The Race for AGI: Hassabis and Amodei Argue About the End of the Familiar World

In Swiss mountains, where the fate of the world economy is usually decided, this time they tried to understand when that same economy will cease to be purely a human affair. In Davos, an event occurred that the industry had long awaited: Demis Hassabis from Google DeepMind and Dario Amodei from Anthropic sat at the same table. The moderator, sparing no pathos, compared them to the Beatles and the Rolling Stones.

The comparison limps, because unlike rock stars, these two are not arguing about charts, but trying to calculate the date when artificial intelligence will match us or leap over humanity's head. The theme was announced ambitiously: "The Day After AGI." But before planning breakfast in a world of superintelligence, we need to understand when this very dawn will come.

Context is more important here than the words themselves. A couple of years ago, the term AGI (artificial general intelligence) was considered the domain of geeks and science fiction writers. Today it is a line item in quarterly reports of the world's largest corporations.

Hassabis, the man who essentially launched the modern wave of AI by acquiring DeepMind for Google, has always been a cautious optimist. He builds systems that play Go, predict protein structures, and now try to reason. His vision of AGI is a tool that will solve problems in physics and biology.

For him, it is evolution, albeit a very rapid one. He sees in this a powerful lever that will help humanity cope with what we ourselves cannot handle.

Dario Amodei from Anthropic plays a different game. His company was born from a split in OpenAI precisely over safety concerns. If Hassabis is a research engineer, then Amodei is a man who looks at exponential curves and sees in them not only progress but also threat. In Davos, he reminded again about scaling laws. We pour billions of dollars into computing power, and neural networks dutifully become smarter. The problem is that this growth is not linear. We might wake up in a world where a next-generation model suddenly displays capabilities we are not ready for—neither legislatively nor morally. Amodei emphasizes that the point of no return is closer than it seems to many regulators.

The difference in their approaches lays bare the main conflict in the industry. On one side—the belief that we will be able to control the process and gradually integrate AI into science and daily life. On the other—the fear that the pace of technology development has already exceeded the pace of adaptation of social institutions. When the moderator asked about specific timelines, tension hung in the air. Hassabis bets on the end of the decade, while Amodei hints that surprises might begin much sooner. And this is not mere coffee-ground divination. On these dates depends where the next trillions of dollars in investment will flow and what laws the EU and USA will pass in the coming months.

Why is this important right now? Because the era of "toy" chatbots is ending. We are moving toward agents that can act independently.

If the leaders of two companies that essentially hold the keys to this technology disagree in their assessment of fundamental risks, that is cause for thought. We are in a situation where the creators of the engine are arguing about whether the machine has brakes and where the cliff edge is. The irony is that both continue to press the gas, because stopping in this race means losing everything.

The Davos meeting showed that there is no consensus even at the very top, and "the day after AGI" can come suddenly for all participants in the process.

Main point: Hassabis and Amodei agree on one thing—AGI is inevitable and near. But if Google sees in it a super-assistant for scientists, then Anthropic warns of a tectonic shift that the world is not ready for. Will we manage to agree on the rules of the game before AI starts writing them itself?