The convention's purpose is to ensure that protections for fundamental human rights, democracy, and the rule of law will also apply to current and future challenges raised by AI.
It represents a significant step forward in international AI governance, addressing the need for an international framework agreement to guide law and policymaking in a field where technology often outpaces governance.
Commitment to human rights and democracy
The convention underscores the Council of Europe’s commitment to unity and cooperation among its member states, emphasising respect for human rights, democracy, and the rule of law.
Recognising the transformative potential of AI, the convention aims to harness these technologies to promote human prosperity, well-being, and sustainable development, while mitigating risks that could undermine fundamental values.
One of the primary concerns addressed by the convention is the potential for AI to exacerbate inequalities and discrimination, particularly against women and vulnerable groups.
By establishing a legal framework, the convention seeks to ensure that AI systems are designed, developed, and deployed in ways that protect human rights, democracy and the rule of law, thereby fostering the trustworthiness of artificial intelligence systems.
International legal umbrella
The AI Convention is distinct from the EU AI Act. While the EU AI Act aims to develop a comprehensive regulatory framework for AI within the European Union's internal market, the AI Convention adopts a broader perspective.
Its objective is to establish an international binding treaty addressing AI in the context of human rights, the rule of law, and democracy.
In other words, the convention is the international legal umbrella under which legislation, such as the EU’s AI Act, sits.
A convention only becomes legally binding on a particular country when that state ratifies it, whereas an EU regulation, like the AI Act, is binding, and directly applicable in all member states of the EU.
Similar definitions
In the convention, a vague definition of AI systems is provided:
“For the purposes of this Convention, ‘artificial-intelligence system’ means a machine-based system that for explicit or implicit objectives, infers, from the input it receives, how to generate outputs such as predictions, content, recommendations or decisions that may influence physical or virtual environments. Different artificial-intelligence systems vary in their levels of autonomy and adaptiveness after deployment.”
This definition is almost identical to the definition of artificial intelligence contained in the EU AI Act:
“‘AI system’ means a machine-based system that is designed to operate with varying levels of autonomy and that may exhibit adaptiveness after deployment, and that, for explicit or implicit objectives, infers, from the input it receives, how to generate outputs such as predictions, content, recommendations, or decisions that can influence physical or virtual environment.”
Both definitions closely adhere to the revised OECD definition of AI, emphasising the machine-based nature of these technologies, and their capacity to ‘infer’ outputs such as predictions, content, and decisions.
Both definitions underscore the varying levels of autonomy and adaptiveness in AI systems, which the convention seeks to comprehensively regulate.
Several central obligations and principles provide the convention’s foundation:
- Parties are required to adopt measures to ensure that AI activities are consistent with human-rights obligations, both internationally and domestically,
- Parties must implement measures to prevent AI from undermining democratic processes, including maintaining judicial independence and access to justice,
- Adequate transparency and oversight mechanisms are mandated, ensuring that AI activities are identifiable and accountable,
- The convention mandates that parties implement measures to prevent AI systems from perpetuating or exacerbating inequalities and discrimination,
- Strong safeguards for privacy and personal data are required, aligning with existing international standards,
- Measures to promote the reliability of artificial-intelligence systems and trust in their outputs, and to establish controlled environments for developing and testing AI systems are required.
Prospects
The success of this convention will depend on the commitment of member states and other signatories to uphold and implement its principles.
As AI continues to evolve, ongoing vigilance will be crucial to addressing new challenges and ensuring that technological progress benefits all of humanity.
The Council of Europe’s initiative offers a model for other regions and international bodies, highlighting the importance of shared values and cooperative governance in the face of rapid technological change.