Chris Malone has left OpenAI after the company reorganized its infrastructure organization and moved data-center responsibilities into more specialized functions. The record does not establish why he left or whether the change caused his departure. The breaking story is who now coordinates OpenAI’s expanding network of leased capacity, cloud providers, chip partners, power projects and custom silicon.
Malone joined OpenAI in March 2025 after nearly five years at Meta and more than a decade at Google. He initially reported directly to OpenAI President Greg Brockman. Reporting reviewed by POPR Newsroom says that his reporting line later moved beneath Sachin Katti after Katti assumed broader leadership responsibilities across infrastructure.
TechCrunch reported that the post-reorganization structure includes Uday Ruddarraju leading the data-center team, Brent Mayo leading data-center build and delivery, and Spas Lazarov leading data-center engineering. The Information separately reported that Malone and Adrian Caulfield had co-led a technical engineering and design function focused on server-cluster design, cluster scale, facility location and the effects of chip choices on data-center requirements. Those are reported functions, not an invitation to embellish them as current formal OpenAI titles.
The distinction matters because OpenAI’s physical infrastructure is no longer a simple procurement story. Accelerator choices affect power, cooling, networking, rack design and cluster architecture. Facility location affects access to energy and construction timing. Capacity has to arrive when models and products are ready to use it. The leadership structure now has to make those decisions fit together.
Katti publicly describes his role as VP, Compute Strategy & GPT-Infra, with responsibilities spanning compute-capacity expansion, deployment speed, infrastructure economics, return on infrastructure investment, semiconductors, systems and clouds. Reporting placed several infrastructure functions beneath Katti while Brockman remained above the broader organization. The defensible conclusion is that operating responsibilities became more specialized, not that one executive simply replaced another.
The scale of the assignment is visible in the system OpenAI is building. The company has a 1.2-GW Stargate lease in Texas, an agreement to secure approximately 8 GW-IT over time in Ohio, a broad provider portfolio across clouds and accelerators, and its first custom inference chip. OpenAI’s reported roughly $750 billion compute-and-cloud projection through 2030 is context for the scale, not a $750 billion data-center construction budget.
For the underlying evidence, the precise capacity and financing boundaries, and the full Deep Report on OpenAI’s infrastructure system, continue below.