Clearview AI has reported that its entire client list has been stolen by an intruder who “gained unauthorized access.”

Clearview has repeatedly been in the news for its controversial practices over the last couple of months. The company has amassed a database of some three billion photos, which it has scraped from millions of websites, including the most popular social media sites on the web. Clearview then sells access to that searchable database, along with its facial recognition software, to law enforcement agencies around the country.

The company is reportedly looking to expand its operation overseas, and has included oppressive regimes on its list of potential countries it may do business with. The potential harm the company’s software could do was illustrated when New York Times reporter Kashmir Hill asked police officers to run her face against the company’s database, which turned up no matches. After running her face, however, the police officers received phone calls from Clearview telling them they shouldn’t be talking to the media.

Now, in a report The Daily Beast reviewed, Clearview says an intruder stole a copy of the company’s entire client list, including the number of user accounts each customer had created and the number of searches they had conducted. The company claims that its servers were not breached and that there was “no compromise of Clearview’s systems or network.”

This breach perfectly illustrates the danger of a company rushing headlong into a potentially dangerous area where many other companies have feared to tread. Google, Facebook and others have certainly had the ability to do what Clearview has done and would no doubt greatly profit from it. Every other company, however, has acted with restraint out of recognition of the harm that could potentially be done.