The Council of Europe Convention on AI: Next steps

πŸ”₯ On September 5, 2024, the European Union (on behalf of its 27 member States), the US, UK, Andorra, Georgia, Iceland, Norway, Moldova, San Marino and Israel, signed the Council of Europe (CoE) Framework Convention on AI and Human Rights, Democracy and the Rule of Law.

πŸ‘‰ Universal geographical scope:

- The Convention has become the first-ever binding treaty on AI.

- Any states in the world who believes in safeguarding human rights, democracy and the rule of law in the digital age is welcome to ratify it.

πŸ‘‰ Human-centric AI perspective:

+ Key guardrails & possibility of red lines based on fundamental rights and principles

β–ΊHuman dignity and individual autonomy

β–Ί Equality and non-discrimination

β–Ί Respect for privacy and personal data protection

β–Ί Transparency and oversight

β–Ί Accountability and responsibility

β–Ί Reliability & Safe innovation

+ Governance mechanisms:

β–Ί Risk & impact management system

β–Ί institutionalised international cooperation

See brochure summarizing key features: https://lnkd.in/dCXZR7-j

See full Convention, https://lnkd.in/evcuQ3uW

🚦 2 key steps are necessary for the Treaty to realize its potential:

1/ it should be signed and ratified by the broadest number of countries so that it becomes a real international benchmark

2/ Parties should commit to ensure the Convention applies to the broadest possible range of activities involving AI, via the declaration they are supposed to submit together with their ratification instrument

Karine Caunes, Executive Director of the Center for AI & Digital Humanism, who participated actively in the CoE negotiations, emphasised: "from a European perspective, the Convention could be a key tool in ensuring a human-centric interpretation of the EU AI Act. This could be a game-changer at a time when:

- the "safety agenda" seems to have pushed the "fairness agenda" to the curb in European circles

- and "international interoperability" talks replace concerns for the public good.

Making sure that European standards & values are respected is a key priority for Digihumanism. It should be as well for European institutions. We will make sure to remind them and hold them accountable."

Previous
Previous

Digihumanism provides expert feedback to the Dutch Department for the Coordination of Algorithmic Oversight on AI Act prohibited practices

Next
Next

Digihumanism Contributes to the General Purpose AI (GPAI) Code of Practice