Picture this: it's post-Google I/O, the confetti has barely settled, and Demis Hassabis — the CEO of Google DeepMind, one of the most respected AI researchers alive, and genuinely one of the people most responsible for the AI revolution you keep reading about — sits down and says, calmly, that humanity is standing in the "foothills of the singularity."
Not a Reddit doomsday thread. Not a fringe tech podcast at 2am. The guy building the thing.
For the uninitiated: the "singularity" is a long-theorised moment when AI becomes so advanced and self-improving that it fundamentally changes everything — civilisation, work, knowledge, power. Think of it less like a sci-fi movie and more like a point of no return on a graph that's currently going vertical. Hassabis isn't saying we're there. He's saying we're at the base of the mountain, lacing up our boots, and the summit is visible.
His updated prediction? AGI — artificial general intelligence, meaning AI that can do essentially anything a human can do intellectually, and then some — could arrive by 2029. That's not a wild guess from a blogger. That's a four-year timeline from one of the most credentialed people on the planet, who has access to the lab results the rest of us don't see.
He also admitted he used "provocative" language deliberately, because — and this is the quote that should be printed and pinned in every government ministry and boardroom — "my economist friends, I feel, are still not taking this seriously enough."
Which raises one genuinely uncomfortable question: if the builders are racing the clock, what on earth are the rest of us supposed to do with the next three years?
