Anthropic Launches Institute to Study AI's Impact on Jobs and Economy

Anthropic, the creator of the AI assistant Claude, announced on Tuesday, March 11, 2026, the launch of The Anthropic Institute, an internal initiative to study the effects of advanced AI on society. The San Francisco-based institute will utilize exclusive company data to anticipate challenges like mass unemployment and economic shifts. This comes at a tense time, shortly after the company sued the US government for being labeled a national security risk.

Focus on Real Risks and Key Hires

The new group combines three existing Anthropic teams, totaling about 30 professionals, including machine learning engineers, economists, and social scientists. They will focus on four main areas: AI's impact on jobs and the economy, threats to social resilience, real-world behavior of AI systems, and AI research governance. For instance, a recent company report from March 5, 2026, indicated that over 70% of customer service tasks are exposed to AI automation, according to real usage data from Claude. However, actual adoption remains low, allowing time for adaptations.

Jack Clark, co-founder of Anthropic and now head of Public Benefit, will lead the institute. He brings significant hires like Matt Botvinick, former research director at Google DeepMind, to study AI and law interactions; Anton Korinek, an economics professor at the University of Virginia, focused on economic transformations; and Zoë Hitzig, who recently left OpenAI to connect economics with model development. According to the official announcement on Anthropic's website, the goal is to share "candidate insights" on what the technology might cause, especially with predictions of "dramatic progress" in the next two years.

Context of Tension with the Government

The timing of the launch is noteworthy. Two weeks ago, on February 27, 2026, the US Department of Defense designated Anthropic as a supply chain risk, blocking federal contracts. This followed the company's refusal to allow Claude's use in mass surveillance or autonomous drones, leading to a lawsuit on March 9. Meanwhile, rivals like OpenAI have taken on the military contract. According to TIME magazine, Anthropic, valued at $380 billion, saw clients migrate to Claude after the incident, bolstering its ethical image but losing hundreds of millions in projected 2026 revenue.

In parallel, Anthropic is expanding its public policy team, tripling its size and opening an office in Washington, DC, this spring. Sarah Heck, former White House national security advisor, leads this effort, prioritizing model transparency and infrastructure investments. Reactions on X vary: some see it as a "genuine governance move," while others, like analyst Anish Moonka, question if it's a PR stunt during the crisis. The institute is already seeking hires to double the team annually, signaling that reports on AI-affected jobs should be released soon.