A cluster of AI startups is now hitting each revenue milestone in less time than it took to hit the last one, a pattern that has become the defining financial signature of the 2026 AI market. Mercor, the domain-expert hiring platform, crossed $2 billion in gross annualized revenue in June, just four months after reaching $1 billion, according to co-founder and CEO Brendan Foody. The company, less than three years old, was at a $500 million run rate as recently as September.
The acceleration is not confined to one company or one business model. Anthropic reported $47 billion in revenue run rate in late May, less than two months after the same metric passed $30 billion. That $47 billion figure is up from $9 billion in late 2025 and just $4 billion in July 2025 — an order-of-magnitude climb in under a year.
Sierra, Bret Taylor's customer service AI agent company, took seven quarters to reach its first $100 million in ARR. It took two more quarters to add the next $100 million. Glean, a seven-year-old enterprise search and AI startup, crossed $300 million in ARR in May. The prior double, from $100 million to $200 million, took nine months; the next $100 million took six.
Key facts
- 01Mercor hit $2B in gross annualized revenue in June, four months after crossing $1B and nine months after a $500M run rate.
- 02Anthropic reached $47B revenue run rate in late May, up from $30B less than two months earlier and $4B in July 2025.
- 03Sierra took seven quarters to reach $100M ARR and just two more quarters to add another $100M.
- 04Glean went from $100M to $200M ARR in nine months, then $200M to $300M in six.
- 05Clio, an 18-year-old legal software firm, went from $200M ARR in mid-2024 to $500M after embedding AI in 2023.
The compression is the story. Traditional SaaS benchmarks — Rule of 40, T2D3 growth curves, the idea that doubling gets harder as the number gets bigger — have been the industry's shared vocabulary for a decade. What these companies are reporting inverts that intuition: doubling is getting easier, not harder, and the gap between milestones is shrinking rather than widening.
One caveat worth surfacing up front: the underlying metric varies across the group. Some companies report annualized recurring revenue from paying customers. Others report annualized run-rate revenue, which projects the most recent month over twelve. Others include committed ARR from signed but not-yet-onboarded customers. Gusto, notably, reported actual trailing twelve-month revenue rather than any run-rate figure. The direction of travel is consistent even if the definitions are not.
Gusto's numbers matter because the 14-year-old HR software company is not AI-native. It reported that revenue accelerated in each of the last five quarters and that trailing twelve-month revenue crossed $1 billion. The company was last valued at $9.3 billion in early 2022. The signal is that AI integration is showing up in the top line at legacy SaaS businesses, not just at model labs and agent startups.
Clio is the clearest version of this pattern. The 18-year-old legal practice management vendor embedded AI into its product in 2023, hit $200 million in ARR in mid-2024, doubled to $400 million by late last year, and recently announced $500 million. The AI wrapper thesis — that mature vertical software companies can re-accelerate by adding AI features to sticky customer bases — has a case study.
The counterweight is that run-rate revenue is a fragile number. It assumes the most recent month persists for twelve. In markets where enterprise buyers are experimenting rapidly and where switching costs between model providers remain low, a single quarter of churn can compress a $47 billion run rate materially. None of the companies on this list have been through a full renewal cycle at their current scale, and gross margins across the model layer remain contested.
There is also a definitional risk. When Mercor reports gross annualized revenue and Anthropic reports revenue run rate and Gusto reports trailing twelve-month revenue, direct comparison is impossible. Investors and reporters are using "ARR" as shorthand for very different underlying cash flows. A $2 billion gross annualized number and a $2 billion trailing revenue number are not the same asset.
What the pattern signals for the AI market is that customer demand is currently outrunning the standard SaaS playbook for measuring it. If Anthropic's run rate holds and Mercor's next milestone lands in another four months or less, the industry will need new benchmarks — the T2D3 growth curve does not describe what is happening here. If the run rates compress on renewal, the same set of companies will define the sector's first serious revenue reset. Either way, the multiples being paid on these numbers assume the acceleration continues, and the first company to break the pattern will reset the entire comp set.
Working on something we should cover, or seeing a story we missed? Send leads, documents, or feedback to hello@aichatdaily.com. For sensitive tips, see our secure tips page for Signal and PGP options.
Spotted an error? Email hello@aichatdaily.com with the URL and the issue, or read our full corrections policy.
