A recent string of issues suggests that facial recognition reliability issues hurt people in times of need. Motherboard reports that there are ongoing complaints about the ID.me facial recognition system that at least 21 states are using to verify people claiming unemployment benefits. People have gone weeks or months without benefits when the Face Match system does not verify their identity, and sometimes have not had a chance to get help through a video chat system intended to resolve these issues.
ID.me chief Blake Hall blamed the problems on users rather than technology. Face Match’s algorithms were “99.9% efficient,” he said, and there was “no relationship” between skin tone and recognition failures. Instead, Hall suggested that people were not sharing selfies properly or following instructions.
Motherboard noted that at least some people have three attempts to pass the facial recognition check, however. The outlet also pointed out that the company’s claims over the costs of unemployment fraud nationwide have exploded rapidly in recent months, from $ 100 billion to $ 400 billion. While Hall attributed this to the expansion of “data points,” he didn’t say exactly how his company calculated the damage. In other words, the real threat of fraud is unclear.
Whatever happens with ID.me’s technology, the incidents highlight one of the reasons federal and state governments are hoping to limit facial recognition. Even though privacy and security aren’t issues, they don’t seem reliable enough to avoid significant issues. This 99.9% success rate could still leave many people without the benefits to which they are entitled. Systems like this may need to be considerably more reliable to eliminate these headaches in the future.