Nvidia has scaled back its financial commitment to OpenAI following pressure from investors. The Wall Street Journal reports that the company initially planned to guarantee $250 billion but will now provide less than $120 billion for the first phase of the data center project. This move comes as critics warn about the risks of over-investing in AI infrastructure. The guarantee covers the construction of about five gigawatts of capacity, with OpenAI separately negotiating a lease for the full 10-gigawatt project developed by SB Energy, a SoftBank subsidiary. Nvidia is also discussing financing for OpenAI’s chip purchases worth up to $350 billion. The reduced investment is seen as a potential signal that even major players in the AI industry are becoming more cautious.
Anthropic’s rapid revenue growth challenges the notion of an AI bubble. According to Reuters, the company’s revenue more than doubled in a single quarter, rising from $4.73 billion in Q1 to over $11.5 billion in Q2, a 14x increase year over year. Financial insiders suggest Anthropic projects revenue of around $190 billion to $200 billion for 2028, far exceeding its previously reported annual run rate of about $45 billion. The company claims it grew revenue by more than 10x each year leading up to early 2026. If these figures are accurate, they indicate sustained demand for proprietary AI services despite political challenges and competition from China.
The situation highlights the divergence in fortunes among AI firms. While Nvidia faces pressure to reduce its exposure, Anthropic continues to grow rapidly, with plans to go public at a valuation near $1 trillion in late September or early October. However, some signs of slowing demand for Anthropic tokens have been reported by financial services firm Ramp, though the overall trend remains positive. These developments underscore the complex landscape of the AI industry, where success and caution coexist.
Source: thedecoder
