Anthropic chief talks AI bubbles and risk
In this piece, Anthropic chief executive Dario Amodei responds to questions about whether artificial intelligence is in a speculative bubble. He argues that, while there is plenty of hype and questionable investment, the underlying technology is a genuine platform shift that will unfold over many years rather than collapsing overnight.
The article also digs into his concerns about safety and rival behavior. Amodei suggests that some competitors are racing to release ever more capable systems without enough testing, driven by fear of falling behind. He calls for a balance where firms still innovate but treat model evaluation and governance as core engineering work, not public relations.
The author uses the conversation to illustrate the tension at the heart of the industry; companies are under pressure to grow fast, but the people building the models know that poorly controlled systems could have real world consequences. The debate is less about whether AI is important and more about how responsibly it will be scaled.
