OpenAI Loses Its Head of Data Centers Amid Infrastructure Pivot and Executive Exodus
Chris Malone, OpenAI's head of data centers, left the company in late August 2026, as first reported by the Wall Street Journal and confirmed by TechCrunch. His departure follows an infrastructure reorganization and brings OpenAI's 2026 executive exit count to 13, per a tally by Business Insider.
Malone joined OpenAI in March 2025, arriving shortly after the launch of the Stargate Project, a $500 billion U.S. data center initiative championed by the Trump administration. OpenAI is a key partner in Stargate alongside Oracle, Nvidia, SoftBank, and Microsoft. Before OpenAI, Malone spent nearly five years at Meta and more than a decade at Google.
His tenure was brief. In March 2026, OpenAI reorganized its infrastructure leadership, shifting its data center strategy and prompting the departure of three senior executives behind the Stargate initiative, according to The Information. That reorganization also marked a strategic pivot: OpenAI moved away from building its own data centers toward renting compute capacity — the processing power needed to train and run AI models — from third-party providers. Malone and Peter Hoeschele, who had co-led OpenAI's data center groups and both reported to President Greg Brockman, were affected by the restructuring, per Data Center Dynamics.
In the most recent reorganization, Malone stopped reporting directly to Brockman and began reporting to Vice President Sachin Katti, who took over leadership of the infrastructure group. OpenAI told TechCrunch it had reorganized its infrastructure organization "to support the scale and pace of its work." Katti joined OpenAI from Intel in late 2025, where he had been an Intel AI leader; his departure followed Intel's retooling of its data center AI strategy under co-CEO Lip-Bu Tan after the semiconductor giant fell short of expectations, CRN reported.
Several other executives now oversee pieces of OpenAI's data center strategy. Uday Ruddarraju leads the data center team, Brent Mayo leads the data center build and delivery program, and Spas Lazarov leads all data center engineering.
Malone's exit is the latest in a string of high-profile departures stretching across the summer. About a month ago, Fidji Simo, OpenAI's de facto second-in-command and product and business chief reporting directly to CEO Sam Altman, stepped down to recover from a chronic illness and remains with the company in an advisory role, TechCrunch reported. In July, OpenAI also lost its head of ethics, Chloé Bakalar. Two weeks before Malone's departure was reported, OpenAI replaced its chief revenue officer Denise Dresser, who had been with the company only about eight months. Two days before Dresser's announced departure, Brad Lightcap, one of OpenAI's longest-serving executives and its chief operating officer, left to start something new, TechCrunch reported.
The broader context here is the intersection of two pressures: an infrastructure strategy in active flux and a leadership bench losing senior operators at a rate that invites scrutiny. OpenAI's pivot from building owned data centers to renting compute capacity, reported by The Information in April, is a meaningful shift for a company whose Stargate partnership was framed around massive domestic infrastructure buildout. That the executive who led data center operations has now departed, after his reporting line changed and his portfolio was redistributed across at least three other leaders, raises a practical question about execution continuity at a time when compute procurement is the defining constraint for frontier AI development.
Thirteen executive departures in a single year is a number that, taken alone, could reflect normal churn at a fast-scaling organization. The composition of the exits, however, is harder to dismiss: the chief operating officer, the product and business chief, the chief revenue officer, the head of ethics, and now the head of data centers have all departed within months of each other. These span commercial operations, product, governance, and infrastructure. When departures cluster across multiple functional areas rather than within a single team, it tends to indicate something systemic rather than isolated dissatisfaction.
What this means for OpenAI's competitive position depends on whether the reorganized infrastructure group under Katti can maintain the pace of compute scaling that Stargate was designed to enable. The partners are still in place, the capital commitment is stated, and OpenAI has not signaled any retreat from Stargate itself. But the operational expertise required to stand up hyperscale data center capacity — the kind Malone brought from Google and Meta — does not transfer overnight. The company now has at least three executives sharing the remit Malone once held, which could dilute decision-making velocity or, alternatively, distribute operational risk. Which outcome prevails is not yet visible from the outside.
