The New Yorkerのインスタグラム(newyorkermag) - 11月14日 08時00分

Proponents view facial-recognition technology as an invaluable tool that can help make policing more efficient. The fact that the software can lead law enforcement to make mistakes—studies have shown that the danger of being misidentified is especially pronounced for members of certain marginalized communities—has been overlooked due to what researchers call “automation bias”: the inclination of people using computer technology to uncritically accept what machines tell them, especially when they perform functions that are inscrutable to them. In part because the use of facial recognition has outpaced a willingness to regulate it, nobody can say exactly how many people have been erroneously identified. Yet, in a striking number of the wrongful arrests that have been documented, facial-recognition searches represented virtually the entire investigation.

In one such case, Alonzo Cornelius Sawyer, a Maryland resident in his mid-50s, was arrested on suspicion of viciously attacking a bus driver. He was visibly older and taller than the assailant captured on security footage, but he’d been identified as a match in a facial-recognition search; he was charged with theft and two counts of second-degree assault, and faced up to 25 years in prison. “All I thought is ‘Black man, gone—lost in the system,’ ” he told Eyal Press. At the link in our bio, read a new investigation into the damage that can be done when police put more stock in fallible facial-recognition searches than the evidence before their own eyes. Illustration by @ibrahirn.

[BIHAKUEN]UVシールド(UVShield)

>> 飲む日焼け止め！「UVシールド」を購入する