Editor

February 20 – Setting a foundation for future legislation, the European Commission launches a public consultation on ethical and legal requirements for artificial intelligence. The initiative produces wide-ranging responses from member states, trade groups, academics, and civil society. In December 2019, incoming President von der Leyen called for “new rules on Artificial Intelligence that respect human safety and rights.” [CAIDP 1.3]

June – Hong Kong’s Chief Executive, appointed by China’s Communist Party, deploys AI-enabled face surveillance, communications tracking, and transit record analysis to suppress opponents of a security law. China has pursued “secure cities” as part of the Belt and Road Initiative

July 22 – OECD Secretary General Angel Gurria told the G-20 Digital Economy Ministers “AI’s full potential is still to come. To achieve this potential we must advance a human-centred and trustworthy AI, that respects the rule of law, human rights, democratic values and diversity, and that includes appropriate safeguards to ensure a fair and just society.” In June 2019, the G20 adopted AI Guidelines based on the 2019 OECD AI Principles.  [CAIDP 1.2]

August – The UK government withdraws biased algorithm after widespread public protest. Researchers determined that the AI-generated scores favored students enrolled at elite schools over actual test results. This disadvantaged high-performing students at underperforming schools.

September 9 – The Boston Global Forum and the Michael Dukakis Institute announce the Social Contract for the AI Age.  The Social Contract sets out a vision for a new society, based on innovation, creativity, and mutual respect.

September 7 – The European consumer association BEUC finds widespread public concerns about reliable AI. In comments to the European Commission, BEUC wrote “a strong regulatory framework is necessary” to guarantee consumers are protected against the risks posed by AI.

September 16-18 – The Boston Global Forum and the Club de Madrid, the world’s largest forum of former heads of democratic governments, organize a global policy dialogue on “A New Social Contract for the Age of AI.”

October 13 – Access Now resigns from Partnership on AI, citing a lack of influence with companies that established the “multi-stakeholder” process. Access Now said, “we support human rights impact assessments and red lines around use of these technologies, rather than an ethics, risk-based, or sandboxing approach.”

October – The Global Privacy Assembly, the worldwide conference of privacy commissioners, establishes accountability measures for AI, including human rights impact assessments. The Privacy Assembly also said that changes to data protection law are necessary for accountability in the development and use of AI. [CAIDP 1.15]

October 21 – The European Council adopts a resolution on protecting fundamental rights in the era of AI. Alone among EU countries, Poland objected to the resolution, citing the inclusion of the phrase “gender equality” among the fundamental rights.

November 3 – California passes Proposition 24, updating the state privacy law and creating new rights for algorithmic transparency. California residents also reject a ballot proposal to establish AI-based predictive policing.

November 5 – The Pope warns that AI could exacerbate economic inequalities around the world if a common good is not pursued. “Artificial intelligence is at the heart of the epochal change we are experiencing.” Earlier in the year, the Pope endorsed the Rome Call for AI Ethics  to “promote a sense of responsibility among organizations, governments and institutions.” [CAIDP 1.19]

December 10 – In a joint report, Transatlantic Approaches on Digital Governance, the Boston Global Forum and the Club de Madrid proposed global policies for a better management of digital technologies and Artificial Intelligence. The groups also urged world leaders to adopt and implement the Social Contract for the AI Age

December 12 – The Center for AI and Digital Policy (CAIDP) releases  “Artificial Intelligence and Democratic Values,” the first comprehensive review of national AI policies and practices. The CAIDP Report followed an extensive review of 30 countries and government organizations, undertaken by a team of international experts. [CAIDP 1.23]

December 12 – Speaking to the Boston Global Forum, President Ursula von der Leyen proposes a Transatlantic Agreement on Artificial Intelligence.  “We want to set a blueprint for regional and global standards aligned with our values: Human rights, and pluralism, inclusion and the protection of privacy,” said von der Leyen. Professor Nazi Choucri, on behalf of the Michael Dukakis Institute, proposed an International Accord on AI in 2021.[CAIDP 1.24]

 

Marc Rotenberg, Director

Center for AI and Digital Policy at Michael Dukakis Institute

The Center for AI and Digital Policy, founded in 2020, advises governments on technology policy

The third plenary session of the CAHAI, an expert group on AI policy, took place this week. [Meeting report.] The CAHAI (the “Ad Hoc Committee on Artificial Intelligence”) was established by the Committee of Ministers in 2019 to examine the feasibility of a new legal instrument based on the Council of Europe’s (COE) standards on human rights, democracy and the rule of law.

