Investors: 2 Tiny AI Stocks You Might Be Overlooking

Artificial intelligence (AI) grabbed the corporate world's attention in 2023, coming to the forefront seemingly from nowhere. Some of the largest companies in the world have been working on the technology for years, but they had yet to produce a mainstream product that captivated consumers.

Then OpenAI unleashed its ChatGPT chatbot platform in November.

ChatGPT attracted 1 million users in the first five days, and it crossed the 1 billion mark in March. Now, the race is on in the corporate sector to apply AI to more products and services, which is shining a light on previously little-known companies.

This article will focus on two AI companies that most investors have probably never heard of. They each have market capitalizations under $1 billion. That means they might carry more risk than most tech stocks, but the upside could be significant if they successfully execute their strategies. 

1. SoundHound AI

SoundHound AI (SOUN -5.35%) is a specialist in conversational artificial intelligence, or the type of AI with the ability to recognize spoken language. It can also respond with voice (as opposed to text, like ChatGPT), which unlocks a whole host of enterprise opportunities.

For example, restaurants can use SoundHound's Smart Answering tool to take incoming phone calls, reducing the burden on employees to answer simple questions from customers relating to operating hours, menu items, and promotions. Similarly, the Smart Ordering tool is a voice assistant capable of accepting orders from customers over the phone. It integrates with leading point-of-sale systems like Block's Square, Toast, and Oracle. 

But that's not all. SoundHound has attracted a major-league customer list, which includes automakers Hyundai, Kia, and Mercedes-Benz, plus media and entertainment companies like Netflix and Snap. SoundHound estimates 90% of new cars will have embedded voice assistants by 2028, and its technology is already helping to make the driving experience more intuitive by powering voice-commanded features. 

To accelerate the improvement of its AI models (in other words, help them learn faster), SoundHound has developed a new technology that allows for integrations with OpenAI and ChatGPT.

Although SoundHound was founded in 2005, it has only traded as a public company for 12 months. It generated $31.1 million in revenue last year, but it had an order backlog worth a whopping $331 million, which is expected to convert to revenue over time (it has an average contract length of 6.5 years). That bookings figure grew 59% compared to 2021, so there is a clear demand pipeline forming from customers.

SoundHound is currently valued at $585 million, which appears cheap against the growth of its order backlog. But it does come with risk because the company lost $115 million in 2022 and faced a cash crunch at the end of the year. It has since secured a financing deal worth up to $125 million, which is great news, but investors should be aware the company's net losses are likely to persist for a few more years.

2. BigBear.ai is chasing a $310 billion opportunity

BigBear.ai (BBAI -1.33%) provides AI-powered solutions to businesses and the U.S. government, designed to help them analyze and draw actionable insights from their data. It uses machine learning to create digital simulations so its customers can game out key decisions, observing the impacts of different scenarios before investing significant amounts of time and money on a final real-world project. 

For example, Baystate Health avoided a $1.2 million expansion of its emergency department after running simulations on the effects. Instead, thanks to BigBear.ai, the hospital managed to reduce the length of stays in its existing emergency ward by 15%, and by 33% in general admission. Similarly, $32 billion chemical giant DuPont used simulations to avoid $480,000 in capital expenditures relating to its rail logistics network. 

Back in January, BigBear.ai secured a 10-year, $900 million contract with the U.S. Air Force, allowing the company to compete for task orders to deliver systems and synthetic environments to critical missions. BigBear.ai is a prime contractor, meaning it will work directly with the government and assume full operational responsibility for the projects it's working on. 

BigBear.ai is valued at just $414 million as of this writing. As is the case with many emerging companies, its annual revenue is incredibly lumpy -- in 2020, it generated just $31 million, which soared by 367% to $145 million in 2021. In 2022, that growth slowed to just 6% with $155 million in revenue coming through the door. In any case, all of those figures represent a mere fraction of what BigBear.ai believes will be a $310 billion opportunity in AI and machine learning by 2026. 

But the company carries significant risks. It isn't profitable, and it logged a net loss of $121 million in 2022. That's a major issue because its cash balance is dwindling; at the end of last year, it had just $12 million on its balance sheet. It successfully closed a $25 million private placement in January to shore up its financial position, which is great, but that's not a long-term solution.

BigBear.ai reports its 2023 first-quarter earnings results on May 9, and investors should clarify the company's strategy to raise capital before buying the stock. If it can find enough cash to last the next 18 to 24 months, that could be enough runway to unlock more growth and, therefore, strong share-price gains for investors.