Monday, February 16


Anthropic CEO Dario Amodei warns rivals on AI spending: It is not cool to …

Anthropic CEO Dario Amodei has warned its competitors about reckless capital expenditures on artificial intelligence (AI). He noted that aggressive capital allocation for AI could lead to catastrophic miscalculations. Speaking in an interview with podcaster Dwarkesh Patel, he criticised rival AI companies for “YOLOing” on spending and “just doing stuff because it sounds cool.”Amodei also defended his company’s more cautious spending approach compared to other AI hyperscalers that commit hundreds of billions of dollars per year.Amodei said even a slight miscalculation could sink Anthropic, explaining why the Claude chatbot developer takes a more measured approach despite his earlier prediction that an AI data centre could one day be a “country of geniuses.” He acknowledged that Anthropic spends less than some competitors and accepts the risk that the company may not be able to meet all AI demand. Without naming specific companies, Amodei said rivals do not fully understand the risks of their spending strategies.

‘AI Works On Data, Humans Create The Unseen’: CBFC Chairman Prasoon Joshi At AI Impact Summit 2026

What Anthropic CEO Dario Amodei AI’s economic returns

Amodei replied that while he is confident the technical milestone is achievable soon, he’s less certain about the timing of AI’s economic returns. “I really do believe that we could have models that are a country of geniuses in the data centre in one to two years. One question is: How many years after that do the trillions in revenue start rolling in? I don’t think it’s guaranteed that it’s going to be immediate. I think it could be one year. It could be two years. I could even stretch it to five years, although I’m sceptical of that,” he noted.Because of uncertainty about how quickly revenue will grow, spending large sums now to build data centres rapidly could be “ruinous” if estimates are off by even a small margin, Amodei warned. In November, Anthropic said it would invest $50 billion in AI infrastructure in the US, starting with data centres in Texas and New York.Meanwhile, the top hyperscalers surprised Wall Street in recent weeks with plans to boost capital expenditures by considerably more than expected. For example, Amazon plans to spend $200 billion this year alone, while Alphabet projects up to $185 billion, and Meta expects capex of up to $135 billion.To illustrate his point about the timing of returns from AI investments, Amodei highlighted the potential for medical breakthroughs that could drive substantial economic value.There’s the question of how much of the gains pharmaceutical companies receive versus AI companies. The research, manufacturing, and regulatory processes also take time. Amodei noted that after the first COVID-19 vaccines were developed, it took about 18 months to achieve broad distribution.When it comes to buying data centres, he looks at Anthropic’s 10-fold annual revenue growth, with 2026 projected at around $10 billion. At the same time, building and reserving a data centre takes one to two years. By then, revenue could top $1 trillion if it continues on its current trajectory, allowing the company, in theory, to allocate a similar amount to data centres.“If my revenue is not $1 trillion, if it’s even $800 billion, there’s no force on Earth, there’s no hedge on Earth that could stop me from going bankrupt if I buy that much compute. Even though a part of my brain wonders if it’s going to keep growing 10x, I can’t buy $1 trillion a year of compute in 2027. If I’m just off by a year in that rate of growth, or if the growth rate is 5x a year instead of 10x a year, then you go bankrupt,”Amodei explained. He also noted that Anthropic’s AI is geared toward enterprise customers rather than fickle consumers, enabling them to rely more on revenue. Overall, Anthropic’s spending on computing capacity is still substantial.“We’re buying an amount that’s comparable to what the biggest players in the game are buying. But if you’re asking me, ‘Why haven’t we signed $10 trillion of compute starting in mid-2027?’ First of all, it can’t be produced. There isn’t that much in the world. But second, what if the country of geniuses comes, but it comes in mid-2028 instead of mid-2027? You go bankrupt,” Amodei added.



Source link

Share.
Leave A Reply

Exit mobile version