GPD has responded to a multistakeholder consultation launched by the Council of Europe’s Ad Hoc Committee on Artificial Intelligence (CAHAI).
The consultation is part of CAHAI’s work to “examine the feasibility and potential elements on the basis of broad multi-stakeholder consultations, of a legal framework for the development, design and application of artificial intelligence, based on Council of Europe’s standards on human rights, democracy and the rule of law”. Following the publication at the end of 2020 of a feasibility study on a legal framework for AI design, development and application, this consultation aims to inform CAHAI’s Legal Frameworks Group, which is in charge of preparing the main elements of the future legal framework.
As well as responding to the direct questions posed, our consultation response sets out some of the considerations that CAHAI should take into account when developing any legal framework. In particular, we recommend that:
- While there is currently no universally agreed definition of AI, any legal instrument should include a definition which, even if non-exhaustive, is future-proof and ensures as great a degree of legal clarity and certainty as possible;
- Any legal framework should ensure that risks to human rights are mitigated against in all areas—not just certain “priority” sectors”. It should also recognise that the same AI system or technology can both protect the rights of some but also put the rights of others at risk;
- Any legal instrument which relates to the private sector should encourage business enterprises to comply with their responsibilities to respect human rights, as set out in the the United Nations Guiding Principles on Business and Human Rights. This would include ensuring that they have adequate human rights due diligence processes in place, and processes to enable the remediation of any adverse human rights impacts they cause or contribute to; and;
- Any legal instrument should, as far as possible, be consistent with and complement other human rights-based initiatives and instruments. This may include (depending on their final form) those being developed under the auspices of the Office of the United Nations High Commissioner for Human Rights, UNESCO and the European Union.
While we welcome the efforts of CAHAI to reach out to other stakeholders, the framing of many of the questions does not allow for the type of nuanced considerations and reflections that are critical to any discussion around AI governance. Many of the most important questions in the consultation—relating to the prioritisation of core thematic areas and elements that the legal instrument should include—only allow respondents to rank between options or to “agree” or “disagree” with broad statements, without any opportunity to explain or expand on their answers. This limits the scope for detailed input and analysis.
Further, the consultation is only open to institutions (rather than individuals) and there have been limited efforts to promote awareness of it outside of Europe, despite the fact that it is likely that any legal instrument will be open to non-European states to endorse.
We believe that processes to develop legal frameworks on AI should be as open, inclusive and transparent as possible, enabling and encouraging participation from a broad and diverse range of stakeholders as possible. It is not clear at this stage whether CAHAI intends to offer stakeholders further opportunities to participate in this process. We would strongly encourage CAHAI to do so—to ensure that its work is able to draw on the important expertise, insight and perspectives from non-governmental stakeholders, including those from outside of Europe.
The deadline for consultation responses was 29 April 2021. These will be considered by CAHAI at a dedicated event on the multistakeholder consultation, provisionally planned for 21 May 2021. They will also be considered by CAHAI’s Legal Frameworks Group as it prepares the main elements of the future legal framework. We’ll be following developments at CAHAI closely—take a look at our AI Policy Hub and sign up to our monthly Digest for regular updates and analysis.