Quote of the weekend, from this TechCrunch article:

Instead, he insisted that OpenAI will go public “when we’re ready, which is when the business is ready, when we feel ready from what the moment is like in society with this technology.” When pressed on whether that means the IPO isn’t happening in 2026, Altman replied, “I would say not 2026, yeah. We’ve got a lot of stuff to do.”

Would say?  Why are we would-saying?

Remember, OpenAI has achieved AGI.  In fact, they’ve achieved it multiple times:

• September 27, 2023: Sam said, “AGI has been achieved internally.”
• January 5, 2025: Sam said, “We are now confident we know how to build AGI as we have traditionally understood it.”
• June 10, 2025: Sam said, “We are past the event horizon; the takeoff has started.”
• December 18, 2025: Sam said, “AGI kind of went whooshing by… okay, fine, we’ve built AGIs.” He also added, “We’re at AGI.”
• February 3, 2026: Sam said, “We basically have built AGI, or very close to it.”

The standing narrative is that once AGI is achieved, ASI happens moments later because as soon as AGI is “the smartest human + .0001%” then it can start recursively self-improving and the curve goes ballistic.  This leads very quickly to datacenters full of PhDs.

Surely, those banks of super-brilliant systems are going to advise OpenAI on the precise moment to go public, right?  They can outthink anything Wall Street has, outthink the market, and outthink all of humanity.

I think it’s fair to hold these AI companies to this standard.  They alternate claiming their systems are a dire threat of mass extinction and the beacon of a golden future, which implies they have unbelievable capability.  So things like IPO timing should be child’s play.  Really, should Anthropic of OpenAI ever make any kind of mistake again?

I’m reminded of a true story about Siebel Systems, who once was the 800 pound gorilla of the CRM software industry, long before Salesforce existed.  In 2002, a small research firm took the 66 marquee reference accounts that had been mentioned in Siebel’s sales literature and conducted a survey of them.  They reported that a large percentage – up to 75% depending on the metric – were very unhappy with Siebel and its software.

This was a major scandal at the time, and for Siebel it was fatal.  They sold software that tracked customer relationships, and the relationship they had with their customers was apparently awful.  A few years later they were bought out by Oracle.

What killed Siebel was the hypocrisy of selling software that was supposed to help companies delight their users, and yet Siebel apparently either wasn’t using it or the software didn’t work.

Likewise, with any misstep OpenAI or Anthropic makes, aren’t the admitting that AGI/ASI is still a distant pipe dream?  Shouldn’t this fabulous technology be right at their side, providing them with infallible advice?

There’s no “would say” about anything when you have Godlike intelligence at your fingertips.

raindog308

raindog308 is a longtime community LETizen, technical writer, and self-described techno polymath. With deep roots in the *nix world, he has a passion for systems both modern and vintage, ranging from Unix, Perl, Python, and Golang to shell scripting and mainframe-era operating systems like MVS. He’s equally comfortable with relational database systems, having spent years working with Oracle, PostgreSQL, and MySQL.

As an avid user of LowEndBox providers, raindog308 runs an empire of LEBs, from tiny boxes for VPNs, to mid-sized instances for application hosting, and heavyweight servers for data storage and complex databases. He brings both technical rigor and real-world experience to every piece he writes.

Beyond the command line, raindog308 has a life-long love of German Shepherd Dogs, high-quality knives, target shooting, theology, tabletop RPGs, playing guitar, and hiking in deep, quiet forests.

His goal with every article is to help users, from beginners to seasoned sysadmins, get more value, performance, and enjoyment out of their infrastructure.

You can find him daily in the forums at LowEndTalk under the handle @raindog308.