Sequoia Capital just put $45 million behind Sable, a startup selling Aidan as an AI employee for sales teams. That's the bet: not better software, but software that takes the seat.
Sable makes Aidan, a system it markets as "the first AI employee powered by real time browser use and vision." Not a chatbot. Not a plugin. Aidan is supposed to run product demos, qualify customers, and walk new users through onboarding without a human sales rep on the call.
According to Fortune, Sequoia led the $45 million investment into the San Francisco startup. That figure is the story's hard center, because Sable isn't pitching a dashboard a salesperson uses between meetings. It is pitching the meeting itself.
Aidan learns a company's product and sales script from the people already on staff, then uses browser control and vision to click through the interface while explaining what it is doing. Sable says it can deploy in days and doesn't require engineering integration. That is the sale to a VP of revenue: don't wait six weeks for implementation, don't hire another junior rep, and don't leave every routine demo on a human calendar.
Shaun Maguire, the Sequoia partner behind the deal, is not a cautious investor. He led the firm's bets on xAI and helped steer its stake in Bridge, the stablecoin company Stripe bought for $1.1 billion last year. Fortune quoted Maguire saying Sable has created a "direct path from frontier capability to how businesses actually reach their customers." Joe Lonsdale, the Palantir co-founder whose firm 8VC also backs Sable, put his name behind the round too.
Look, that language is grand. The product is narrower. Aidan is not replacing every salesperson. It is trying to take the repetitive, high-volume work that already feels scripted: the twentieth demo of the week, the onboarding walkthrough with the same buttons, the qualification call where the questions barely change.
Sequoia is buying outcomes, not tools
This fits the way Sequoia has been talking about AI all year. Business Insider reported in March that Sequoia partner Julien Bek told founders they should worry if they are "just an iteration away" from a model replacing what they do. His answer was blunt: sell services and outcomes, not tools that sit in the path of the next model release.
Sable is that thesis with a name badge. You don't buy Aidan so your sales team can click faster. You buy it because Sable says Aidan can handle part of the job. That is a much cleaner pitch than most AI software companies make, and frankly, it is also a more dangerous one. If the product works, the budget comes from headcount as much as software. If it fails, buyers will notice in the most public place possible: in front of customers.
Sequoia has already seen this movie in customer service. Sierra, the AI support company founded by Bret Taylor and Clay Bavor, became one of the clearest examples of enterprise buyers paying for agents that take over customer interactions rather than merely summarize them. Aidan points the same idea at sales, where the upside is obvious and the tolerance for awkward failure is much lower.
Aidan still has to survive real calls
The hard question is whether Aidan holds up when the demo is no longer staged. Demos are staged. Real calls aren't. Vision-based agents that click through live software can break on ugly details: a redesigned menu, a permissions pop-up, a slow-loading page, a customer who asks to jump three screens ahead.
A trained rep handles those moments by improvising, stalling, asking a follow-up, or admitting the product needs a human answer. Sable's whole pitch depends on Aidan doing enough of that, every time, across products it did not help build and customers it has never met. That is not a small technical claim. It is the company.
Sequoia's check says the firm believes the reliability is close enough to sell now. Enterprise buyers will make a harsher judgment. They won't care whether "AI employee" becomes a category if Aidan loses a prospect inside a broken onboarding flow.
Sable has the funding now. It also has a short clock. The company doesn't need to prove AI can replace every salesperson. It needs to prove one AI worker can take a real slice of a sales team's week and not embarrass the buyer while doing it.
Also read: A Leaked Kimi K3 Promo Page Shows Moonshot Chasing an Edge Claude Already Has • A Former Red Bull F1 Engineer Just Raised Germany's Biggest Seed Round Ever • Jason M Riggs on Why AI Spending Keeps Outrunning AI Leadership
