Semiconductor stocks experienced a broad decline on Tuesday, with Nvidia (NASDAQ: NVDA) and its peers coming under significant selling pressure after fresh concerns emerged about the financial health and growth trajectory of OpenAI, a key customer for the AI chip sector.
Nvidia shares fell approximately 1.5%, reversing gains from the previous session when the stock had closed at a record high. The selloff was widespread across the semiconductor sector. Advanced Micro Devices (AMD) dropped 3.1%, while Broadcom (NASDAQ: AVGO) declined 3.29%. Arm Holdings slid about 6.7%, with Intel, Micron Technology, and Applied Materials also posting notable losses.
The downturn comes on the heels of a strong rally for chip stocks, with the Philadelphia Semiconductor Index climbing nearly 50% from its late-March low, a surge fueled by optimism surrounding sustained artificial intelligence infrastructure spending.
OpenAI’s reported struggles shake the AI investment narrative. Investor sentiment shifted dramatically after a report by The Wall Street Journal revealed that OpenAI had missed its own internal targets for new user growth and revenue. The report also highlighted growing concerns among company executives about whether the firm can sustain its heavy spending commitments on data center infrastructure.
Chief Financial Officer Sarah Friar reportedly expressed concerns to leadership that the company might not be able to meet its future computing contract obligations if revenue growth does not accelerate. Additionally, board members have increased their scrutiny of data-center deals and have questioned CEO Sam Altman’s aggressive strategy to secure additional computing capacity. In response, OpenAI acknowledged its investment stance, stating it was 'buying as much compute as we can.'
This is particularly alarming for chip stocks because Nvidia, AMD, and Broadcom all have supply agreements with OpenAI, making them highly sensitive to any changes in the company's financial outlook. Nvidia has also made significant financial commitments, including a $30 billion investment in OpenAI’s latest funding round, though it scaled back an earlier plan to invest as much as $100 billion.
Adding to the pressure, OpenAI is facing increasing competition from rivals such as Anthropic, which is gaining traction in coding and enterprise markets, and AI models from Alphabet, which have also received strong reception. In a joint statement, Altman and Friar pushed back against the narrative, calling any suggestion that they are divided or pulling back on securing new computing resources 'ridiculous.'
Profit-taking adds to pressure ahead of Big Tech earnings. Market participants also pointed to profit-taking as a significant factor behind the decline, given the sector's strong recent performance. 'This morning’s moves in individual stocks indicated some profit-taking across semiconductors, which seems reasonable given their incredible run since the end of March,' wrote David Morrison, senior market analyst at Trade Nation.
The pullback comes ahead of a crucial earnings period for major technology companies like Alphabet, Microsoft, Amazon, and Meta Platforms, which is expected to provide further clarity on AI spending trends. Collectively, large US technology firms are expected to commit more than $700 billion toward infrastructure investments this year, a figure that underpins the bullish case for semiconductor demand. Chip-stock investors will be looking to these reports for reassurance that spending plans remain intact.
Despite the selloff, the long-term outlook for Broadcom, which is carving out a niche with custom ASIC chips for hyperscalers like Google and Meta, remains strong. Broadcom’s last quarter showed AI revenue reaching $8.4 billion, up 106% year-over-year, and the company has a $73 billion AI backlog. CEO Hock Tan stated the company has a 'line of sight' to over $100 billion in AI chip revenue by 2027. However, the immediate market reaction was driven by fears that OpenAI’s financial troubles could signal a slowdown in the overall AI spending boom.
