Business & Economics
Nvidia-OpenAI $100 B Data-Center Deal Stalls, Talks Pivot to Smaller Equity Stake
On 30-31 Jan 2026, internal pushback at Nvidia froze a non-binding September 2025 pact to fund and build 10 GW of compute for OpenAI—slashing the prospective outlay from $100 B to a still-undecided equity infusion of merely “tens of billions.”
Focusing Facts
- The September 2025 memorandum called for Nvidia to finance and guarantee up to $100 B and lease GPUs to OpenAI while constructing at least 10 GW of capacity—roughly the output of ten large nuclear reactors.
- After the freeze, current talks contemplate an equity cheque below $30-40 B, concurrent with OpenAI’s attempt to raise $100 B at a pre-money valuation near $830 B.
- Nvidia separately disclosed a commitment of up to $10 B in rival Anthropic in Nov 2025, signalling hedging against OpenAI’s slowdown.
Context
Capital-heavy tech love affairs often cool once spreadsheets arrive: in 1985, IBM’s plan to pour $1 B into the up-start Microelectronics & Computer Technology Corp collapsed within months when internal bean-counters balked at unclear returns— eerily similar to Nvidia’s jetzige volte-face. The stall exposes two structural currents: (1) the AI arms race is morphing into a raw infrastructure contest where chipmakers, cloud titans and model labs must lock in multi-decade power and fabrication capacity, and (2) hardware suppliers are no longer passive vendors but quasi-investment banks shaping their customers’ balance sheets. Whether this deal revives or not, the moment signals that even trillion-dollar optimism has a cost-of-capital governor; the railroads of the 1870s likewise over-promised track mileage before a financing crunch forced consolidation. On a 100-year arc, the episode will matter only if it marks the point when AI compute ceased to be a limitless credit fountain and began following the boom-bust cadence of every previous General-Purpose Technology build-out—from electrification (1880s) to fiber optics (1990s).
Perspectives
Mainstream business press
Mainstream business press — They interpret the frozen $100-billion NVidia-OpenAI deal as a warning sign that Sam Altman’s sky-high spending plans and the wider generative-AI boom may be outrunning hard financial reality, complicating OpenAI’s hoped-for 2026 IPO. Traffic-hungry financial outlets profit from surfacing boardroom friction and market doubt, so they highlight internal dissent and potential market pullbacks while giving little space to management’s counter-arguments that the talks are merely being resized, not abandoned.
Tech-optimist consumer tech sites
Tech-optimist consumer tech sites — They frame the pause as only a temporary hiccup in an otherwise furious fund-raising race, stressing that Amazon, SoftBank and others are eager to pour tens of billions more into OpenAI and that the company could still hit a $1-trillion valuation at IPO. These outlets court gadget fans and bullish retail investors, so they accentuate eye-popping valuations and ‘next big thing’ momentum, relying on second-hand reports and glossing over the unanswered questions about profitability or governance.
Watchdog-oriented tech criticism media
Watchdog-oriented tech criticism media — They see the stalled mega-deal as just one symptom of an ever-more muscular OpenAI that is willing to intimidate critics and skirt its original altruistic mission in pursuit of scale and profits. By focusing on subpoenas and door-stepping rather than the financing itself, they risk painting every corporate move as sinister, reinforcing their advocacy brand but offering limited context on the commercial stakes.