Investor sentiment toward artificial intelligence and big tech has taken a hit following the initial success of China-based DeepSeek’s large language models, V3 and R1.
A report by UBS highlighted that these models are reportedly trained at significantly lower costs, offering deep discounts on inference, raising questions about the high capital expenditure (capex) intensity of established tech giants.
The report said, “Market sentiment on AI and big tech has turned negative following the initial success of China-based DeepSeek’s V3 and R1 large language models.
“DeepSeek’s aggressive pricing strategy has put pressure on big tech companies, sparking fears of an impending price war in the AI sector. However, the UBS report points out that while DeepSeek has made an impressive entry into the market, its models still have significant limitations when compared to frontier models developed by leading technology firms.
Additionally, there is limited transparency around DeepSeek’s training methodology, which raises questions about its long-term potential and scalability.
The report further stated that investors are now grappling with broader concerns, including the high capex requirements of the “Big 4” tech companies and how to allocate investments across the various layers of AI development. This includes focusing on enabling technologies versus application-driven solutions.
The report also advised investors to remain cautious and avoid overreacting until more clarity emerges about DeepSeek’s operations and its business model. The report suggested focusing on the upcoming quarterly results of major tech companies for better insights into how they plan to address this new competition and sustain their dominance in the AI space.
While DeepSeek’s pricing strategy challenges the status quo, its overall impact on the AI ecosystem remains uncertain. The report recommended a measured approach, emphasising that the market needs more concrete data before concluding the potential shift in dynamics. As the tech giants gear up to respond to this challenge, the next few weeks will be crucial in shaping the future landscape of AI investments and competition.
