OpenAI Details Full-Stack AI Strategy to Scale at Lower Cost

OpenAI is framing its compute strategy as an integrated stack spanning data centers, chips, frontier models, its developer platform, consumer and enterprise products, and AI-native devices. Chief Financial Officer Sarah Friar said co-design across those layers can improve performance and energy efficiency while lowering the cost of serving AI. The strategy matters because cheaper, more dependable inference could broaden adoption, while higher usage and revenue would fund further research, infrastructure and safety investment.

On Aug. 25, 2026, OpenAI released the first measured results for Jalapeño, its first custom inference chip. On the public InferenceX benchmark running GPT‑OSS 120B, Jalapeño delivered higher peak throughput per kilowatt and lower token latency than the commercial systems tested, OpenAI said, while also performing strongly on DeepSeek R1 and Kimi K2. The company also said GPT‑5.6 Sol set a new high on the Artificial Analysis Coding Agent Index while using 54% fewer output tokens than another leading model. It disclosed no investment figure or overall cost-reduction percentage.