Elad Gil That there are winners in the AI ​​markets – which are still wide open

said investor extraordinaire Elad Gil at Solo VC On stage at TechCrunch Disrupt AI has been one of the least predictable technological breakthroughs ever.

Generation is at the cap table for almost every successful company of the past decade, including many of today’s leading AI companies.

However, he believes that over the past year, some AI markets appear to have been almost completely sewn up by market leaders. Beyond these areas, a wide range of AI remains anyone’s game.

“I started investing in generative AI in 2021… At the time, not a lot of people were paying that much attention to it,” Gill said. But he saw the huge jump in capabilities between GPT 2, It was launched in 2019and GPT 3, were released In 2021. “The step between 2 and 3 was so large that if you extrapolated the scaling laws, or the curve, you could really assume that this would be very significant,” he said.

This convinced him to start supporting early-stage startups to build products powered by large language models. His bets included both established model makers like OpenAI and Mistral, as well as application companies like Perplexity, Harvey, Acter.ai, Decagon, and Abridge. However, throughout 2024 and most of 2025, the capabilities of the foundational models jumped with each release, turning AI on its head every few months.

“I used to say at the time that AI was the one market where the more I learned, the less I knew about it,” he said. “Normally, the more you learn about something, the better you know it, the easier you can predict the future, etc. But AI was just nebulous. There is a lot of uncertainty. I think there are still markets like that in AI.”

However, he now also sees markets with clear winners. The most obvious example is the foundational models themselves. Although there are hundreds of models, some countries like South Korea exist It is still working on developing sovereign models By local companies, leaders emerged. “Google, Anthropic, OpenAI, maybe xAI, maybe Meta, maybe Mistral — it’s like a bunch of winners,” the winners predicted.

Following models, AI-assisted programming is believed to have runaway winners making it difficult for new entrants to catch up. Not only foundational model makers have moved on (Anthropic with Claude Code, OpenAI with Codex) but also startup leaders like Anysphere indicator Devine perception (which acquired Windsurf) It will be difficult to beat him. There are well-funded startups Like Magic (which Gill described as potentially “weird”)) or Poolside on their tails.

He sees that medical transcription is under siege, With Abridge in first place And a bunch of others like Ambiance Being “important”.

He notes that customer support—an early goal of both traditional AI and the new crop of AI startups—is having market leaders that are difficult to catch, like his portfolio company. Decagon. (It has been uploaded $131 million at a valuation of $1.5 billion In June.) President of OpenAI Brett Taylor’s startup, Sierracompeting in this space. This is also an area where established companies — Salesforce, Hubspot, and many others — are adding AI offerings.

So which markets seem wide open? Financial instruments (fintech), accounting, AI security, and “other markets that we know are by default very interesting. But we don’t know who will do it,” says Gill.

Ironically, rapid growth is not the sign that a company was going to be a huge success that once was. “CEOs at every big company are saying to their teams, ‘Hey, we have an edict. We need to figure out our AI strategy,'” Gill said. “These giant companies are willing to try things they wouldn’t have tried two years ago, and it’s only because of AI.”

So, new AI markets can generate a lot of revenue from big-name corporate clients quickly, “but that doesn’t mean they’ll stick around,” Gill points out.

Only after the market goes through the boom cycle in the beta phase can startups and investors know if these revenues will survive and grow. “There’s a wrong signal, and then there’s things that just work,” Gill said. He describes legal AI startup Harvey as one of the market leaders that “just works.” It raised three massive rounds in 2025, jumping from A $3 billion valuation to $5 billion to 8 billion dollars, In just a few months.