This article was updated on July 10, 2017, and originally published on Feb. 28, 2017.
NVIDIA (NVDA 1.40%) was one of the hottest tech stocks of 2016 and has continued that run into 2017 with a share price jump of 194% over the past 12 months. The company makes the vast majority of its revenue from gaming -- about 53% in the fiscal first-quarter 2018 -- but NVIDIA is much more than just a gaming processor company.
The artificial intelligence (AI) market is quickly expanding, and NVIDIA is positioning itself to make big gains in the space. According to the company's own investor presentation published back in May, NVIDIA has six key AI opportunities that it will benefit from between now and 2025.
Take a look:
Data source: NVIDIA.
NVIDIA's total addressable market for AI is about $39 billion between 2020 and 2025, when you combine all of the above categories together.
Just for reference, NVIDIA defines deep learning training as a computer "learning a new capability from existing data" and deep learning inference as "applying this capability to new data."
The company is already tapping into these AI markets by using is dominant position in the graphics processors chip space. According to Goldman Sachs' Toshiya Hari, NVIDIA already holds nearly 90% of the market for chips that are used for training tasks in machine learning (which is part of the broader Ai market).
NVIDIA's DGX-1 supercomputer uses machine learning to process information faster than previous deep learning machines of its kind, and is currently being used by SAP for enterprise solutions for 320,000 customers. Additionally, the company said that it's latest Pascal chip architecture is "purpose-built for AI" and is already being used by IBM and Microsoft for some of their AI services as well."
In the self-driving car space, NVIDIA's AI opportunity comes from its Drive PX 2 supercomputer, which uses deep learning to process image information for driverless cars and helps them decide how they should react in certain situations.
About 225 automakers, Tier 1 automotive suppliers, researchers and start-ups already use a version of Drive PX, including Tesla. The electric-auto maker is rapidly moving toward a self-driving car future and has already added key driverless car technologies -- including NVIDIA's computers -- to make it a reality. And NVIDIA has already partnered with several large automakers this year to make their driverless car ambitions a reality.
Finally, NVIDIA' AI city opportunities include everything from traffic management to retail analytics to surveliinance and forensic analyzation. The company is using its GPUs to process things like security camera footage AI image analytics.
A growing opportunity
While NVIDIA's AI opportunity is huge, investors should know one thing: AI makes up only a small percentage of revenue right now.
In the fiscal first quarter 2018, NVIDIA brought in just $409 million from its data center division -- 21% of total revenue-- and even its automotive business (which includes the Drive PX 2 AI supercomputer) brings in about 7% of total revenue.
That means the company will need to continue releasing new hardware and adding more customers to reach its full AI potential. But at this point, NVIDIA is already well on its way to fully tapping into the AI market, and it's doing so at the perfect time.
This year may be one of the best years to invest in AI, as tech companies and governments around the world are ramping up investments in artificial intelligence. NVIDIA's current position in AI, potential market share, and focus on new AI hardware should push the company to the top of the list for anyone looking to invest in AI over the long term.
