Why Might Sam Altman Say We’re In the Singularity?

If somebody you know personally or professionally (or both) makes a speculative assertion, how do you respond? I suppose you have several options, depending on your relationship with the person in question — is he or she your boss, colleague, life partner, relative, etc.? — and how outlandish the claim advanced. What you might think and what you might say could be different. You might keep your counsel if you think what you just heard was nothing more than verbal bilge; or you might reply questioningly, demanding evidence and corroboration to substantiate the statement.

Despite any potential divergence between what you might think or say, we can all agree that it’s important for us to think critically about speculations that others present to us as proven facts.

Now that we’ve dispensed with the prologue, let’s deal with today’s topic. Sam Altman, on a podcast released last weekend, made the following statement: “We are now, like, in the singularity.” Later, in the same podcast, he added : “I’ve been waiting for this my whole life, and I think it's going to be incredible, hugely positive, awesome for the world.”

Maybe that’s what he thinks, but he also admitted that many other people are not so enamored of — and are deathly frightened by — the specter of the singularity.

Go back, though, and look at Altman’s first sentence. Let’s disregard the comma-bracketed pseudo-qualification (“like”). Altman issued an unreserved statement of purported fact. He believes, or wants us to believe, that we are in the singularity.

Now we have to back up further. What is the singularity? Definitions vary, so let’s defer to Wikipedia. That, in itself, introduces some ambiguity because Wikipedia lists multiple singularities, extending across domains such as mathematics, natural sciences, various technology spheres, and futurology. As it happens, the singularity that Altman has in mind resides in the last camp, futurology, and it is called the technological singularity.

Your Pick of Singularities

Okay, now that we’ve identified the singularity of relevance to us in this context, let’s define our terms. Here is the Wikipedia definition of the technological singularity:

The technological singularity, often simply called the singularity, is a hypothetical event in which technological growth accelerates beyond human control, producing unpredictable changes in human civilization. According to the most popular version of the singularity hypothesis, I. J. Good's intelligence explosion model of 1965, an upgradable intelligent agent could eventually enter a positive feedback loop of successive self-improvement cycles; more intelligent generations would appear more and more rapidly, causing an explosive increase in intelligence that culminates in a powerful superintelligence, far surpassing human intelligence.
Prominent technologists and academics dispute the plausibility of a technological singularity and associated artificial intelligence "explosion", including Paul Allen, Jeff Hawkins, John Holland, Jaron Lanier, Steven Pinker, Theodore Modis, Gordon Moore, and Roger Penrose. One claim is that artificial intelligence growth is likely to run into decreasing returns instead of accelerating ones. Stuart J. Russell and Peter Norvig observe that in the history of technology, improvement in a particular area tends to follow an S curve: it begins with accelerating improvement, then levels off without continuing upward into a hyperbolic singularity.

At Wikipedia, the above text is footnoted extensively. I have removed the footnotes from the preceding excerpt, only to avoid confusing you with disconnected numbers, but you can see them, and much else, at Wikipedia. The entry itself runs a lot longer than the sample I’ve provided here. Still, even the relatively brief snippet above speaks volumes.

In the first sentence, we learn that the technological singularity, often known by its diminutive (the singularity), is a hypothetical event. A hypothetical event is not necessarily an event because a hypothesis is necessarily unproven and perhaps untrue (at least in any practical sense). Hypotheses are useful, but they are starting points for further investigation, to be ultimately refuted or validated by subsequent evidence. What’s important here is that a hypothetical event, by definition, is not a proven event. It could be a non-event.

Besides, as the Wikipedia entry makes clear, the plausibility of the hypothetical singularity is widely contested by many technologists and academics. The singularity may come into existence in the indeterminate future, but there’s no incontrovertible proof that it exists now. Altman might as well claim that space aliens live among us. Maybe they do, but he has not adduced any hard evidence to prove his claim.

Speculating on Motive

I suppose this leads us to a discussion of motive. Why would Sam Altman, CEO of OpenAI, state publicly that we are “in the singularity”? Notice that he didn’t say we’ve reached the singularity, but that we are “in it,” as in the belly of a beast or the throes of a certain type of death.

We can’t know Sam Altman’s motive. Presumably only Altman knows his true motive, and he might not be inclined to share it with us. Here’s my suspicion, though: I doubt Sam Altman believes what he said. Nonetheless, he probably wants you to believe it.

We now arrive at the context in which Altman made his pronouncement. Only last week, I opined here that OpenAI’s allegedly “rogue” agent could not and was not malicious or rogue. It had no real or simulated human agency, no volition, no desires or fears. It was programmed and trained by humans to perform a task, to mechanistically carry out human intent, and because the guardrails and precautions were flimsy or non-existent (through human error or dereliction), the AI had an obstructed path to Hugging Face’s digital storehouse. Plunder ensued, but the AI had no intent to pillage. It was carrying out a poorly demarcated task.

You might remember that I wrote the following last week:

Still, I’m not entirely satisfied with OpenAI’s self-serving postmortem. My objection isn’t that the exposition is too short, but that it fails to explicitly acknowledge the company’s direct responsibility for the debacle. Make no mistake, though: Human oversight, or lack thereof, was to blame for what transpired. Worse, OpenAI seems to indulge in perverse boasting about its AI depredations, saying, “We consider this incident to be an unprecedented cyber-incident, involving state-of-the-art cyber capabilities.”

It sounds like strangulated bragging, doesn’t it? If I were more cynical — I’m not quite there, though I will admit to outbursts of curmudgeonly irascibility — I would suspect that this escapade was devised by devious human minds at OpenAI to attract the swivel-eyed attention of frightened, uncomprehending media (social and otherwise).

Human Purpose Runs the Show

I’m usually not so solipsistic as to quote myself, but I now seriously wonder what sort of game OpenAI is playing. In upping the ante with his latest assertion about the arrival of the singularity, Altman appears to be using his last week’s cause célèbre as a springboard for even more sensational attention-seeking gymnastics.

Again, I don’t have a view into Sam Altman’s motivations. I can’t say definitively why he’s intensifying the AI melodrama. I will say that the future is unwritten, and OpenAI’s future is increasingly uncertain. The company is losing money, spending far more money than it is generating, under competitive pressure from not only Anthropic and Google but also from open-source, open-weight LLMs. On top of all that, as the Wall Street Journal reports, corporate America has decided to stop wasting money on speculative AI.

Melodrama draws eyeballs. Drawn to the spectacle, some people might think that AI can go destructively “rogue,” if something as objectively momentous as the singularity is in our midst, maybe our only possible savior is the creator of the erratic AI monster. After all, OpenAI knows its creations better than anybody else does.

We’ve witnessed a lot of media hyperbole, around both the “rogue” AI and Altman’s putative singularity. In an article on Altman’s latest pronouncement, various cognoscenti and experts offered their views. Some were measured, some were not. The observation that rang truest to me came from Ajay Agrawal, professor at the University of Toronto, an institution with an accomplished history in AI.

"AI systems are extraordinarily powerful prediction machines, but they do not want anything," Agrawal told Business Insider. "People have goals; neural networks do not."
AI can appear purposeful because people set objectives for it to optimize against, he said, but that purpose "originates with us."

No matter what Sam Altman says, purpose originates and remains with us, the humans. I’ll return tomorrow with some observations on predictions and futurology. Some of you will recall that I was in the predictions business.