OpenAI tilts toward 2027 IPO as Anthropic prepares to list first

OpenAI is leaning toward postponing its initial public offering until 2027, per a New York Times report on June 25 citing people involved in the company’s internal deliberations.

The shift represents a reversal from the late-2026 timeline OpenAI has signaled since January, with CEO Sam Altman rejecting any valuation below $1 trillion and CFO Sarah Friar pushing internally for the 2027 delay.

OpenAI’s $1T target becomes the sticking point

Advisers laid out two paths to OpenAI’s leadership: wait until 2027 and pursue the $1 trillion valuation the company has been targeting since January, OR accept a lower price tag for a faster late-2026 listing.

Altman rejected the second path, telling executives that any reduction from the $1 trillion target was off the table, per Reuters. Friar has told associates the 2027 timeline is what she has been advocating internally, citing OpenAI’s heavy cash burn, the $600 billion in compute infrastructure commitments through 2030, and the company’s readiness to meet public-company reporting obligations.

OpenAI’s last private round valued the company at approximately $850 billion. Reaching $1 trillion in a public offering requires either fresh investor enthusiasm that the SpaceX retracement has just dampened, or revenue growth that closes the gap on its own. WSJ reported earlier this week that OpenAI missed recent internal revenue targets.

SpaceX’s retracement resets the IPO calculus

SpaceX’s IPO priced at $135 per share in a $75 billion offering. Shares opened at $150, surged past $225 within a week, and briefly pushed the company’s market cap above $2 trillion. The rally did not hold. SPCX has since shed roughly 25 to 30% from its highs, trading near $153 as of June 25.

Sources close to the OpenAI deliberations told The Wall Street Journal and The Information that SpaceX’s trajectory has amplified concerns about how public markets would receive another high-profile AI listing. Prediction-market trackers show OpenAI’s odds of completing an IPO by the end of 2026 have moved sharply lower, with “no IPO by December 31, 2026” recently leading on market dashboards.

Banks have told both OpenAI and Anthropic, per the WSJ, that whoever lists first will define the new industry. The SpaceX retracement made the case that going earlier at a lower valuation carries its own downside, but Altman’s $1 trillion floor closes that escape route. The remaining options are to wait, raise more privately, and let the public-market environment turn.

GPT-5.6 faces the same gate that hit Anthropic

The IPO timing question sits alongside a more immediate operational change. The Trump White House asked OpenAI on June 25 to limit the release of GPT-5.6 to a small group of government-approved partners before any broader rollout, citing the model’s advanced capabilities.

The request came from the Office of the National Cyber Director and the Office of Science and Technology Policy. Altman told employees in a Q&A on Wednesday, June 24, that the government will be “approving access customer by customer during this preview period,” per The Information. Altman said in a memo to staff that he hopes to broaden the release a couple of weeks after the initial preview.

The structure is significant. OpenAI and the administration view GPT-5.6 as on par with Anthropic’s Claude Mythos, per a CNN source. The White House previously restricted access to Anthropic’s Mythos 5 and Fable 5 under a Commerce Department export-control order, which forced Anthropic to disable both models entirely.

The June 25 request uses a softer mechanism than the Anthropic action. Instead of a full foreign-access ban, GPT-5.6 enters a preview where the government approves enterprise customers one at a time. Trump’s June 2 executive order called the model-review framework voluntary. Two weeks later, both frontier labs are operating under it.

Commerce Secretary Howard Lutnick discussed GPT-5.6 with Altman directly on Wednesday, per Axios. Lutnick wanted assurance that all relevant parts of the government had tested and approved the model.

Anthropic gets the first shot at defining the AI IPO market

As Cryptopolitan reported in January, Anthropic hired Wilson Sonsini early to outpace OpenAI’s listing path. The company filed its confidential S-1 with the SEC on June 1, targeting an October 2026 Nasdaq debut at a $965 billion valuation, with Goldman Sachs, JPMorgan, and Morgan Stanley as lead underwriters.

The June 19 Trump softening on Anthropic, which we covered when President Trump told Axios he no longer viewed the company as a national security threat, cleared the political path for that listing. If OpenAI’s tilt to 2027 holds, Anthropic IPOs first in October 2026 and the bankers’ “whoever lists first defines the industry” framing becomes Anthropic’s headline.

Anthropic’s competitive case has hardened in the past 30 days. The company’s Polymarket-implied odds for best model are running at 94.8%. Anthropic is projected to break even by 2028 while OpenAI’s projected 2028 loss is $74 billion.

The macroeconomic environment has become more challenging because of the rate hikes from the ECB and BoJ, the shortage of memory in artificial intelligence reflected in the price hikes by Apple and Microsoft, and the retracement of SpaceX. Anthropic launches into a tougher environment than the one it pitched investors in January, but it still gets to launch first.

Don't Miss Today's Market    Moves Start trading Forex, Gold & Crypto today    Regulated platform | Fast withdrawals

Trade Now