Photo: Anete Lusina / pexels.com

A ban that sends a strong message

When one of the world's largest technology companies bans its own employees from using a well-known AI tool, the entire industry takes notice. That is exactly what is happening now at Alibaba. The company is prohibiting its employees from using Claude Code and other AI products from the U.S. company Anthropic on company computers.

The background sounds like something out of a spy movie: There are suspicions that the software collects information about its users, which could allow inferences to be drawn about their location or work environment. Anthropic explained that these features were part of an experiment designed to curb abuse and unauthorized use. Nevertheless, Alibaba pulled the plug and deemed the software too risky for corporate use. According to reports, the company plans to use its own AI solutions instead going forward. The decision also comes amid growing tensions between Chinese and American AI companies.

It's no longer just about data protection

When people think of artificial intelligence, many initially think of practical tools for writing, programming, or email. But in business, there’s much more at stake. Those who use AI often transmit large amounts of information. This can include trade secrets, source code, customer data, or internal documents.

Even if a provider assures users that it will handle this data responsibly, the question remains: What information leaves the company? What data is stored? And who can analyze it later?

Large companies, in particular, now view these issues as a strategic priority. After all, data is often more valuable today than machinery or buildings. In the worst-case scenario, losing control of one’s information means losing a key competitive advantage.

The global AI race is heating up

The case also illustrates just how intense the competition between American and Chinese AI companies has become.

In recent weeks, there have been mutual accusations. Anthropic accused Alibaba of improving its own AI systems using output from other models. At the same time, it was revealed that certain features of Claude Code were intended to help better detect usage from China and restrict unauthorized access. These developments have further heightened mutual mistrust.

For companies, this means one thing above all else: Choosing an AI is no longer merely a matter of personal preference. Security, trust, and the political landscape are increasingly playing a decisive role.

What Companies Should Learn From This Now

Even though the current case involves a major international corporation, the real message is at least as important for small and medium-sized businesses.

Many employees today use ChatGPT, Claude, or other AI services as a matter of course in their daily work. Often, this happens without the employer’s explicit consent. As a result, internal information is quickly entered into external systems.

That is why companies should establish clear rules:

• Which AI applications are permitted?
• What data should never be entered?
• What alternatives does the company itself offer?
• Who regularly checks the systems in use?

Such simple guidelines not only prevent unnecessary risks; they also provide reassurance to employees who want to use AI effectively without unintentionally causing problems.

Technology alone is not enough

The Alibaba case illustrates something else as well: Even the most advanced technology is no substitute for a healthy dose of skepticism.

AI is evolving at a breathtaking pace. Its capabilities are constantly changing. New possibilities emerge almost daily. At the same time, providers regularly update their systems, often without users being able to understand every technical detail.

That is why companies should not blindly follow every new trend. Anyone who wants to use AI needs regular checks, clear lines of responsibility, and an understanding of what information truly belongs in external systems—and what information should never leave the company.

An honest look behind the scenes—one that’s okay to be a little painful

Our position is clear: The real scandal isn’t the ban at Alibaba. The scandal is that many companies still have no rules at all governing the use of AI. They blithely try out every new chatbot, copying and pasting confidential information—and it’s only when a major corporation sounds the alarm that everyone suddenly wakes up.

AI is a powerful tool. But anyone who works with it should treat it with the same care as they would a company bank account or their house keys. Anyone who believes that free or third-party AI services are automatically the right place to store internal data is taking unnecessary risks. Innovation is important—but common sense should always take precedence over the latest hype.