The international digital and human rights organization Access Now has withdrawn in protest from its role as a member of the Partnership for AI (PAI) due to lack of change among companies affiliated with the group or incorporation of opinions from civil society organizations. PAI was formed in September 2016 by a consortium of Big Tech companies and corporate giants such as Apple, Amazon, Facebook, Google, IBM and Microsoft. Since then, PAI has grown to include more than 100 member organizations, more than half of which are now non-profit, community or human rights-focused groups such as Data & Society and Human Rights Watch.
“We have learned from the conversations with peers, and PAI has given us the chance to contribute to the larger discussion on artificial intelligence in a new forum,”
Access Now also withdrew because the group advocates a ban on face recognition and other biometric technology that can be used for mass surveillance. Earlier this year, Partnership on AI produced an educational resource on face recognition for decision makers and the public, but PAI has not taken a position on whether the technology should be used. Access Now joined PAI about a year ago, and in the letter to the PAI management team, Access Now leaders concluded that PAI is unlikely to change its position and support a ban on face recognition.
“The events of this year, from the public health crisis to the global bill on racial justice, have only underscored the urgency of addressing the risks of these technologies in a meaningful way,” the letter said. “As more public authorities around the world are open to directly banning technologies such as face recognition, we will continue to focus our efforts where they will be most effective in achieving our priorities.”
Governments’ use of surveillance technology has increased in both democratic and authoritarian nations in recent years. The Freedom of the Net report released today by Freedom House in 2020 found a decline in Internet freedom in many parts of the world over the years, and that governments are increasingly using COVID-19 as an excuse to enable surveillance.
The American Civil Liberties Union (ACLU), Amnesty International and the Electronic Frontier Foundation (EFF) – all members of the PAI – have led or supported face recognition bans in major cities, state legislatures and in the US Congress. Conversely, PAI members such as Amazon and Microsoft are some of the most well-known providers of face recognition in the world. During the largest protest in U.S. history in June, Amazon and Microsoft announced temporary moratoriums on face recognition sales to U.S. police. Reform work may be on the agenda for the next congress to address privacy, racial displacement and freedom of expression raised by face recognition.
More than two years after its founding, PAI began engaging in specific policy and AI ethics issues, such as advocating for governments to create special visas for AI researchers. PAI also opposed the use of algorithms in pre-trial risk assessments, which the Federal Bureau of Prisons used earlier this year to determine which prisoners were released early due to COVID-19. PAI publicly shares the names of members, but rarely shares the names of specific members who contributed to political position documents produced by PAI employees.
In response to the Access Now resignation letter, PAI CEO Terah Lyons told VentureBeat that PAI is working closely with technology companies to address and adjust their behavior, and hopefully the work will work over the next year. But, she said, engaging in a multi-stakeholder process and trying to reach agreement between different voices to ensure that AI benefits people and society can be challenging and time consuming.
“It has definitely been a learning journey for us,” she said. “It’s also something that takes a lot of time to achieve moving industry practices in meaningful ways, and because we’ve just had program work for two years as a fairly young nonprofit organization, I expect it will still take us some time to really meaningfully move the needle. in that regard, but I think the good news is that we have laid a very important foundation, and we are already beginning to see evidence that the payment of dividends and some of the incremental choices our corporate members have made as a result of their commitment. ”
Examples of the kind of incremental change she refers to come from companies such as Facebook and Microsoft that are participating in the deepfake detection challenge, which was overseen by PAI’s steering committee. She also pointed to specific examples from PAI’s work in the areas of justice, accountability and transparency, but declined to share the names of specific companies or organizations that participated.
“A lot of the work we did with them on this issue, specifically, I think really affected how they thought about and internally addressed the challenges they face related to these issues, as well as some of the other companies involved,” she said.
Lyons said that PAI chose not to take a position on face recognition because the non-profit organization assesses each topic on a case-by-case basis to determine where PAI can best impact.
“It is not necessarily the case that we will be in the best position to take a position on every single question. But we try our best to ensure that we deliver a kind of service and value to ensure that these debates, when they take place in a public or private setting, are as well informed and evidence-based as possible, and that we equip and provide all our organizations the opportunity to be in direct conversation with each other about these tough issues, ”she said.
In other AI ethics and policy issues, Lyons said PAI has not produced any research or formed a steering committee to address the role AI plays in the concentration of power from technology companies. Last week, a subcommittee on antitrust in the House of Representatives concluded a 16-month investigation with a lengthy report concluding that Amazon, Apple, Facebook and Google are monopolies. The report concluded that power consolidated by Big Tech companies threatens competitive markets and democracy. It also states artificial intelligence and the acquisition of start-ups in AI and new fields as instrumental parts of continuing to increase the competitive advantage of Big Tech companies. However, the PAI created a joint prosperity initiative that will try to address how to more distribute power and wealth in a more even way, so that continuous concentration of power from technology companies is no longer seen as an inevitability. The shared prosperity group includes a number of renowned AI ethics researchers and described in a blog post last month.