OpenAI · 2026-04-13 · rumor
RUMOR — Leaked OpenAI Memo Reveals 'Spud' Model Plans and Claims Anthropic Is Overstating Revenue by $8B
OpenAI CRO Denise Dresser's internal Q2 memo leaked via The Verge: reveals 'Spud' model codename (confirmed as GPT-5.5, shipped 10 days later), agent platform 'Frontier', deployment service 'DeployCo', and claims Anthropic overstates revenue by $8B.
A leaked strategy memo exposed OpenAI's next model codename, new platform ambitions, and an unusually aggressive attack on Anthropic's revenue figures.
What is it?
An internal Q2 strategy memo from Denise Dresser, OpenAI's Chief New Revenue Officer, was obtained by The Verge and reported on April 13, 2026. It describes OpenAI's transition from product vendor to 'operating infrastructure.'
How does it work?
The memo lays out five strategic priorities including a model codenamed 'Spud' (described as 'an important step in the intelligence foundation for the next generation of work'), an agent platform called 'Frontier', a deployment service called 'DeployCo', and an Amazon partnership for production-ready execution. It also alleges that Anthropic overstates its ~$30B revenue run rate by approximately $8B through 'grossing up' revenue-sharing agreements with Google and Amazon.
Why does it matter?
The memo was confirmed credible — GPT-5.5 launched April 23, ten days later, matching the Spud description. The Anthropic revenue allegations drew significant industry attention, with Anthropic declining to comment. Source: The Verge (obtained leaked document), reported April 13, 2026. Confidence: High — confirmed by GPT-5.5 launch.