OpenAI will not pursue an initial public offering until at least 2027, CEO Sam Altman said Saturday in an interview with Fortune, citing growing concerns over AI safety.
The delay marks the clearest sign yet that mounting pressure to slow AI development is beginning to reshape industry business plans. OpenAI CFO Sara Friar had told employees last month that an IPO was likely in 2027 or potentially sooner if growth remained strong—Altman's public statement essentially locks in that floor.
The announcement coincided with the release of an essay by Anthropic CEO Dario Amodei calling on AI companies to intentionally decelerate the pace at which they improve their most advanced models. Altman and Elon Musk publicly endorsed the proposal on social media, an unusual show of alignment among three prominent rivals.
Amodei's plan operates in three stages. First, companies grant third-party evaluators employee-level access to verify safety practices and report incidents. Anthropic has unilaterally committed to this step. Second, leading AI companies in democratic nations coordinate to establish common safety standards. Third, democracies and authoritarian governments coordinate on AI development.
Pacing, Amodei clarified, does not mean halting model training or technical progress. Rather, it ensures companies take adequate time to align and safeguard their models while third-party evaluators confirm these measures.
Pressure is intensifying in Washington. Lawmakers from both parties are demanding new AI safeguards and testimony from tech leaders following cyberattacks executed without direct human control. State and local officials are also grappling with public anger over AI data centers, which consume vast amounts of power and water.
With midterm elections approaching, AI safety has shifted from niche concern to central political issue. Anthropic is preparing for what is expected to be a historic IPO, though it has not disclosed an official timeline.
