Sam Altman Delays OpenAI IPO to 2027 on Safety Concerns, Worldcoin (WLD) in Focus

IPO Off the Table Until 2027

Sam Altman said OpenAI will not go public this year, pushing the most-anticipated tech listing to at least 2027 — news that lands directly on the Worldcoin (WLD) ecosystem, the crypto project the OpenAI chief co-founded and whose token often trades as a proxy for his broader agenda. “Given everything happening with safety, right now would be an ill-advised moment to go public, and we don’t feel pressure on that,” Altman said in an interview published Friday, September 12. He added that the company still has “a lot of stuff to do,” from meeting what will be required on safety and alignment to determining how the industry and governments can work together.

The caution fits a defensive stretch for the AI sector’s leadership. Anthropic CEO Dario Amodei publicly called on Saturday for a slowdown in the AI race, with both Altman and Elon Musk quickly agreeing — an alignment of rivals that would have seemed improbable a year ago. The statement also extends a thread this desk has tracked: Altman previously backed an AI pacing plan endorsed by 1,300 staff, and an OpenAI safety researcher has publicly assigned 70% extinction odds to advanced AI — the same risk framing Altman now cites for staying private. For Worldcoin (WLD) holders, the practical read is that Altman’s attention and capital stay locked inside a private OpenAI: there is no listing event to trade, and the token side of his empire — distributed to verified humans through the World App wallet rather than an ICO-style public sale — keeps moving on his headlines alone.

GPT-6 Astra Faces ‘Nerf’ Claims

OpenAI’s newest flagship, GPT-6 Astra, is facing a wave of “nerfed” complaints just one week after launch — and the pattern matters for WLD watchers because every OpenAI quality controversy now ripples into Altman-linked assets. Astra wowed users at launch with demos including rebuilding Manhattan street by street inside a game engine, but sentiment on X flipped sharply this week. Pseudonymous developer synthwavedd wrote that Astra “feels significantly dumber” and called it “The Post-Launch Lobotomy.” Developer Pranjal Paliwal, an early admirer, re-read the code the model had written him and reversed course, posting on X that the company does not have AGI — artificial general intelligence — but “a regression.”

@betterclever · X post

Posting on X.

View on X

The complaint cluster shares a shape: answers arrive faster but worse, and users suspect OpenAI quietly cut the “juice value” — the computing effort the model spends thinking before it replies. Two independent users ran identical prompts against launch-day Astra and today’s build, and both got worse results from the current version. Dax Raad’s team behind the coding tool Opencode switched back to Astra’s predecessor, GPT-5.6 Sol, saying costs had doubled for downsides not worth paying. Skeptics push back: the most detailed rebuttal argues Astra is “as dumb as it was on launch,” and the honeymoon simply ended. The playbook is familiar — Sol went through the identical cycle in July, when an OpenAI executive denied deliberately weakening the model while confirming experiments with the reasoning-effort setting. Astra still prices at $10 per million input tokens and $50 per million output tokens, 2.5 times Sol’s launch rate, and it remains OpenAI’s first model to cross the company’s cybersecurity-risk threshold, a capability restricted to vetted defenders under the Daybreak program. Readers tracking the market in real time can follow live spot and futures prices on Bitget.

WLD’s Next Catalyst Is OpenAI Itself

The reversal marks a sharp turn from OpenAI’s own timeline earlier this year. In May, reporting indicated the company had been working with Goldman Sachs and Morgan Stanley on listing documents, with a secret draft filing possible within days and a target of going public before September 2026; Altman himself was said to favor a fourth-quarter 2026 debut. Chief financial officer Sarah Friar, by contrast, pushed for 2027 to give the company more time to strengthen its financial structure. Her caution centered on a stark gap: OpenAI has signed nearly $1 trillion in compute and data-center commitments against roughly $25 billion in annualized revenue — a mismatch of more than 40 times that would face quarterly scrutiny as a public company. Altman’s stated rationale has now shifted accordingly, from financial readiness to safety.

(as of 00:00 UTC) Taken together, the two items sketch OpenAI at a governance crossroads: a CEO declining a public float until 2027 on safety grounds, while users accuse his flagship model of regressing a week after launch. Both claims are verifiable at the primary layer — Altman’s own published statement and Paliwal’s public post documenting the code regression. In our reading, the tradable consequence sits with Worldcoin (WLD), which has repeatedly re-rated on Altman headlines, including the recent Google RSI model leak rumor, with sentiment proxies ranging from prediction markets to spot order books. Until OpenAI addresses the Astra complaints or revisits a listing, WLD remains a headline-driven sentiment proxy rather than a fundamentals story.