Anthropic CEO: Some Tech Giants Are Burning Cash on AI in a YOLO Manner, Risks Could Skyrocket

Anthropic CEO Dario Amodei admitted at the New York Times DealBook Summit that the entire AI industry’s data centers and infrastructure require massive capital investment and years of construction, yet the market remains uncertain about how quickly AI’s economic value will grow. He stated that Anthropic has managed its spending as responsibly as possible, but acknowledged that some companies in the industry are taking a “YOLO-style” approach—recklessly sprinting forward and ramping up risk to excessive levels.

A Dilemma for the Industry: AI Requires Huge Spending, but Payback Timeline Is Uncertain

Amodei pointed out that the core issue in the AI industry now is the growing gap between investment and return.

He said that data center construction is massive in scale, taking years from planning to completion, and requires tens or even hundreds of billions of dollars in capital expenditures just to get started. The problem is, no one can guarantee that AI’s commercial value will explode in a short time.

He described this as a real industry predicament, because everyone knows this is the direction for the future, but the current pace of investment may already be outpacing the market’s capacity to absorb it.

Some Companies Are Investing Without Regard for Risk, Setting Risk to the Max

Amodei went on to say that while Anthropic is also ramping up investment, it is deliberately controlling the pace, keeping risk within manageable bounds.

However, he then revealed that some companies have already begun investing in a “YOLO” manner—with a “you only live once” attitude, knowingly taking on extremely high risk and accelerating their spend. He did not name any companies, but described that some participants have taken risk to excessive levels, which could bring subsequent problems.

A Surge of Massive Investment Spurs AI Bubble Concerns

When Amodei’s remarks are put in the context of recent industry developments, it quickly becomes clear what he is referring to. Meta, Alphabet, and OpenAI have all recently announced even larger-scale AI investments, including data center expansion, chip procurement, and enhancing AI model computing power.

In particular, OpenAI’s announced AI infrastructure investment of up to $1.4 trillion has sparked widespread discussion about an “AI bubble.” This kind of massive spending is exactly what Amodei meant by “risk may be set too high,” pushing the entire industry’s capital intensity to historic extremes.

Anthropic Also Expanding in Preparation for 2026 IPO, Competing With OpenAI

While Amodei emphasized that Anthropic is relatively conservative in its investment pace, the company has still recently announced a $50 billion investment to build its first customized data centers at multiple locations in the US, demonstrating its ambition for steady expansion.

Amodei said that Anthropic has always maintained a “steady growth” strategy and will not ramp financial risk to the maximum. However, according to previous reports, Anthropic has hired Silicon Valley heavyweight law firm Wilson Sonsini to begin preparing for a 2026 IPO, with the company valuation potentially reaching $300 billion, making it one of the largest tech IPOs in history. At the same time, OpenAI is also making pre-IPO preparations, setting the stage for fierce competition between the two AI giants.

(Anthropic preparing for 2026 IPO, valuation could reach $300 billion, racing ahead of OpenAI)

This article, “Anthropic CEO: Some Tech Giants Are YOLO-Spending on AI, Risk May Be Off the Charts,” first appeared on Chain News ABMedia.