Data Backup Digest

The team with Clearview AI aims to build the first-ever “search engine for faces” thanks to a federal patent that was recently filed – and granted – for its proprietary facial recognition technology. While it’s certainly not the first facial recognition system we’ve seen, Clearview AI’s software is expected to bolster law enforcement efforts in the future.

Highly Controversial Software

Although similar systems have been proven as effective tools in identifying wanted criminals, there are plenty of critics that cite numerous concerns with facial recognition technology as a whole. Not only do they claim that these systems prone to a multitude of errors, but they also claim that they often lead to incidents of wrongful arrest.

However, Clearview AI insists otherwise. According to a recent audit by the NIST, the accuracy of their facial recognition technology is highly accurate. Moreover, a company spokesperson has verified that there are no cases of wrongful arrest as a result of their software. They also placed within the top 10 of current facial recognition tech vendors in recent federal testing.

How it Works

As mentioned, Clearview AI’s latest software is likened to a “search engine for faces.” It works by scraping, or gathering, images that are publicly available – including social media profile pictures. These images are then shared with law enforcement personnel or governmental entities for use in large-scale databases and surveillance systems – all without the individual’s consent.

Potential Legal Issues

Since these images are all scraped from public sources, there are no legal repercussions for the Clearview AI team. But this isn’t to say that their efforts have gone unchallenged.

A recent investigation by an Australian privacy regulator has found Clearview AI in breach of their citizen’s rights to privacy. The company has since been ordered to remove any and all images of Australian citizens within their database.

The Office of the Australian Information Commissioner, OAIC, recently issued a press releasing stating, in part: “The covert collection of this kind of sensitive information is unreasonably intrusive and unfair. It carries significant risk of harm to individuals, including vulnerable groups such as children and victims of crime, whose images can be searched on Clearview AI’s database.”

While that investigation was performed in tandem with the United Kingdom’s Information Commissioner’s Officer, or ICO, there hasn’t been a ruling regarding the legal status of Clearview AI’s systems within the UK.

Clearview AI insists that all of the images were publicly available – ultimately clearing them of any legal responsibility. They also reiterate the fact that all of the images were published within the U.S., thereby making them exempt from Australian laws.

It’s also important to note that this all comes at a time when Meta, Facebook’s new parent company, effectively ended their facial recognition technology. They cited growing concerns over facial recognition technology as well as a recent settlement for breaching privacy laws in the state of Illinois that cost the company approximately $650 million. Whether or not they had plans to end their program before the settlement is unclear.