Anthropic co-founder predicts AI will achieve Nobel prize-winning discovery within a year

Jack Clark, co-founder of the company behind Claude, thinks an AI system will collaborate with humans to produce a discovery worthy of a Nobel Prize within the next 12 months. He also predicts that bipedal robots will be assisting tradespeople within two years.

The Anthropic thesis, explained

His timeline aligns with projections from Anthropic CEO Dario Amodei, who has forecast that by 2026-27, frontier AI models could exhibit intelligence surpassing that of Nobel laureates across multiple fields. Biology, engineering, and other disciplines are all on the table, according to Amodei.

Nobel Prize-worthy discoveries don’t happen in a vacuum. They require experimental validation, peer review, and often decades of incremental work. An AI system that identifies a novel protein structure or proposes a new theoretical framework still needs humans to verify, replicate, and contextualize the finding. Clark’s prediction specifically involves AI “working with humans,” which is an important qualifier that tends to get lost in the headline.

The labor market problem nobody wants to talk about

Alongside the optimistic forecasts, Amodei has issued warnings that are harder to put on a keynote slide. He has cautioned that AI could push unemployment to 10-20%, with the heaviest impact falling on young workers in AI-exposed fields.

That’s not a hypothetical. Studies have already documented a 14% drop in job-finding rates for younger workers in AI-exposed roles since 2022. Entry-level white-collar positions, the traditional on-ramp to professional careers, are the most vulnerable category.

Critics have argued that the complexity of many jobs may actually underestimate AI’s potential impact on employment. The argument goes like this: people assume AI will struggle with nuanced, judgment-heavy work, but the latest generation of models is increasingly competent at exactly those tasks.

What this means for the AI investment thesis

Anthropic itself is privately valued in the tens of billions of dollars, backed by Amazon and Google. The company doesn’t have a token, but its predictions ripple through every adjacent market, from AI-focused crypto protocols to GPU suppliers to the decentralized compute networks that pitch themselves as alternatives to centralized AI infrastructure.

What investors should watch is not whether an AI system literally wins a Nobel Prize in the next 12 months. The Nobel committee moves slowly and the award often recognizes work done years earlier. The more useful signal will be whether AI systems produce results that the scientific community recognizes as genuinely novel and significant, the kind of work that would be in the conversation for major prizes if a human had done it.

Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our Editorial Policy.