once litigation forces disclosure.

For competitors and challengers in the AI-assistant space -- Google's Gemini, Anthropic's Claude, and xAI's own Grok -- the practical lesson is that owning distribution through a dominant platform doesn't automatically translate into engagement if the integration adds friction a user has to actively opt into.

ShareXLinkedInEmail

Key Sources

2 sources

Reported by MacRumors · Analysis by Value Add Pulse.

← Back to Pulse

THE WIRE in your inbox— Tech, startup & VC news with Trace's take. Free, no spam.