War of the Chips
Who will win? NVIDIA? GOOGLE?
There are more as well, including
Leading AI chip designers
Considered the market leader, primarily known for its high-performance GPUs like the H100 and Blackwell, which are crucial for AI model training and inference.
AMD:
A major competitor to Nvidia, offering its own line of AI accelerators, including the Instinct series.
Recently announced new AI accelerator chips to compete with established players like Nvidia and AMD.
Big tech companies developing their own chips
Designs its own Tensor Processing Units (TPUs) for AI tasks in its cloud services.
Develops its own AI chips, including Inferentia and Trainium, for its AWS cloud offerings.
All of these large platforms have subsidaries owned by them, such as Mediatek, TSMC is a foundry which manufactures the wafer silicon discs and forges the chips. They are akin to a large steel forging plant, producing the basic silicon.
Creates its own AI accelerators for its Azure cloud platform.
Meta:
Designs custom chips, such as the MTIA, to support its internal AI initiatives.
Develops custom silicon, like the M3, M3 Pro, and M3 Max processors, with capabilities for AI and machine learning on its devices.
Other notable companies
A major player in the custom AI processor space, a role complemented by Nvidia’s dominance in GPUs.
Another key competitor in the semiconductor industry, though Nvidia currently holds a leading position in the AI chip market.
TSMC:
The world’s largest contract chip manufacturer, it produces advanced chips for many of the companies listed above, including Nvidia, AMD, and Apple.
A major chip manufacturer that has secured large deals to produce AI chips for companies like Tesla.
NVIDIA of course has a significant lead on all of these manufacturers. However their chips are focused on their specific industry. NVIDIA is a generic GPU developed from their very successful gaming chip.
Who will dominate this niche.