Dario Amodei admits AI suffers from a crisis of trust, saying people worry companies or governments are ‘cooking up some new way to screw them over’

Anthropic's CEO Dario Amodei addressed the public's overall sense of apprehension surrounding AI in a recent X post, clarifying that he is not the sole cause of this anxiety. He acknowledged that there are trust issues, delving into the topic of AI regulation. Amodei presented a false dichotomy between those who argue regulation leads to power concentration and those who believe open-access models are the key to controlling AI.

He pointed out that institutions like courts can decentralize power, and Anthropic has advocated for policies that slow down frontier AI while giving smaller competitors an edge. Amodei conceded that AI inherently tends to concentrate power due to scaling laws, but claimed that well-designed rules can address AI's risks, constrain powerful companies, and accommodate open-weight models.

He supported the creation of a FINRA-like entity and the Trump administration's stance on AI testing. Addressing his own communications, Amodei denied that he has been overly negative about AI, pointing to essays that present both risks and benefits. He noted that social media often mischaracterizes his statements to increase clicks.

The Anthropic CEO admitted that the public has a negative view of AI, which he considers a significant problem, but emphasized that the solution lies in delivering on promises to benefit the world, rather than making empty promises. Anthropic is working towards improving trust by focusing on advancements in biology and medicine.

Until they achieve tangible results, Amodei vowed to be honest about AI risks and avoid misleading rhetoric.