Summary:
-
Anthropic’s controversial ad “There’s hope in hard questions” aired during a World Cup match, sparking confusion and criticism.
-
The film is based on extensive research and public input, but critics question Anthropic’s credibility and track record.
-
Despite the high-stakes approach, Anthropic’s ad raises doubts about trustworthiness and accountability, overshadowing its intended message.
There’s a rule in advertising that you don’t open on a burning house. Anthropic opened on a burning house.
The AI company’s newest spot, “There’s hope in hard questions,” went live July 9 and aired to a World Cup audience during Argentina vs. Switzerland. It moves through surveillance footage of faces being scanned in a crowd, and — the shot everyone screenshotted — rows of military gravestones, timed to a voice asking, “Who’s going to hit the brakes if we need to?” It closes on someone staring out at the ocean.
Agency Mother made it. Myles McAuliffe directed. It is, technically, beautiful. It is also the reason a lot of people spent Monday asking whether Anthropic’s social accounts had been compromised.
Sam Altman got there first, posting on July 14 that he assumed the campaign was satire and kept scanning for a parody handle. Then he pivoted to the actual knife: hard questions are great, he suggested, provided Anthropic deems you worthy enough not to get silently downgraded. The two CEOs have been circling each other since Dario Amodei left OpenAI in 2020, and Altman had already called Anthropic’s Super Bowl ads “deceptive” earlier this year.
The film isn’t freestyling. It’s built on the Anthropic Public Record, a survey of 52,000 Americans on their hopes and fears about AI, plus interviews with 81,000 Claude users across 159 countries and 70 languages. The company is asking the public to submit their own hard questions and has committed to publicly reporting what it does about them — including where it falls short.
ADVERTISEMENT
Which makes this a familiar maneuver dressed in unusually expensive clothes: name the harm loudest, and the naming becomes the credential. Critics point out the track record complicates it — a safety commitment dropped in February, and reporting that Claude was used to help select strike targets in Iran even as the company publicly resisted surveillance and autonomous weapons applications.
The bet is that visible discomfort builds more trust than a polished feature reel. The bet may still pay — but only if the answers show up. Otherwise the ad’s own question does the work its critics never had to: can Anthropic be trusted?
Which, as more than one person noticed, is not the question the ad meant to leave you with.
