Binance’s CZ: Most AI Startups Will Fail as Anthropic Nears $1T

Changpeng Zhao, founder of Binance, posted on X on Friday that most artificial intelligence companies will fail, calling the sector overcrowded even as large funds flow to a few firms.

Zhao wrote: “AI will stay and grow exponentially. But most AI companies will go bust. There are just too many. Even survivors will see huge price fluctuations. There will be new survivor entrants too.” He framed the pattern as typical of early-stage industries that attract waves of capital and then consolidate.

Anthropic announced a $65 billion Series H round that left the company with a $965 billion post-money valuation and reported a $47 billion annualized revenue run rate. The fundraising substantially increased Anthropic’s implied value and placed it close to the trillion-dollar mark used to rank the largest private tech firms.

OpenAI is valued at about $852 billion after a March funding round and has begun preparing a confidential S-1 with investment banks, with a possible public debut as early as September at an expected valuation above $1 trillion.

Some corporations buying AI services have slowed hiring and reviewed budgets tied to the technology. A ride-hailing company’s chief executive said the firm reduced hiring to absorb AI investments, while its technical leaders reported exhausting a 2026 budget for third-party developer tools in four months and questioned whether higher token usage has improved consumer products.

A February study from an economic research group found 90% of surveyed firms reported no measurable productivity gains from AI. OpenAI has guided to operating losses through at least 2028, including a projected $74 billion in operating losses that year, and has outlined roughly $1.4 trillion in datacenter spending over eight years.

Major cloud providers have recorded large forward commitments linked to leading AI labs. Two top labs account for more than half of about $2 trillion in future cloud commitments from major hyperscalers.

Anthropic is reportedly on track for its first operating profit this quarter. Most other firms in the sector continue to spend faster than they earn, according to company disclosures and investor reports.

Zhao’s post came during a busy period of fundraising for AI startups and labs, where a handful of firms have attracted the bulk of capital while many smaller companies compete for customers and ways to cover heavy infrastructure and development costs. Investors and analysts are awaiting OpenAI’s S-1 to review how a company with an implied trillion-dollar valuation presents revenue, costs and cash flow on a public balance sheet.