KorvaJoin the waitlist
Startups

Anthropic May Hit $100 Billion in Sales With a Twelfth of Google's Staff

The gap reflects heavy spending on AI computing, and fuels a debate over what it means for jobs.

AI image
The Twenty Minute VC (20VC): Venture Capital | Startup Funding | The Pitch · 20VC: Will Jensen Return to NVIDIA | Anthropic Set for Largest IPO of All Time | Kalshi, Mercor & Fractile: Mega-Rounds Continue

Anthropic is expected around Q1 to reach a $100 billion revenue run rate with fewer than 6,000 employees, in investor Jason Lemkin's rough estimate.

At the same $100 billion sales mark, he noted, Google had about 81,000 employees and Meta had about 71,000. That is roughly one-twelfth the staff for the same sales.

Many people say that comparison is no big deal, citing inference and AI costs and different margins from regular software. Lemkin calls that argument all true, but says the human implications are still stark and calls the one-twelfth ratio a warning about AI's effect on jobs.

Investor Rory O'Driscoll countered that it is a play with far more computing leverage and less human leverage. With fewer people, he said, roughly the same pay pool can be shared among a smaller group, which is why those who remain are paid well.

Questions people ask

Does having fewer employees mean Anthropic is more profitable?

Not necessarily. AI companies spend heavily on computer chips, data centers, and electricity to train and run models. So lower headcounts do not automatically mean higher profit margins than traditional software.

Where does the money go if it does not go to more workers?

The source points only to heavy spending on inference and AI costs, not to a detailed split of revenue. It also says the smaller staff is paid well, with roughly the same pay pool shared among fewer people.

Will ordinary companies become this lean?

Probably not to the same degree. Most businesses still need people for sales, service, and physical work. Anthropic sells AI itself, so it benefits most directly from substituting computing power for hiring.

More in Startups