Anthropic’s latest funding round reportedly valuing the artificial intelligence group at $380bn was not just another large private tech deal. The round was led by major institutional investors and also included previously committed capital from Microsoft and NVIDIA, part of a broader $15bn strategic agreement struck late last year. It was interpreted as a signal that AI companies are no longer experimental platforms. They are becoming economic infrastructure.
The reaction was immediate. Shares in outsourcing groups, freight brokers and high growth software companies fell sharply across several markets. Investors were not simply digesting a valuation headline. They were repricing business models.
From Tool to Service Provider
The speed of Anthropic’s rise is central to the anxiety. The company disclosed a $14bn annual revenue run rate, having generated almost no revenue at the beginning of 2023.
Much of that growth is reportedly driven by enterprise customers rather than consumer subscriptions.
For investors, the implication is clear: large corporations are no longer just experimenting with AI. They are deploying it at scale.
One product in particular has focused attention
Anthropic’s coding agent, which is said to be generating roughly $2.5bn in annual revenue on its own. Unlike earlier “copilot” tools that assisted developers, this system is described as capable of managing multi step software projects with limited supervision.
That distinction matters. Assistance improves productivity. Autonomy challenges employment structures.
Pressure on the Outsourcing Model
The sharpest reaction came in India, where IT services groups form a core part of the equity market. The traditional outsourcing model depends on large teams of junior engineers billing clients by the hour. If companies begin buying finished software output directly from AI systems, the economics change.
Investors fear a shift away from “time and materials” contracts towards outcome based pricing. In that world, scale would depend less on headcount and more on computing capacity and proprietary data.
Whether that transition happens quickly is uncertain. But markets tend to price risk before it materialises.
Logistics and the Middle Man Problem
The sell off also spread to freight brokerage companies.
The concern here is similar: if AI systems can autonomously negotiate shipping rates, reroute cargo and update contracts in real time, the traditional role of the human broker could shrink.
Adding to the volatility was the rise of Algorhythm Holdings, which recently pivoted into AI powered logistics with a platform known as SemiCab. Until last year, the company operated under the name The Singing Machine, producing karaoke hardware. Its transformation and subsequent surge in share price has amplified the sense that barriers to entry are eroding.
If software can scale operations without adding staff, investors naturally question the durability of commission based models.
The “Service as a Software” Shift
What ties these reactions together is a broader shift in how AI is being monetised.
For decades, companies purchased software tools and employed people to operate them. Increasingly, AI systems are delivering the output directly writing code, analysing contracts, managing supply chains.
Some investors describe this as “service as a software”: instead of paying for labour supported by software, clients pay for outcomes produced by algorithms.
If that model holds, revenue could concentrate in the hands of a small number of AI platform providers, while labour intensive intermediaries face margin pressure.
A Regulatory Wildcard
Anthropic also announced a $20mn donation to Public First Action, a group advocating expanded AI regulation. The company presents this as a commitment to safety. Critics, however, argue that stricter regulatory frameworks may favour well capitalised firms that can absorb compliance costs.
This has introduced a further concern in markets: that frontier AI groups may not only dominate technologically, but shape the rules under which they operate.
Overreaction or Structural Shift?
Not all analysts believe the sell off is justified. Large enterprises remain cautious about fully autonomous systems, particularly in regulated industries. Human oversight, accountability and trust still carry weight. Established companies also hold decades of proprietary data that AI models rely upon.
Yet even sceptics concede that the baseline has shifted. AI is no longer being valued as a productivity enhancement layered onto existing systems. It is being priced as a substitute for parts of those systems.
The $380bn valuation may prove ambitious. But the market reaction suggests investors are grappling with a deeper question: if AI firms become the new backbone of digital infrastructure, which existing business models remain intact and which quietly dissolve?
That uncertainty, more than any single funding round, explains the turbulence.