The Forecast for 2027

A member of Anthropic's technical team, Tao Lin, indicated that by the end of 2027 artificial intelligence could reach or surpass the level of the best human experts in almost all sectors, from programming to film production. The estimate, reported by IT之家, was made on November 15 in an interview with Palisade Research and places the threshold at about 1.3 years away.

The source specifies Lin's role: he has worked at Anthropic for about a year and a half as a member of the technical staff, working on automated alignment research, AI control and safeguards, with the goal of making the Claude model safe. The forecast therefore comes from someone inside the company, but it is not presented as an official position of the company.

Employment: The Long-Term Scenario

In the same interview, Lin outlines a scenario in which, as technology advances, almost all people could permanently lose their jobs and depend on public income support. The mechanism described is gradual: AI takes over existing tasks, workers move to other activities, which in turn can be automated.

In support of his reasoning, Lin cites Anthropic internal data according to which 68% of observed uses of Claude involve tasks that in theory AI could perform on its own. The source emphasizes, however, that this does not automatically imply the replacement of real jobs, an important distinction for assessing the scope of the statement.

Why 2027 Will Not Be the Year of Mass Unemployment

The researcher himself plays down the short-term alarm: he considers it highly unlikely that large-scale unemployment will occur in 2027, because the process is constrained by the speed at which robots spread. Even assuming that AI absorbs much of intellectual work, replacing physical work would require advanced robots on a large scale, automated factories and adequate infrastructure.

A clear distinction emerges between two horizons: AI's ability to match experts is placed in the near term, while the concrete impact on employment depends on material factors and remains, by the admission of the author of the forecast himself, highly uncertain. The basis for the 2027 estimate and the representativeness of the 68% usage figure remain to be verified.

Source and attribution: this original summary was prepared by AICronaca based on the report by IT之家 ↗. The statements are attributed to the source; any informational limitations: The information comes from a single source, IT之家, which reports an interview with an Anthropic employee: these are personal forecasts and statements, not official positions of the company. No external data, independent replications or methodological details on the 2027 estimate and the 68% survey are available..