Opening of the 1st meeting of the Ad Hoc Committee on artificial intelligence (CAHAI)
Ladies and gentlemen,
It is a pleasure to open the first plenary session of the Ad Hoc Committee on Artificial Intelligence.
Everyone in this room is aware of the transformative power of AI.
We all share an enthusiasm for the ways in which it has the potential to change lives for the better.
But we also share a determination that it should not cause unintended harm:
That human rights, democracy and the rule of law should be enhanced, not hindered, by the rise of AI.
So, this intergovernmental committee is the first with the expertise and the mandate to examine the feasibility and elements of a legal framework based on the common standards agreed by this Organisation and applied across our member states.
Certainly, there is a growing appetite for action.
Back in 2006, there were only four or five ethical charters, worldwide, relating to the use of AI.
Today, there are nearly two hundred.
These charters have been issued by governments, industry and multi-stakeholder groups including, for example, the Montreal Declaration for the Responsible Development of Artificial Intelligence.
But these are instruments of self-regulation, and there is an increasing awareness that they do not address all of the challenges presented by AI.
That is why the Council of Europe’s Committee of Ministers took the decision earlier in the year to establish this Committee:
To send a clear signal that the time has come to look beyond charters, and towards how we might further protect and promote our hard-won rights, at the pan-European level.
Why should it be the Council of Europe that does this?
This Organisation is comprised of 47 member states.
And for the 830 million people who live within them, the European Convention on Human Rights, the European Social Charter, and more than 200 additional conventions owned by this Organisation form the basis of their human rights protection in law.
Of these conventions, two already apply directly to new technologies.
Our Budapest Convention on Cybercrime is the only legally binding international instrument on that theme.
Today, its 64 states parties reach far beyond the borders of Europe, and 130 countries worldwide benefit from the Council of Europe’s capacity-building activities on this subject.
And our Convention on Data Protection – Convention 108 – was recently modified to take into account changes in this fast-evolving area.
The original Convention has been ratified by 55 countries and, for this treaty too, we operate capacity-building around the world.
So, we have the common legal space, and we have the standard-setting expertise to ensure that we produce an effective legal framework.
There have of course been questions about whether new standards will hinder innovation.
In fact, it has the potential to support it.
Because a common legal framework for Europe – with the potential for use around the world – will be better than small-scale or sub-regional legal standards.
Our approach can provide certainty and coherence.
It will also ensure greater trust in Artificial Intelligence itself.
In the years to come, AI is set to play an ever-greater role in the way that governments and public institutions operate.
When citizens are assured that this innovation is being overseen by legal standards that protect their rights, they will feel more reassured and confident about the future of technology.
And a legal framework would also be an asset when it comes to liability questions, because products that are made according to clear standards protect both the producer and the consumer, ensuring that these generate greater public trust too.
So, the task of CAHAI – your task – is to say what that binding legal framework might look like if we choose to follow that path.
Over the next two years there is much work for you to do.
How ambitious should we be?
Should there be a framework convention?
How would new standards be checked?
Certainly, certification is an important issue that will need to be addressed, along with a range of other issues including whether there are some AI applications for which self-regulation will suffice.
By May of next year, you are tasked with producing a progress report for the Committee of Ministers, which will include specific proposals for further action.
So, progress must be swift.
But what we have in this Committee is exactly the right group of people to make that happen.
Member states, observer states, other international organisations, civil society, and industry representatives:
Your input is required to get this right.
We have every confidence in you, and I thank each and every one of you for your commitment.