AI’s Biggest Players And Rising Stars |
AI is everywhere, so it seems. Just about every enterprise is using it in various ways to increase efficiency, accuracy and research, and the technology itself is approachable enough for anyone to explore new uses to improve their work. For the last eight years, Forbes has honored companies making great strides in developing the AI ecosystem with the AI 50 list. Our 2026 AI 50 launched today and features private companies spanning the AI universe. It includes behemoths that use AI to do almost everything, like OpenAI, and much smaller, newer companies with specific functions, like drug designing firm Chai Discovery, founded in 2024 by two former OpenAI staffers.
For the first time, we also published a companion list: the AI 50 Brink List. This list highlights the most promising startups that have completed Seed and Series A funding rounds. These companies, which are an average of 24 months old, are newer than their peers on the main list—but their ambition, inspiration and capacity to change the way AI is used is the same.
This is the published version of Forbes’ CIO newsletter, which offers the latest news for chief innovation officers and other technology-focused leaders. Click here to get it delivered to your inbox every Thursday.
Amazon is increasing its competition in the satellite internet space, now dominated by Starlink—a challenge from one of the few companies that can keep up with Elon Musk in spending. This week, Amazon announced it was buying satellite firm Globalstar for $11.57 billion. This acquisition, expected to close in 2027, will enable Amazon Leo's commercial satellite business to provide direct-to-device cellular service without antennas or cellular towers.
Starlink, which launched in 2019 and has more than 10,000 satellites in its constellation, has a big head start in this business. Amazon Leo is working to catch up, though, with more than 200 satellites in its constellation today. The Globalstar acquisition adds more satellites—plus significant expertise—to Amazon Leo, which is scheduled to officially launch to consumers later this year. According to TechCrunch, Amazon Leo is slated to have 3,200 satellites in orbit by the time the service becomes widely available. But even with fewer satellites, PCWorld reports Amazon is promising faster download speeds: up to 1 Gbps, while Starlink’s are between 45 and 280 Mbps.
Once Amazon Leo launches, it will have another distinct advantage: Amazon Web Services. AWS is the cloud infrastructure leader, with about 28% of total market share, according to Statista and Synergy Research Group. In February, Amazon announced a partnership between AWS, Amazon Leo and AT&T to strengthen connectivity, increase resilience and add coverage.
ARTIFICIAL INTELLIGENCE
The newest trend among engineers working with AI has nothing to do with function, platform or complexity. It’s all about spending lots of tokens—known as tokenmaxxing. Leading AI companies are celebrating employees who spend the most on AI tokens. Forbes contributor Tim Keary writes that more token use can be extremely pricey—a Meta engineer used 281 million tokens in 30 days, which could have cost more than $1 million.
But the cost is not a deterrent for company leaders. Nvidia CEO Jensen Huang said on a podcast that he’d be “deeply alarmed” if an engineer making $500,000 didn’t consume at least $250,000 worth of tokens in a year, writes Forbes’ Richard Nieva. The founder of AI real estate tools startup Alven, which had fewer than 10 employees, wrote on LinkedIn that the company spent $16,000 on tokens in February and planned to buy $60,000 worth in March.
But does this metric mean anything? Not really. AI token use doesn’t add up to getting more value out of AI—all it means is that an engineer is making AI do more. Companies that defend the practice told Keary that tokenmaxxing encourages employees to experiment more and get more use out of AI, which can naturally lead to greater value. And while it makes sense for companies not to limit tokens, the large bill that comes from tokenmaxxing might not be seen by investors as the best use of company money—especially because many AI companies aren’t yet profitable.
From a policy perspective, National Cyber Director Sean Cairncross told an audience at the Semafor World Economy conference in Washington, D.C., this week that the U.S. is ready to tackle cybercrime. Last month, the administration published President Donald Trump’s Cyber Strategy for America, laying out six areas in which the federal government plans to take action to combat cybercrime.
But, Cairncross added, Trump signed an executive order on the same day that directs different departments to work with members of the cybersecurity community to research how to improve domestic cybersecurity frameworks and present an action plan to improve defenses and better identify perpetrators; provide training to law enforcement; plan a cybercrime victims’ restitution program; and focus on prosections—both in the U.S. and abroad—of suspected cybercriminals. He said the administration is taking a “whole government approach” to cybercrime, making it a priority.
The strategy, he indicated, will also work to combat cybercrime from adversarial foreign countries—including Iran and China.
“What we are doing collectively as a federal government is saying, ‘How can we strategically approach this problem and maybe put these guys on the defensive,’ rather than sitting and waiting for them to act?” Cairncross said. “Because those response options, once something has happened, aren’t that great.”
Cairncross provided no specific initiatives or insights into what these departments were finding out or how these goals would be implemented. But he did say there will be more execution and action in line with the strategic goals “soon.”
Introducing the 2026 Forbes AI 50
In eight years, AI has gone from obscure to vital. Today, almost every company in almost every industry is using AI technology to be more efficient, write code and debug it faster, refine writing and ideas, and do deep research in the blink of an eye—among other functions. And in the last eight years, Forbes has compiled a list of the most promising privately held companies applying artificial intelligence to solve real-world challenges.
This year’s list, presented with sponsoring partner Mayfield, highlights the companies that are performing among the best in technical potential and rigor, corporate culture, company traction, market competitiveness and fundraising. The statistics are impressive: The 50 companies on the list have raised a combined $306.5 billion, although 80% of that belongs to behemoths OpenAI and Anthropic. The honorees represent all kinds of AI companies—from general purpose models and coding apps for music creators to data labeling and robotics models.
As AI technology advances and companies become larger—both financially and in their sphere of influence—competition for this year’s list was the tightest to date. Forbes put out a companion list this year: the AI 50 Brink List, highlighting 20 up-and-coming startups. These companies are, on average, about two years old and have completed Seed and Series A funding rounds. But the company’s ages have nothing to do with their vision and possibilities. Many of them are competing for top talent and approaching billion-dollar valuations. They’ve raised a total of $3.5 billion. Some of these are "neo labs": companies created to advance AI research, often founded by a former leader at a big tech company. And they may appear on future Forbes AI 50 lists—if they aren't acquired first.
Corporate services firm Vistra hired Damian Leach to be its new chief AI & digital officer, effective April 8. Shing previously worked as global CIO at Seaco.
Data and analytics company Alteryx appointed Julie Irish as its new chief information officer, effective April 7. Irish joins the firm from Couchbase where she worked in the same role, and she has also held leadership roles at New Relic and Harvard Business Publishing.
Application security provider Black Duck selected Dom Glavach as its new chief information security officer, effective April 9. Glavach joins the company from CyberSN where he served as chief information security officer and chief security strategist, and he spent two decades with Concurrent Technologies Corporation prior to that.
Steve Jobs is remembered as a visionary who could execute bold new plans—and tell the world about them in ways that were easy to understand. Here are some of his secrets to effective communication.
AI is technology for everyone, but there is a gender gap: Fewer women are using it, which could create problems as they climb the career ladder. Here’s what you need to know to encourage women to become comfortable with AI for the sake of leadership today and tomorrow.
Which company saw its valuation skyrocket after it announced earlier this week that it was abandoning its former line of business and pivoting into an AI company?
See if you got the answer right here.