Anthropic Employees Expect AI Nobel-Level Discovery by 2030

Anthropic's internal survey of 1,334 employees finds 91% expect AI could make a Nobel-level discovery by 2030, compared with 39% of 10,980 U.S. respondents. The report also details divergent expectations on workplace automation.

Anthropic Employees Expect AI Nobel-Level Discovery by 2030
Reading time: 2 Minutes
Follow on Google

Anthropic said 91 percent of its employees expect artificial intelligence could achieve a Nobel-level scientific discovery by 2030, according to an internal survey of 1,334 staff published in a company research report.

The report compared those results with responses from 10,980 U.S. adults; 39 percent of the U.S. respondents said they expect AI to reach that level by 2030.

The survey asked whether AI could make a scientific discovery that would markedly change human understanding of a phenomenon. The report notes that winning a Nobel Prize is a separate outcome that depends on the decisions of prize committees.

Survey results also covered workplace automation. More than half of Anthropic employees said they expect AI to perform at least 80 percent of their current tasks by 2030, while only 5 percent of U.S. respondents gave the same answer.

Anthropic employees on average expect AI to handle about 75 percent of their current work by 2030; the average expectation among U.S. respondents was 28 percent.

The report suggests one possible reason for the gap is staff members exposure to advanced models in their daily work, since they develop and use the technology and may observe signs of progress earlier than the general public.

The report emphasizes that closer experience with AI does not necessarily make forecasts more accurate and frames the findings as a contrast between those who work daily with advanced models and the wider public on AI's future role in research and workforce automation.

Chloe Nakamura

“I love exploring gadgets, apps, and trends that redefine how we connect, work, and play in a digital world.”

Leave a Comment

Comments

No comments yet. Be the first.