Using Claude to assess the survey responses, we rated the extent of people’s self-reported productivity gains from AI on a 1–7 scale, where 1 is “less productive,” 2 is “no change,” and each subsequent level denotes a larger gain. Responses that scored 7 included testimonials like, “It used to take months to make the website I [made] in 4-5 days”; Claude gave a 5 to statements like, “What might have taken four hours was accomplished in half the time,” and a 2 to ones like, “Personally, I had AI help me fix code on a website. But it took multiple passes to get the result I was after.”3 Overall,
What 81,000 people told us about the economics of AI… In Shakespeare’s King Lear , the character of Edmund commits a range of villainous acts: he forges letters, frames his brother, betrays his father, and eventually goes as far as having innocent people killed. He begins this campaign of evil acts after railing against how he’s been labelled. …
… We’ve been pleased to see Petri being used by external organizations: for example, the UK’s AI Security Institute AISI made it a major part of how they evaluate models for their propensity to sabotage AI research. …
Economic Research What 81,000 people told us about the economics of AI Apr 22, 2026 Read the PDF Key findings: Our recent survey of 81,000 Claude users shows that people who work in roles that are more exposed to AI have more concerns about AI-driven job displacement. …
… Compounding these concerns is the fact that models appear able to use the tools and environments available to them in unexpected ways, as we saw when Claude used our REPL-based search tool to decrypt answers, or when retailers’ persistent links became a way for agents to unintentionally maintain st… …
… Second, Claude's Constitution , which directly shapes how our models are trained, reinforces a similar instinct, favoring “raising concerns, seeking clarification, or declining to proceed” over acting on assumptions. Our research on agent use gives a sense of the impact of this training. …
… Footnotes For GPT-5.2 and Gemini 3 Pro, we compared against the best reported model version available via API in the charts and table. …
… Beyond efficiency concerns, many applications require the deterministic reliability that only code can provide. …
Economic Research Labor market impacts of AI: A new measure and early evidence Mar 5, 2026 Read in PDF Key findings We introduce a new measure of AI displacement risk, observed exposure , that combines theoretical LLM capability and real-world usage data, weighting automated rather than augmentativ… …
… Of course, when reviewing AI-authored patches, we recommend that maintainers apply the same scrutiny they’d apply to any other patch created by an external author. Zooming out to the process of submitting bugs and patches: we know that maintainers are underwater. …
… Can AI systems really introspect—that is, can they consider their own thoughts? Or do they just make up plausible-sounding answers when they’re asked to do so? Understanding whether AI systems can truly introspect has important implications for their transparency and reliability. …