Sam Altman, CEO of OpenAI, says the singularity is not a distant event to anticipate. He says it is already happening. In a blog post titled The Gentle Singularity, Altman wrote: “We are past the event horizon; the takeoff has started.” He described this moment not as a sudden rupture but as a gradual shift that society is already inside.
His framing directly pushes back on the science fiction version of singularity — the idea of a sudden, uncontrollable moment when AI surpasses human intelligence and everything becomes impossible to predict. Altman described the actual experience as something quieter: “Wonders become routine, and then table stakes.” The change is real, but it does not look the way most people imagined.
He wrote that “humanity is close to building digital superintelligence, and at least so far it’s much less weird than it seems like it should be.” That statement captures the core tension in his argument — that something historically significant is underway, but it’s unfolding in an ordinary-seeming way.
His predictions are specific. For 2026, he expects AI systems to begin generating genuinely novel scientific insights — not just summarizing or retrieving information, but producing original discoveries. For 2027, he points toward robots capable of operating physically in the real world at meaningful proficiency. Over a longer horizon — within a decade — he suggests intelligence could become “too cheap to meter,” meaning the cost of AI capability would eventually become so low that price stops being a barrier to deployment.
The phrase “too cheap to meter” is borrowed from early predictions about electricity. It implies a future where access to intelligence is as ambient and inexpensive as electric power is today. OpenAI has been working to position itself for that scale, and its structural and investment decisions over the past year reflect that ambition. The company has also moved into hardware development, acquiring Jony Ive’s design firm io for $6.5 billion to build AI-native devices that could serve as interfaces for this kind of ubiquitous intelligence.
Altman’s perspective has shifted over the years. Earlier, he acknowledged existential risks from AI development as a serious concern. The Gentle Singularity essay marks a move toward a more optimistic and gradual framing — still acknowledging transformation, but presenting it as navigable rather than catastrophic.
What he has not done is predict specific breakthrough dates with certainty. His predictions come with implicit room for compression or delay. The singularity, in his view, is already underway — the question is how fast each stage unfolds.