Anthropic bets on trust as a brand: the ad campaign that turns fear of AI into a selling point

By ADWEEK · July 10, 2026.

Anthropic has debuted "Hard Questions," a new advertising film from the agency Mother directed by Myles McAuliffe, within its "Keep Thinking" campaign, launched in September 2025. The ad opens with the image of a house burning at night and a voice asking "can AI be trusted?", stringing together uncomfortable shots—a child absorbed in front of a screen, facial recognition software—before turning toward a more hopeful tone, with questions such as whether AI could help people feel less misunderstood. According to the company, those questions weren't written by any ad copywriter: they come from conversations with more than 12,000 people around the world about their hopes and fears regarding AI. Alongside the film, Anthropic is launching an initiative for the public to keep asking questions and receive answers about the technology.

The context is no accident. In January 2026, OpenAI announced beta tests of advertising inside ChatGPT, and weeks later Anthropic aired two ads during Super Bowl 60 mocking how intrusive an AI with ads interrupting the conversation would be. Sam Altman called them "misleading"—a criticism worth treating as what it is, the response of a direct competitor, not an objective verdict—but the creative sector rewarded them: they took the Film Grand Prix at Cannes Lions 2026. "Hard Questions" is not an isolated ad, but the continuation of a coherent brand strategy: Anthropic positions itself as the AI that doesn't monetize the user's attention or sell them advertising, against a rival that is starting to do so.

Our reading is that Anthropic is selling something more valuable than a model: trust as a differentiated product in a market where consumer sentiment toward AI remains, as the article itself acknowledges, mixed. It's a smart move because it attacks the sector's weakest point in the short term: people don't distrust AI's capability, they distrust its incentives—who controls it, who profits from their data, who decides when to "hit the brakes." Turning those citizen doubts into the literal script of the ad, instead of neutralizing them with innovation slogans, is an unusual communication bet in an industry that tends to sell AI as frictionless magic. Whether it works will depend on whether Anthropic's business model—ad-free, with subscriptions and enterprise contracts—proves sustainable against the financial appeal of monetizing attention, something OpenAI is already starting to explore.

Over the longer term, this positioning war matters because the winner won't be simply whoever has the most powerful model, but whoever gets hundreds of millions of people to trust the technology enough to let it into increasingly intimate medical, financial or educational decisions. That trust is what will, over time, allow AI to be used fully to speed up diagnoses, discover drugs or manage resources with the abundance we defend as this technology's horizon. The short term brings real friction—distrust, cross-fire advertising scandals, aggressive brand competition—but the fight to be "the responsible AI" is, at bottom, a healthy sign: the industry is starting to compete on credibility too, not just on benchmarks, and that is a necessary condition for the public to accept delegating more to it.