Elad Gil: The AI Markets With Winners vs. Wide-Open Opportunities
▼ Summary
– Elad Gil considers AI the least predictable tech boom, with much of the market still open for competition despite some areas having clear leaders.
– He began investing in generative AI in 2021 after observing the significant capability leap from GPT-2 to GPT-3, which indicated its future importance.
– Gil invested in both foundational model companies like OpenAI and Mistral and application firms such as Perplexity, Harvey, and Abridge.
– He identifies foundational models, AI-assisted coding, medical transcription, and customer support as markets with established leaders that are hard to catch.
– Gil views financial tooling, accounting, and AI security as wide-open markets, noting that rapid enterprise adoption doesn’t guarantee long-term success for startups.
Venture capital investor Elad Gil recently observed that the artificial intelligence sector presents a dual reality: some markets have already crowned definitive leaders, while others remain completely up for grabs. Speaking at a major technology conference, the influential backer of numerous successful tech firms described AI as one of the most unpredictable technological expansions he has witnessed. His investment history includes foundational AI model creators and a range of application-focused companies, giving him a broad perspective on the industry’s rapid evolution.
Gil began placing bets on generative AI back in 2021, a time when the field attracted relatively little attention. What convinced him was the dramatic performance jump between GPT-2 and GPT-3. He reasoned that if this steep improvement trajectory continued, the technology was destined to become profoundly significant. This insight led him to fund early-stage startups leveraging large language models, including both infrastructure players like OpenAI and Mistral, and application-layer companies such as Perplexity, Harvey, Character.ai, Decagon, and Abridge.
Reflecting on the market’s inherent unpredictability, Gil noted, “AI was the one market where the more I learn, the less I know.” Typically, deeper knowledge of a sector allows for better forecasting, but AI has consistently defied this pattern due to its immense complexity and pace of change. He believes certain segments of the AI landscape still operate with this high degree of uncertainty.
However, he now identifies several areas where winners have clearly emerged. The market for foundational models is a prime example. Despite a global proliferation of models, including national efforts in countries like South Korea, Gil predicts the field will be dominated by just a handful of players. He names Google, Anthropic, OpenAI, and potentially xAI, Meta, and Mistral as the likely long-term leaders.
Another sector showing clear front-runners is AI-assisted coding. This space is not only crowded with offerings from foundational model companies like Anthropic’s Claude Code and OpenAI’s Codex, but also features specialized startups such as Anysphere’s Cursor and Cognition’s Devin that have built a formidable lead. Gil also pointed to well-funded companies like Magic, which he considers a potential “outlier,” and Poolside as serious contenders.
In medical transcription, Gil sees the market as effectively cornered, with Abridge leading the pack and a few other players like Ambience maintaining important positions. The customer support sector, an early target for AI automation, also appears to have hard-to-catch leaders. His portfolio company Decagon is a major player in this space, having secured a $1.5 billion valuation. They compete with Sierra, a startup from OpenAI chairman Bret Taylor, while established giants like Salesforce and HubSpot are aggressively integrating AI into their existing platforms.
So where do the wide-open opportunities lie? Gil highlights several fields that appear ripe for disruption. He points to financial tooling, accounting, and AI security as markets with immense potential but no established winners yet. These are domains where the inherent value is obvious, but the companies that will ultimately dominate have not yet been determined.
An interesting shift Gil notes is that rapid revenue growth no longer reliably signals a future breakout success. Corporate leaders worldwide have issued mandates to adopt AI, leading large enterprises to experiment with new solutions they would have previously ignored. This environment allows new AI companies to quickly secure substantial contracts from major clients. However, Gil cautions that this early revenue does not guarantee long-term customer retention.
The true test comes after the initial trial-phase boom. It is only then that startups and their investors can discern whether the early revenue signals a sustainable business or just a temporary spike. Gil distinguishes between “false signal” and ventures that are “just working.” He cites legal AI startup Harvey as a prime example of the latter. The company experienced explosive growth in 2025, skyrocketing from a $3 billion valuation to $8 billion in just a few months through three successive funding rounds, demonstrating the powerful traction possible in these emerging AI markets.
(Source: TechCrunch)
