OpenAI’s Head of Data Centers Exits as Executive Churn Hits 13 in 2026

Chris Malone, OpenAI's head of data centers, left the company last week, the Wall Street Journal reported on August 25. Malone joined OpenAI in March 2025, shortly after the announcement of the Stargate Project, a $500 million U.S. data center initiative. His departure brings the total count of senior executive exits at OpenAI in 2026 to 13, according to Business Insider, and lands at an awkward moment as the company prepares for an IPO now expected in 2027.

What happened

Detail	Fact
Executive	Chris Malone, head of data centers
Date of departure	Week of August 18, 2026
Tenure at OpenAI	Joined March 2025 (approx. 17 months)
Prior employers	Nearly 5 years at Meta; 10+ years at Google
Total 2026 exec departures	13 (per Business Insider)
Related initiative	Stargate Project, a $500 million U.S. data center program
OpenAI IPO timeline	Pushed from 2026 to 2027

OpenAI confirmed the departure to TechCrunch with a statement saying it had “recently reorganized” its “infrastructure organization to support the scale and pace of our work.” Before leaving, Malone stopped reporting to OpenAI President Greg Brockman and began reporting to VP Sachin Katti, who now leads the infrastructure group.

Three executives are now said to be sharing responsibility for data center strategy: Uday Ruddarraju (data center team lead), Brent Mayo (build and delivery program), and Spas Lazarov (data center engineering), who has a background in both the data center and energy sectors.

Why a data center departure matters right now

AI infrastructure buildout has become one of the highest-stakes functions at any major AI lab. OpenAI is a key partner in the Stargate Project alongside Oracle, Nvidia, SoftBank, and Microsoft. Losing the executive who oversaw that work, with no obvious single successor, creates a coordination risk at exactly the wrong time.

Malone’s exit is also not an isolated event. The pace of departures has accelerated sharply. In the past month alone, several senior leaders have walked out:

• Brad Lightcap, one of OpenAI’s longest-serving employees and its COO, said he is “starting something new” but has not shared details.
• Denise Dresser, the chief revenue officer, was replaced after approximately eight months in the role.
• Fidji Simo, who served as product and business chief and reported directly to CEO Sam Altman, stepped down citing a chronic illness. She stays on in an advisory role.
• Chloé Bakalar, the head of ethics, left in July.
• OpenAI’s preparedness team, which assessed catastrophic risks from AI models, was reportedly disbanded last week.
• Kate Rouch, chief marketing officer, left in April, also reportedly for health reasons.
• Bill Peebles, who led the Sora image generator project, left after that product was shut down.

What OpenAI says about the churn

Co-founder Greg Brockman has pushed back on the narrative, saying the intense “spotlight” on OpenAI means “every departure gets scrutinized in a way that it doesn’t otherwise.” The company’s remaining leadership has broadly downplayed the exits as normal turnover at a fast-moving organization.

That framing is harder to sustain when the departures include the COO, the CRO, the product and business chief, the head of ethics, and now the data center lead, all within the same calendar year. The disbanding of the preparedness team, a unit focused on assessing whether OpenAI’s own models could cause catastrophic harm, is a separate concern that intersects with the broader safety and governance questions regulators and investors are watching.

Our take

Thirteen senior exits in eight months at a company approaching an IPO is not normal churn. Some departures have clear explanations (health, project shutdowns), but others, like Lightcap and Malone, remain vague. The pattern matters because OpenAI is simultaneously asking investors to believe it can execute on a $500 million data center program, maintain AI safety oversight, and justify a valuation that critics already call inflated. Losing the person who ran data center strategy, with no single replacement named, weakens that story on the infrastructure side.

For businesses evaluating OpenAI’s products as long-term dependencies, this is worth watching but not a reason to change tools today. What it does underscore is the value of building with multiple providers rather than betting entirely on one. If you are thinking through your own AI integration strategy, vendor stability is a real factor alongside capability.

We have covered the broader pattern of enterprise hesitation around AI adoption. Leadership instability at the largest AI lab adds one more reason enterprises move cautiously.

Source: TechCrunch · AI