Logo Assembly Logo Hemicycle

Speech on artificial intelligence and public functions - La Sapienza University, Rome, Italy, 23 November

(Anglais uniquement)

I’m Very pleased to be here today alongside Deborah Bergamini, Undersecretary of State for Relations with Parliament. In fact, Ms Bergamini, was one of the key rapporteurs of the Assembly on the issue of Artificial Intelligence and presented - on behalf of the Committee of Political Affairs and Democracy – the report on "The Need for democratic governance of artificial intelligence". She has proactively promoted the views of the Assembly on the important issues related to AI – I look forward to hearing more from her today.

It is a matter of fact, that AI influences our lives and transforms societies in a variety of ways.

On the positive side, technology boosts progress and innovation – this is especially relevant today – for example, finding an effective vaccine against COVID-19 relied considerably on scientific progress and new technologies.

On the negative side, AI-based systems bear risks, including regarding the respect for our fundamental rights and freedoms. For example, AI can be used to control and filter information flows or to exercise automated censorship of content published on social media.

Lack of transparency in data collection and its use by algorithms reduces the ability of human users to take fully informed decisions.

In the criminal justice field, predictive policing and algorithms may increase discrimination and bias.

Lack of knowledge may lead to lack of trust in AI and its benefits. Breaking this vicious circle is probably one of the major challenges we face when dealing with AI. And we can face this challenge by focusing on solutions rather than on the problems.

Therefore, it is of utmost importance to strike the right balance between mitigating the risks and making full use of the advantages that AI can offer, and I believe that one of the first steps to build citizens trust in AI and its benefits is to set the rules of the game.

A transparent and clearly defined legal and ethical framework that must be respected by all.

I. Regulation as a response; role of Parliamentary Assembly

The role of Parliaments is to adopt laws, put forward normative standards and regulations. This is why the issue of regulating – in a human-rights-compliant manner - the use of AI is at the centre of the priorities of the Parliamentary Assembly of the Council of Europe.

In October 2020, the Parliamentary Assembly adopted a set of resolutions and recommendations, examining the opportunities and risks of AI for democracy, human rights and the rule of law. We examined a variety of issues, such as:

  • the need for democratic governance of artificial intelligence;
  • preventing discrimination caused by the use of artificial intelligence;
  • the role of artificial intelligence in policing and criminal justice systems;
  • medical, legal and ethical challenges posed by the use of AI;
  • artificial intelligence and labour markets;
  • legal aspects of ‘autonomous’ vehicles;
  • the brain-computer interface: new rights or new threats to fundamental freedoms.

Our research and our debates show that self-regulation and ethical principles are not sufficient and appropriate.

A common European legal framework for artificial intelligence is needed.

The Assembly also endorsed universally accepted and applicable core ethical principles upon which AI regulations should be based:

  • transparency, including accessibility and explicability;
  • justice and fairness, including non-discrimination;
  • human responsibility for decisions, including liability and the availability of remedies;
  • safety and security;
  • privacy and data protection.

Clearly, the application of AI in certain areas has a particularly serious potential to negatively impact our human rights. To address this and ensure that fundamental human rights safeguards are respected, a framework which is clear and transparent for all is needed.

This will increase transparency which is essential to build up knowledge, as well as predictability and legal certainty, which is essential to build up trust.

The Parliamentary Assembly therefore called on the Committee of Ministers to elaborate a "legally binding instrument governing artificial intelligence [that is] based on a comprehensive approach, deals with the whole life cycle of AI-based systems, is addressed to all stakeholders, and includes mechanisms to ensure the implementation of this instrument".

II. The Assembly strongly supports the work of the CAHAI

CAHAI has conducted impressive research. Its feasibility study has identified the rights that may be affected and has mapped the existing soft law and binding standards that may apply.

Indeed, we are not in a legal vacuum. Many international legal instruments, including CoE treaties such as the European Convention on Human Rights, the CoE Convention on Data Protection (Convention 108), or the Budapest Convention on Cybercrime, are already applicable to AI systems.

This, however, is not sufficient – a consolidated and coherent legal framework is needed.

The draft feasibility study proposes elaborating a comprehensive legal framework combining binding and non-binding legal instruments, that complement each other.

In the Parliamentary Assembly, we believe this is the right way forward.

Parliamentary input at all stages is particularly important because the role of Parliaments is to translate normative standards and guidelines into laws within our domestic jurisdictions.

When the international legal framework is finally in place, Parliaments will have to ratify it and adapt the relevant legislation. More important to ensure that the Assembly is fully on board right from the start.

III. What role for the Council of Europe?

When it comes to human rights, democracy and the rule of law, there is no doubt that it is the CoE standards that are internationally recognised and serve as benchmarks to measure progress or lack thereof. The European Convention on Human Rights provides a common and uniform legal framework for the protection of fundamental rights and freedoms.

Naturally, the CoE should lead the way in addressing the "new generation of rights" that we see emerging, including AI.

That’s why the Assembly welcomed the fact that, the 131st Ministerial Session in Hamburg in May earlier this year, decided to give priority to the work on an appropriate legal framework for the development, design and application of artificial intelligence.

The Parliamentary Assembly is providing support to the process, both in terms of content but also equally important, political leadership.

The CoE has the opportunity and the potential to become the regional leader in developing a comprehensive and human-rights-compliant legal framework which could serve as a model for developing global regulations.

Many of our Conventions – especially the most innovative and ground-braking ones – such as for instance the Budapest Convention on Cybercrime or Oviedo Convention on bioethics and human rights are open for non-member states.

When working on AI, there are three issues which are essential to keep in mind:

  • First, adopting a holistic approach: the entire life-cycle of AI-based systems, starting from the design phase, its application and its effects and consequences have to be addressed.
  • Second, the importance of the multi-stakeholder approach: all those who are concerned – governments, parliaments, private sector, scientific experts, as well as users – have to have a say. This is already the approach adopted by CAHAI and I would encourage them to hear the views and voices of all stakeholders concerned. And today’s event, contributes to the process of bringing together all parties concerned – including the academic world.
  • Third, striking the right balance between regulation and technological development: over-regulation may kill innovation; on the other hand, not enough regulation creates the risk of overstepping the “red lines”.

The Parliamentary Assembly will continue working on AI through follow-up to the adopted resolutions and recommendations, further reports, its Sub-committee on Artificial Intelligence and Human Rights and its observer status with the Ad hoc Committee on Artificial Intelligence (CAHAI).

I thank you for your attention and look forward to the contributions and discussions.