Anthropic CEO Dario Amodei believes AI “will disrupt 50% of entry-level white-collar jobs over 1-5 years”.

Framed as a warning and couched in concerned tones, his essay last month nonetheless doubled up as a terrific sales pitch for a company seeking the best possible terms ahead of an upcoming IPO.

And it’s had quite an impact. Over the past month, Anthropic has sat at the centre of AI-related market turmoil amid speculation its Claude Code plugins could replace swathes of white-collar workers making up the backbone of the global services economy.

Initially limited to SaaS players, the threat of disintermediation has spread across the professional services universe, as investors worry about corporates vibe-coding legal documents, financial research, and even advisory work across capital markets services provided by investment banks.

This may seem fanciful in the current state, with white shoe firms, the big four, and bulge bracket banks alike so important in bringing credibility and prestige to financial offerings across equities and fixed income instruments.

But whether Claude can or can’t replace the work of these firms now doesn’t matter – Anthropic just needs enough people to believe it possibly could in order to secure a gargantuan IPO valuation.

It’s well on the way. On 12 February, Anthropic raised USD 30bn in a Series G round that valued it at USD 380bn on a post-money basis, this was over double the USD 13bn raised in September during a Series F fundraising round.

This valuation arrived at 27x Anthropic’s annual revenue run rate of USD 14bn. Growth is certainly impressive, with ARR growing 10x annually over the last three years.

And Anthropic has raised its 2026 revenue target to USD 18bn, with a 2027 goal of USD 55bn, rising to a 2029 figure of USD 148bn.

On the face of it, these revenue targets make investment at a USD 380bn valuation a great buy – and might support an even loftier market capitalization should the business proceed with its well-publicised IPO attempt this year. 

The major question mark over the equity story is the same as that being grappled with via the broader market shake-out: is Anthropic’s total addressable market within the software solutions space it already serves, or far broader in trampling over and devouring whole services industries – including many businesses within its client base?

AI software spending totalled USD 283bn in 2025, according to tech consultancy Gartner, which predicts growth to around USD 452bn in 2026 and USD 636bn in 2027. Gartner further predicts half of all IT spend in 2027 will be spent on AI.

Were Anthropic to achieve USD 18bn in 2026 revenues, that would represent around 4% of the total market; to hit USD 55bn in 2027, it will have to grow to an almost 9% market share against fiercely competitive rivals like Mag 7 stalwarts Microsoft and Alphabet, which are also investing billions into their own enterprise AI software models.

This also comes at a time of growing interest in open-source AI models that will allow established software businesses to guard against their workflows and IP being rinsed by agentic AI champions.

But if Anthropic, through Claude Code and other tools, is looking to take a slice of the whole global professional services industry – estimated to be worth around USD 7.4tn by 2027, according to market research player The Business Research Company – its revenue estimate would only be a paltry 0.7% of that total addressable market.

Growing in that space could prove a far easier story to sell at an IPO given its agentic AI solutions would be replacing old world legacy systems and businesses that are ripe for modernisation.

Far better to supplant these businesses entirely than to fight hard for every dollar in the enterprise software market against other AI giants.

Of course, this presupposes that hyperscalers will not also play that game; and equally that replicating tasks undertaken by professional services firms will aggregate to a service level that out-competes these businesses, with their layers of client relationships and trust.

Many will consider this to be more in the realm of science fiction than hard business fact, especially given modest AI productivity gains predicted by executives both sides of the pond.

Still, the more investors believe this could be a vision of the future – or are convinced that others believe it – the higher the valuation Anthropic will be able to achieve by the time it lists.

With retail interest in US IPOs increasing, this flood of capital could also warrant an almighty post-IPO pop.

As Anthropic executives talks up the impact of its industry – and therefore its own prospects – it’s worth remembering this is one of the oldest dances in venture capital. But this time, the waltz is being performed on the biggest stage of all.

Amodei admits many critics accuse him of falling foul of the lump of labour fallacy, taking issue with his prophesies on white-collar work elimination. In his essay, Amodei conceded previous innovations in human history have led to eventual workforce expansions as Jevons’ paradox accelerated demand.

“It’s possible things will go roughly the same with AI, but I would bet pretty strongly against it.”

Which is just another way of saying: this time it’s different.

Maybe it is. But it’s always worth taking such pronouncements with a dose of market realism.

Beyond Meat’s US listing in 2019 heralded the end of animal consumption, before serving all involved a huge helping of humble pie.

Anthropic’s technology and operational delivery already lift it into another level. Just remember that technological disruption at current speeds may involve the occasional crash.