Anthropic против юристов: почему акции сервис-провайдеров летят в пропасть

The stock market is rarely merciful to those who fall behind progress, but on Tuesday we witnessed a true public execution. As soon as the Anthropic startup announced the launch of its new tool for the legal sector, the quotes of companies engaged in data service maintenance plummeted in sync. This is that very moment when theoretical discussions about replacing humans with artificial intelligence turned into concrete losses in investors' portfolios. We are witnessing the beginning of the end for a business model that for decades has fed entire corporations — the model of selling man-hours for performing boring office work.

Anthropicdidn't just update its Claude model, it created a specialized solution capable of analyzing legal documents, identifying risks, and matching facts with the efficiency of an experienced lawyer. For the business process outsourcing industry, this sounds like a death sentence. Why pay a huge staff of employees in India or Eastern Europe for primary contract audits if an algorithm does it in seconds and practically free? Investors instantly calculated this risk, understanding that the margins of classic service providers will soon plummet to zero.

The context of this situation is rooted in last year, when shares of the educational service Chegg crashed after it admitted that students preferred ChatGPT to its services. Now the story is repeating, but already on a more serious, corporate scale. If earlier AI was perceived as an amusing assistant that could write a letter or create a workout plan, now it's entering the territory of complex expertise. Legal analysis is just the first swallow. Next come financial audits, consulting, and deep data analytics, where previously "the human factor" was considered irreplaceable.

Why is this happening right now? Anthropic and their competitors from OpenAI realized that general chatbots are just the tip of the iceberg. The real money lies in creating narrowly focused agents that integrate into corporate workflows. When AI stops merely "talking" and starts "doing" specific work, the value of intermediaries who previously provided this service through people is wiped out. We are seeing a transition from a cloud computing economy to an economy of intelligent actions.

Many executives of data processing companies try to keep a stiff upper lip, claiming to be implementing AI within their structures. However, the market doesn't believe them. The problem is that if you implement AI to do work ten times faster, you can no longer bill for time spent. This is a fundamental conflict of interest that destroys the old service economy. While service providers try to reorganize, startups like Anthropic simply take their bread, offering direct access to technology.

In the coming months, we will likely see several more such crashes in other sectors. Any company whose main asset is "people at computers following instructions" is at risk. Investors will now look not at revenue, but at how well the business model is protected from the direct impact of language models. The irony of fate is that the very data that these companies collected and processed for years ultimately became fuel for training their digital killers.

The bottom line: Anthropic showed that specialized AI agents are ready to kill entire service markets. Who's next in line to pack their bags — accountants or junior analysts at investment banks?