Last week (14 February), GPU manufacturer NVIDIA disclosed in a filing that it held stocks in five companies linked to artificial intelligence (AI): ARM Holdings, SoundHound AI, Recursion Pharmaceuticals, Nano-X Imaging and TuSimple.
The news that Nvidia had invested in these five companies immediately boosted their share price, reflecting Nivdia’s leading position in the field and the deep respect it commands. Verdict examines how these companies are using AI and what they have done to draw the tech giant’s favour.
Recursion Pharmaceuticals
Nvidia holds around $76m in Recursion stocks, the news of which helped boost the company share price 30% from Monday 12 to Friday 16. The company had previously signed a deal to sell $50m of shares to Nvidia in July of 2023. Recursion uses machine learning and AI to discover drug candidates that might otherwise be missed as well as theoretically reducing lead time from discovery to commercialisation.
The statistics the company offers are impressive. Its website claims its automated lab robotics enables it to conduct “up to 2.2 million experiments each week for up to 50 weeks per year,” as well as generating and storing over 21 petabytes (each 1,024 terabytes) of “proprietary high-dimensional data”. The machine ‘learns’ through predicting drug candidates for a particular indication, synthesising the most promising ones and testing to see if they work. The results of these tests, whether positive or negative, inform the algorithm’s future decisions.
The one thing the company does not yet have is a marketed drug. It has numerous candidates in phase II or II/III trials, but setbacks have led the company to fall behind expectations. As Endpoints News noted last year, Recursion gave itself a goal of discovering 100 clinical-stage drug candidates in its first ten years. The total currently stands at four.
Nvidia’s investment is also not much compared to total sector engagement in 2023. Deals relating to AI were worth over $20bn last year while Recursion’s total market capitalisation is only $2.83bn, far below its $6bn peak in 2021.
Arm Holdings
Semiconductor manufacturer Arm is the largest company that Nvidia has invested in, with a nearly $132bn market capitalisation, following a huge rally in early February. It is also the company in which Nvidia holds the most investment at $147.3m. This is unsurprising given that Nvidia had hoped to buy Arm in a $50bn deal until it fell through in 2022 following concerns over competition.
The company went public following the deal’s failure, which was expected to be a consolation prize for its controlling interest SoftBank. Since then, however, skyrocketing demand for chips and the growing US-China trade war centred around them has led Arm’s stock to jump to nearly double since its initial valuation making it a very good deal and leading the Financial Times to critique the failure of the London Stock Exchange to entice the British company to float on it.
SoundHound AI
Vocal AI company, SoundHound, has perhaps benefitted the most from Nvidia's investment, with its stock climbing over 90% over the past month and growing from relative obscurity to a valuation of almost $1bn. The company itself has been around for almost 20 years, beginning in 2005 as a rival to Shazam which could identify songs based on an audio sample or a user’s own rendition.
It has since grown to offer AI voice recognition and generation across a range of products, from automated ordering in restaurants to what it calls ‘intelligent transcription’ – a tool that not only converts speech to text but also understands meaning, allowing it to identify conversation topics, according to its website.
The stock growth is particularly impressive given that the company made a $115m loss in 2022 according to GlobalData and “expects to continue to incur additional substantial losses in the foreseeable future”, per its Q3 2023 10-Q filing. The company has never turned a profit, although it did see record revenue during this period.
Nano-X Imaging
Nano-X was another big winner from Nvidia's investment being made public, with its stock jumping 67% since the announcement and nearly 90% over the course of the month. The company offers a range of medical imaging and diagnostics tools, including its flagship ARC tomosynthesis machine. Tomosynthesis works similarly to CT technology, taking x-ray images from a range of angles to produce a sharper and more detailed image.
Nano-X claims its machine is “less complex and smaller,” and though it is not clear exactly what it is less complex than, the touted flexible business model fits with the company’s aim to democratise imaging. One of the key selling points of Nano-X’s package is its cloud-based platform, which allows the compute-intensive process of compiling the separate X-rays into a single image to be performed off-site.
It is its AI platform however that has piqued Nvidia’s interest. Or rather, Zebra Medical’s AI platform. Bloomberg reports that Nvidia’s roughly $380,000 stake in Nano-X in fact comes from a deal arranged after the latter bought out Zebra in 2021. In exchange for its stock, Nvidia was granted shares in Nano-X, meaning it did not actively pursue the company. The platform itself works to offer early detection of asymptomatic chronic disease through algorithmic processing of CT data.
TuSimple
Autonomous trucking company TuSimple has not fared quite as well as the other companies benefiting from Nvidia's investment. While Nvidia’s report did boost the company’s stock 34%, it will likely do little to shore up a company that has experienced a nearly 99% loss in value since 2021, falling from a high of $62.60 to 47 cents. TuSimple announced its intention to delist from the Nasdaq in mid-January.
This is the latest in a string of changes and follows its decision to pull out of the US late last year following poor performance. Though founded in the country, its original creators are Chinese and that is where the company is now focusing operations.
The autonomous vehicles market has faced challenges across the board in recent years due to slow technology progression, regulatory hesitancy and safety concerns. Trucks face further hurdles due to their additional size and weight.