• OpenAI plans to keep investing in AI data centers despite political resistance, CFO Sarah Friar tells CNBC (NASDAQ:VSNT).
  • The company unveiled new community commitments on November 5, including paying full electricity costs and avoiding NDAs with public officials.
  • Voters rejected or restricted data-center projects in 40 of 41 local ballot measures across 10 states, CNBC reports.

OpenAI will press ahead with its massive data-center expansion even after Tuesday’s election results dealt a blow to the company’s infrastructure ambitions, Chief Financial Officer Sarah Friar told CNBC. The commitment to continue building comes as the artificial-intelligence giant faces growing resistance at the local level over energy costs, water usage and transparency.

Friar’s comments, reported by CNBC, signal that OpenAI is not backing away from its roughly $600 billion computing spending plan through 2030. But the company is also moving to address the political headwinds that have already stalled dozens of projects across the country.

On November 5, OpenAI published a set of commitments for future data centers it owns and operates, pledging to pay their full electricity costs, protect other ratepayers, seek community input, publish construction information and avoid nondisclosure agreements with public officials—subject to exceptions for proprietary information. The company also urged policymakers to turn similar protections into law. OpenAI said the policy work began in August, so the package was not drafted solely in response to the election.

The electoral backlash was pronounced. CNBC reported that voters supported restrictions or rejected pro-development proposals in 40 of 41 local data-center ballot measures across 10 states. Incoming Democratic governors in Texas, Georgia and Ohio campaigned on pausing permits or tightening construction rules, according to Reuters (NASDAQ:TRI). Those are election outcomes and policy commitments—not evidence that every proposed restriction is already law, but they underscore the shifting political landscape.

The stakes are high. OpenAI’s buildout supports spending on construction, chips, electrical equipment and power infrastructure, and the company argues it creates lasting local employment. Chris Lehane, OpenAI’s global-affairs chief, recently cited an expected 2,500 long-term jobs at a southern Ohio facility, averaging $80,000 annually. Those are company projections, not independently verified figures.

At the same time, the economics of the buildout are coming under closer scrutiny. OpenAI told investors its September annualized revenue was nearly $50 billion, according to Reuters—a revenue run rate, not revenue actually earned over a full year. Reported second-quarter 2026 revenue was $6.7 billion, up about 18% from the first quarter, according to reporting citing The Wall Street Journal (NASDAQ:NWSA). But secondary reporting also puts the second-quarter operating loss at $12.3 billion, including stock-based compensation. OpenAI does not publish regular public-company financial statements, so these figures warrant more caution than filed earnings.

Financing conditions appear to be tightening. On November 2, reporting said OpenAI reduced its latest fundraising goal from $30 billion to as much as $20 billion, with a proposed post-money valuation around $1.25 trillion rather than $1.4 trillion. Those were reported negotiations, not a confirmed final closing. The reduced target makes the tension between infrastructure commitments and revenue growth particularly relevant.

Financial exposure extends beyond OpenAI. CNBC reported on November 6 that ING (NYSE:ING) was seeking buyers for approximately $500 million of data-center loans, including an interest associated with a Stargate project. Loan sales do not by themselves establish distress, but they show how AI infrastructure risk is being distributed through financial markets.

The industry is adjusting broadly. Microsoft (NASDAQ:MSFT), Google (NASDAQ:GOOGL), Meta (NASDAQ:META) and Anthropic have outlined comparable community or ratepayer commitments, shifting competition beyond model quality and computing capacity toward securing power, financing, permits and public acceptance.

The political issue is the contrast between federal support for AI infrastructure and state or local resistance. Reuters reported that OpenAI has cultivated ties with the Trump administration, while its latest outreach also targets Democrats and local communities. National Economic Council Director Kevin Hassett said November 6 that he expected data-center growth to continue despite the election results.

Internationally, fundraising and compute access are becoming intertwined. Reporting on the latest financing discussions identified UAE-backed MGX among expected leading investors and suggested the arrangement could provide additional computing capacity in the UAE. That creates a cross-border capital-and-infrastructure connection, though it does not establish that OpenAI is relocating U.S. projects abroad.

A major limitation of OpenAI’s new commitments is coverage: Oracle (NYSE:ORCL) and other partners own much of the infrastructure OpenAI uses. The promises specifically covering OpenAI-owned-and-operated campuses should not automatically be assumed to cover all partner facilities. That distinction leaves questions about how far the protections extend.

The broader backlash has been building. CNBC cited research identifying at least 46 AI data-center projects across 20 states—representing more than $170 billion of announced investment—that were blocked, withdrawn or stalled following community opposition between January 2024 and May 2026. Those categories are not synonymous with permanent cancellation, but they illustrate the depth of local resistance.

Analyst Dan Ives argued on November 6 that much of the next 18 months’ pipeline would proceed, while acknowledging that political resistance would slow the wider buildout. Conversely, market commentary cited by CNBC warns that construction moratoriums could compound already disappointing financial returns. Neither outlook is settled.

The headline therefore signals commitment, not immunity: OpenAI can maintain its investment strategy while still facing slower approvals, higher community obligations and tighter capital discipline. The key developments to watch are whether campaign promises become enforceable permitting restrictions, whether OpenAI’s protections extend to partner-owned capacity, whether fundraising closes on workable terms, and whether enterprise revenue growth improves the economics of its compute commitments.

OpenAI did not respond to a request for comment on Friar’s CNBC remarks. A spokesperson pointed to the company’s November 5 policy announcement.