Amazon announced Wednesday that it would place a one-year moratorium on police use of its facial recognition software, Rekognition. The brief statement comes just two days after IBM announced in a letter to Congress that it would no longer offer general purpose facial recognition software.
Research by Joy Buolamwini and Deborah Raji in 2018 and 2019 was instrumental in revealing racial and gender bias in Rekognition and other facial recognition software. A 2019 study by the National Telecommunications and Information Administration supported their conclusion.
“The move by Amazon, as sudden as it appears, comes after years of pressure by civil society groups, investors, and other advocates calling on the company to stop selling Rekognition to law enforcement,” said Isedua Oribhabor, U.S. Policy Analyst at Access Now.
During Amazon’s Annual General Meeting on May 27, shareholders voted on multiple proposals relating to Rekognition and other surveillance technology the company produces. Though the shareholder proposals did not pass, this move indicates that Amazon is responding to the concerns of its shareholders and of civil society.
Amazon has given little indication of how it plans to use this year-long moratorium or what the moratorium means for the company. “Hitting the pause button is a small step in the right direction, but is not enough,” said Oribhabor. “At a minimum, Amazon must create a cross-functional human rights team and perform human rights impact assessments to truly evaluate the effects that its tech products have on civil and human rights, particularly on the rights of the Black and Brown communities that have suffered the most from use of its technology.”
“Although strong regulation and safeguards can mitigate certain harms, we need to accept that certain uses of technology, such as biometric recognition systems that enable mass surveillance, are so incompatible with the protection of fundamental rights that these systems simply should not be used,” said Daniel Leufer, Mozilla Fellow at Access Now.
Amazon’s announcement also underscores the need for regulation by the U.S. government. Facial recognition and other surveillance software pose serious risks for human rights including freedom of assembly and privacy, some of which can be mitigated only by comprehensive and robust privacy and data protection laws in the United States. Regulating specific tools, including facial recognition, without protecting fundamental rights still leaves individuals and communities at risk.