- Advances in artificial technology enable new technologies like facial-recognition
- Microsoft says regulatory focus should include racial and gender bias, privacy and mass government surveillance
- Many believe new laws should get permission, when facial-recognition is being used
Microsoft is urging governments around the globe to regulate facial-recognition technology in 2019 with independent assessment of accuracy and bias and prohibition of ongoing surveillance of specific people without a court order.
The company competes with Amazon, Facebook, and Google in the emerging market for facial-recognition products and has been very vocal on the need for government regulations. In July, Microsoft argued with a blog post that change will not happen if only a few companies adopt new standards while others ignore them.
Brad Smith, Microsoft’s president and chief legal officer, underscored the urgency in a blog post stating that that delays to enacting new rules could “exacerbate societal issues.” Society is ill-served “by a commercial race to the bottom, with tech companies forced to choose between social responsibility and market success.”
Microsoft’s push for regulation highlights the ambivalence over powerful new technologies enabled by advances in artificial intelligence. Adoption of facial recognition is proceeding quickly—especially in China, where the government uses it extensively for surveillance—creating concerns about potential misuse.
Earlier this year, civil rights activists called on Amazon to stop selling its facial-recognition technology to law enforcement agencies. Andy Jassy, Amazon Web Services, chief executive said the company hasn’t seen any abuses of the technology and suggested that “countries themselves have to decide” on the rules.
Mr. Smith identified benefits of facial-recognition technology including identifying missing children and driver authentication for Uber Technologies, Inc. He also emphasized specific areas where governments should focus legislation: racial and gender bias, privacy and mass government surveillance. He invoked George Orwell’s dystopian novel, 1984, in which a government tracks its citizens’ every move, as a cautionary tale.
He proposes that new laws notify people and get their consent when facial-recognition is being used. Smith acknowledged that Microsoft’s interests are competitive as well. Without regulation responsible companies could lose sales to competitors with technologies that increase discrimination or abuse human rights. This could tip the market towards an approach that is less socially responsible.
- Facial-recognition is everywhere – airports, police stations, and integrated into the largest cloud platforms in the world with few federal rules to govern how it’s used.
- Many facial recognition algorithms still show higher error rates for African-Americans, women, and young people, suggesting the systems might be entrenching societal biases.
- In addition to Microsoft, others calling for regulation of facial-recognition include rights advocacy groups such as the American Civil Liberties Union and the AI Now Institute.
Quests and Actions (Q&A):
- Should police use facial recognition? Do we have the civil society or governmental infrastructure to ensure that law enforcement would not abuse this technology?
- Will other business leaders ask the U.S. Congress to create a bipartisan expert commission to identify the best way to regulate this software as Microsoft proposes?
- How can we find ways to balance the public benefits of facial recognition with the obvious privacy concerns?