OpenAI pulls out of a second Stargate data center deal

In Norway, OpenAI had been in talks with neocloud provider Nscale, but pulled out of those talks and the data center was instead leased by sometime-OpenAI partner Microsoft, according to sources involved in the discussions. Nscale and Microsoft on Wednesday confirmed the Norway deal, with a source familiar with the negotiations saying that Microsoft will rent the facility in Narvik, Norway, and will then provide compute power to OpenAI through an unspecified agreement. OpenAI did not respond to a request for comments. The UK data center pullout also seemed to reflect some financial belt tightening. Jeremy Roberts, senior director at Info-Tech Research Group, said the move makes sense from an accounting perspective. “OpenAI is embattled on several fronts. Anthropic has been doing very well in the enterprise, and OpenAI’s cash burn might be a problem if it wants to go public at an astronomical 800 billion+ valuation. This is especially true with higher energy prices due to geopolitics, and the public and regulators increasingly skeptical of AI companies, especially outside of the United States,” Roberts said. “I see these moves as OpenAI tightening its belt a bit and being more deliberate about spending as it moves past the interesting tech demo stage of its existence and is expected to provide a real return for investors.” He added, “I expect it’s a symptom of a broader problem, which is that OpenAI has thrown some good money after bad in bets that didn’t work out, like the Sora platform it just shut down, and it’s under increasing pressure to translate its first-mover advantage into real upside for its investors. Spending operational money instead of capital money might give it some flexibility in the short term, and perhaps that’s what this is about.” All in all, he noted, “on a scale of business-ending event to nothingburger, I would put it somewhere in the middle, maybe a little closer to nothingburger.” Acceligence CIO Yuri Goryunov agreed with Roberts, and said, “OpenAI has a problem with commercialization and runaway operating costs, for sure. They are trying to rightsize their commitments and make sure that they deliver on their core products before they run out of money.” Goryunov described OpenAI’s arrangement with Microsoft in Norway as “prudent financial engineering” that allows it to access the data center resources without having to tie up too much capital. “It’s financial discipline. OpenAI [executives] are starting to behave like grownups.” Forrester senior analyst Alvin Nguyen echoed those thoughts. “Microsoft picking up the capacity [in Norway] makes sense. It is a quick way to gain additional capacity for in-demand services without a heavy investment [given that] they are renting the capacity, not building out new data centers,” Nguyen said. “For Nscale, this shows that the demand for AI infrastructure remains high, regardless of OpenAI pulling away from these projects. Hyperscalers are willing to absorb the GPU tranches to preserve Nscale’s pricing power and financing terms.” Nguyen also saw the move as one of financial discipline at OpenAI. “[They are] being more selective about where they deploy, and seem to be moving from splashy deployments to growing incrementally and intentionally,” he said. Independent technology analyst Carmi Levy said he also sees OpenAI’s abandonment of these two data center deals as evidence of “the wide gulf between promising to spend on data centers and actually seeing those projects through to completion. OpenAI isn’t the only AI player following the build-it-big strategy, of course, with all major vendors falling all over themselves to announce ever larger projects in what has become a race for long-term AI infrastructure dominance.” However, he noted, data center capacity means nothing without equivalent power generation. “Local and regional regulatory pushback is strengthening, often turning what initially promised to be straightforward data center builds into risky, complex projects, sometimes with no realistic end date,” Levy said. “Supply chain challenges arising from unrest in the Middle East and tariff worries everywhere else don’t help matters, either. As individual projects turn red amid this increasingly turbulent environment, OpenAI is making the logical decision to cut its losses and find another way to scale up.” And although the deal with Microsoft to lease back data center space at the Norway facility may temporarily pretty up OpenAI’s financial books, he pointed out that it might cost OpenAI down the road if Microsoft continues to become more of a competitor than a partner. “Allowing partners like Microsoft to bear the brunt of data center construction and management headaches is a pragmatic change in direction, [but] it leaves OpenAI less in control of that same compute capacity in the long-term,” Levy said. “It’s a lot like IT decision makers deciding to leverage existing cloud services platforms for compute capacity instead of building the infrastructure themselves.” It avoids headaches around capex, he noted, as well as removing the distractions of managing infrastructure instead of focusing on more salient AI-related core competencies, and it also allows for more timely and cost effective management of capacity-related investment as demand grows at often unpredictable rates. “But,” he said, “it means less independence, as well, as whoever built and owns the infrastructure gets to make the rules.” This article originally appeared on NetworkWorld.