AI-related stocks experienced a significant downturn on Monday following calls from prominent AI leaders, including Anthropic CEO Dario Amodei and OpenAI CEO Sam Altman, for a slowdown in the pace of AI development due to safety concerns. This has rattled investors who fear an industry-wide slowdown could have widespread ripple effects, impacting everything from chip purchases to computing power, and ultimately threatening the impressive fund flows into AI-related assets.
Global chip stocks were particularly hard hit. In Asia, South Korean heavyweights SK Hynix and Samsung Electronics closed down more than 6% and 4% respectively, while SoftBank, a major investor in OpenAI, saw its shares plummet 10% in Japan. European semiconductor companies also suffered, with chip equipment giant ASML falling over 4%, Nokia down around 5%, and Infineon dropping more than 6%. Companies tied to data center buildouts, such as Siemens Energy and Schneider Electric, also saw declines.
In premarket trading in the U.S., memory chipmaker Micron was down approximately 5%, Intel dropped nearly 6%, and Nvidia, a key player in AI hardware, was more than 2% lower. Hyperscalers like Microsoft, Amazon, and Alphabet also experienced slight dips. The sector-wide outflow was notable, with Information Technology leading outflows at $1.53 billion for the day.
Financial analysts are expressing concern that a slowdown in AI development could derail the equity market rally, which has largely been fueled by anticipated AI growth and productivity gains. Zoe Gillespie, a senior director at RBC Brewin Dolphin, warned that if the expected earnings growth of these companies comes under threat, it could destabilize overall equity performance. While Amodei emphasized that progress would still seem fast and not a complete halt, the immediate market reaction reflects investor apprehension regarding the future trajectory of AI adoption and its financial implications.
Despite the immediate market reaction to the potential slowdown in AI development, some analysts suggest that the demand for AI inference, which is the use of trained AI models, still far outstrips supply. This implies that even if the training and rollout of new AI models are slowed, company revenues, particularly in areas addressing this inference demand, may not be significantly impacted.
