Nvidia has dominated the market for AI accelerators since the ChatGPT wave in late 2022, thanks to its strong foundation in accelerated computing and CUDA, a proprietary software toolkit. However, with tech giants and start-ups moving fast to develop their own custom AI chips, Nvidia’s dominance is eroding.
The company’s data center segment has been particularly successful, generating over $35 billion in revenue during the latest quarter. But as customers like Microsoft, Meta Platforms, and OpenAI spend billions of dollars on Nvidia’s GPUs, the math is changing. Companies are now considering developing their own AI chips to save costs and stay competitive.
Amazon and Alphabet have already created custom AI chips, while others are following suit. Meta is reportedly testing a custom AI chip for AI training workloads, which are more computationally intensive than AI inference workloads. This trend could put pressure on Nvidia’s pricing and margins, as well as reduce demand for its chips.
Nvidia faces competition from AMD in the AI accelerator market, but its own customers represent a bigger threat. As tech giants rethink their AI hardware and software stacks, they may choose to design their own custom chips instead of relying on Nvidia’s GPUs. This could lead to a decline in Nvidia’s market share and growth as demand cools down.
Nvidia stock has already slumped from its all-time high due to concerns about the resiliency of AI demand and global economic worries. As custom AI chips gain traction, the stock may face additional pressure.
Source: https://finance.yahoo.com/news/heres-nvidias-incredible-ai-success-144500116.html