Anthropic launched an initiative on Wednesday called “Inviting Hard Questions,” asking the public to submit its toughest concerns about artificial intelligence and committing to publish, and track, how it responds to them.
What the company is asking
According to Anthropic’s announcement, the company set up a website where anyone can read questions already submitted by others or add their own. The themes it is highlighting include whether AI will cost people their jobs, how it affects creative work and human agency, what happens if powerful AI capabilities end up in the wrong hands, and whether AI can help cure diseases. Anthropic says it will “publicly track and report the specific actions we’re taking to address those questions—and we’ll be clear about the ways in which we might fall short.”
The research behind it
The initiative draws on several rounds of public engagement Anthropic says it has already run. The company surveyed 52,000 Americans about their hopes and concerns about AI as the first round of what it calls the Anthropic Public Record. Separately, its Anthropic Interviewer project surveyed 81,000 Claude users across 159 countries and in 70 languages, alongside in-person focus groups and studies of anonymized Claude usage data.
Why it matters
Anthropic is structured as a public benefit corporation, overseen by an independent Long-Term Benefit Trust, and the company has been leaning on that structure to argue it should be held to public accountability rather than judged on commercial results alone. Publishing a running account of unresolved public concerns — and its own admitted shortcomings — is a more exposed position than a typical corporate transparency report, at a time when governments and the public are still debating how AI companies should be governed and who gets to set the rules for the technology’s rollout.