Applied AI companies are moving beyond chips to solve real-world problems in voice, power, and delivery.
Revenue growth ranges from 46% to 217%, though all three companies remain unprofitable.
High risk meets high reward as these stocks target massive markets with early-stage technologies.
Artificial intelligence (AI) isn't just about graphics processing units anymore. The next wave of winners will come from companies that apply AI directly to daily life, power the massive energy demands of AI data centers, and utilize autonomy to slash delivery costs.
That's why three very different names -- SoundHound AI (NASDAQ: SOUN), Navitas Semiconductor (NASDAQ: NVTS), and Serve Robotics (NASDAQ: SERV) -- deserve a closer look. Read on to find out more about these three top AI stocks.
Where to invest $1,000 right now? Our analyst team just revealed what they believe are the 10 best stocks to buy right now. Continue »
Image source: Getty Images.
SoundHound has carved out a niche as the leading independent voice and conversational AI platform, with deployments across restaurants, automakers, and customer service centers. The company's technology powers everything from Stellantis vehicles to White Castle drive-thrus, processing millions of voice interactions monthly.
Revenue growth has been explosive of late. Second-quarter 2025 sales hit $42.7 million, up 217% year over year, prompting management to raise full-year guidance to $160 million to $178 million. That's a promising sign that the AI voice pioneer's platform is indeed gaining traction in the marketplace.
Key catalysts include automotive manufacturers expanding embedded voice features and restaurant chains rolling out AI-powered drive-thru systems to combat labor shortages. The recent Amelia Technologies acquisition adds enterprise AI agents, expanding the addressable market.
Risks include ongoing cash burn and competition from tech giants, but for investors seeking operational AI exposure beyond chips, SoundHound offers high-growth leverage to voice interfaces.
Navitas designs gallium nitride (GaN) and silicon carbide (SiC) power semiconductors that dramatically improve efficiency in data center power supplies. The company claims its GaN technology can reduce data center power consumption by up to 40%, versus traditional silicon, which is critical as AI training pushes facilities toward gigawatt scale.
Second-quarter revenue hit $14.5 million, down significantly (29.2%) from the same period a year ago, as the company pivots from consumer electronics to AI and automotive markets. Management recently unveiled platforms for hyperscale power units and secured $100 million in fresh capital. With the GaN power-device market projected to reach $4.4 billion by 2030, design wins with hyperscalers could transform the narrative overnight.
The risks are clear and include a tiny revenue base, ongoing losses, and dilution from recent financing. But Navitas offers early-stage leverage to AI's power bottleneck.
Serve Robotics brings AI to the sidewalk. Spun out from Uber in 2021, the company deploys autonomous robots for last-mile delivery in Los Angeles, Atlanta, and Chicago. Second-quarter 2025 revenue reached $642,000, up 46% sequentially, with delivery volume surging 80% quarter over quarter.
With $183 million in liquidity, Serve targets deploying 2,000 robots by year-end, up from roughly 100 today. Management estimates each robot can generate $48,000 in annual revenue at maturity, creating a massive tailwind as the company starts to deploy its fleet across the country.
This is the speculative play of the group -- revenue remains negligible, regulatory approval varies by city, and competition from Amazon and drone start-ups looms large. However, if Serve cracks the low-cost autonomous-delivery market, the addressable market could render today's $623 million valuation a rounding error.
These three stocks offer different angles on AI's real-world deployment. SoundHound brings voice technology to industries desperate for automation. Navitas provides the power infrastructure AI can't scale without, and Serve represents a bet on autonomous systems replacing human labor.
All three companies are unprofitable and carry significant execution risk. But for investors willing to venture beyond Microsoft and Nvidia, they represent pure plays on AI's next chapter -- where technology moves from research labs to restaurant drive-thrus, data center power supplies, and city sidewalks.
Before you buy stock in SoundHound AI, consider this:
The Motley Fool Stock Advisor analyst team just identified what they believe are the 10 best stocks for investors to buy now… and SoundHound AI wasn’t one of them. The 10 stocks that made the cut could produce monster returns in the coming years.
Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, you’d have $661,220!* Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,114,162!*
Now, it’s worth noting Stock Advisor’s total average return is 1,069% — a market-crushing outperformance compared to 185% for the S&P 500. Don’t miss out on the latest top 10 list, available when you join Stock Advisor.
See the 10 stocks »
*Stock Advisor returns as of August 25, 2025
George Budwell has positions in Microsoft, Navitas Semiconductor, Nvidia, and Serve Robotics. The Motley Fool has positions in and recommends Amazon, Microsoft, Nvidia, Serve Robotics, and Uber Technologies. The Motley Fool recommends Stellantis and recommends the following options: long January 2026 $395 calls on Microsoft and short January 2026 $405 calls on Microsoft. The Motley Fool has a disclosure policy.