Dario Amodei Admits AI Companies Have Not Delivered on Their Promises
Dario Amodei concedes the industry has not met its lofty claims, arguing that only real scientific breakthroughs will shift public opinion, as his company, Anthropic, prepares for a listing valued at $2 trillion based on anticipated results.
August 16, 2026 – 9:01 pm
Credit: TechCrunch
Amodei wrote on X that the most accurate criticism of AI companies, including Anthropic, is their failure to deliver on promises. This comes as Anthropic prepares for a listing expected to value the company at $2 trillion in October, with its worth largely based on future potential rather than past achievements.
"The most accurate criticism of AI companies, including Anthropic, is that we haven’t yet delivered on our big promises to benefit the world."
Amodei’s perspective is clear-cut:
"At this point, saying that AI will cure cancer is more a cliché than it is inspiring… The thing that will work is actually curing cancer."
The timing of his statement adds weight to his argument.
The public sentiment supports Amodei’s view. Pew Research Center found that half of American adults are more concerned than excited about AI in daily life, while only 10% are more excited, up from 37% concern in 2021.
He dismisses the idea that his warnings contributed to this public distrust. "I do not agree that my messaging has been disproportionately negative," he wrote, referencing his 2024 essay Machines of Loving Grace, which argues AI could transform society for the better.
Instead, Amodei attributes public skepticism to a broader trend of mistrust towards corporations dating back decades, stating, "AI is just the latest iteration."
However, critics argue that AI companies will face scrutiny beyond just their results. An OpenAI employee noted that while pharmaceutical companies consistently deliver advancements, they remain one of the least trusted industries. They questioned whether AI companies can win public trust solely based on breakthroughs, highlighting concerns about pricing, access, lobbying, opacity, and power dynamics in addition to results.
Yann LeCun offered a contrasting view, advocating for widely available, shared, and open AI as the only viable path forward. This aligns with Meta’s push towards regulatory pressure in Washington.
Amodei concluded by setting expectations: Anthropic hopes to deliver "incredible results in the coming years and some early glimmers in the coming months," comfortably preceding any tangible signs of progress.