This week the CAHAI published Toward Regulation of AI Systems and convened a panel to discuss Democratic Governance of AI. The CAHAI report explored global perspectives on the development of AI legal frameworks, and highlighted recent developments in Israel, Mexico, and Japan. The report emphasized that the COE has “a crucial role to play to ensure that Artificial Intelligence (AI) complies with the Organisation’s standards on human rights, democracy and the rule of law.”

The Council of Europe is a leader in the realm of international legal instruments and human rights. All Council of Europe member states have signed up to the European Convention on Human Rights. And Article 8 of that Convention has done much to shape modern privacy law. COE Conventions are also open for ratification by non-member state. The original COE Convention on Privacy (Convention 108) was ratified by 56 countries.

A key outcome from the December 2020 meeting of the CAHAI was the approval of the Feasibility Study on AI Legal Standards. The Feasibility Study will soon be published online.

The COE also announced an online event on 20 January 2021 “Human Rights in the Age of AI – Europe as an International Standard Setter for AI,” organised with the support of the Council of Europe.

In CAIDP Update 1.17 (Oct. 31, 2020) we reported that the COE Parliamentary Assembly adopted a Resolution on the Need for Democratic Governance of Artificial Intelligence.  The Assembly called for “strong and swift action” by the Council of Europe. The parliamentarians warned that “soft-law instruments and self-regulation have proven so far not sufficient in addressing these challenges and in protecting human rights, democracy and rule of law.”

Marc Rotenberg, Director

Center for AI and Digital Policy at the Michael Dukakis Institute

The Center for AI and Digital Policy, founded in 2020, advises governments on technology policy.

The Center for AI and Digital Policy has published the first comprehensive review of national AI policies and practices.  The report “Artificial Intelligence and Democratic Values: the AI Social Contract Index 2020” was released in conjunction with the presentation, by the Boston Global Forum and the Michael Dukakis Institute, of the World Leader Peace and Security Award to European Commission President Ursula von der Leyen.

The CAIDP Report is based on an extensive review of AI policies and practices in 30 countries, undertaken by a team of international experts. The CAIDP Report found that (1) the OECD/G20 AI Principles have shaped global policies, (2) AI safeguards build on data protection law, (3) there are “red lines” for certain AI applications, such as facial surveillance and criminal sentencing, and (4) AI policy is in the early days but the pace is accelerating.

The CAIDP Report set out five recommendations: (1) Countries must establish national policies for AI that implement democratic values; (2) Countries must ensure public participation in AI policymaking and also create robust mechanisms for independent oversight of AI systems; (3) Countries must guarantee fairness, accountability, and transparency in all AI systems; (4) Countries must commit to these principles in the development, procurement, and implementation of AI systems for public services; and (5) Countries must halt the use of facial recognition for mass surveillance.

Among the countries reviewed, Germany was at the top. The experts cited Germany’s high level of public participation in AI policy making, strong data protection infrastructure, multiple independent agencies focused on AI, and leadership on AI policy within the European Union and globally.

Leading AI policy experts participated in a discussion of the AI Social Contract Index, including Amb. Karen Kornbluh (ret.), Wonki Min, Merve Hickok and Prof. Frank Pasquale. Representative Robin Kelly (D-IL), cosponsor of the House Resolution on a National AI Strategy, also spoke at the event.

The CAIDP plans updates of AI policy throughout 2021 and the release of a new AISCI-2021 in December 2021. Additional information about the report is available at caidp.dukakis.org/aisci-2020.

 

Marc Rotenberg, Director

Center for AI and Digital Policy at the Michael Dukakis Institute

The Center for AI and Digital Policy, founded in 2020, advises governments on technology policy.