On 9 October 2024, the Council of Europe’s Committee of Ministers issued a new Recommendation regarding the ethical and organisational aspects of the use of artificial intelligence and related digital technologies by prison and probation services.

The Recommendation aims to establish principles and rules that guide the Council of Europe member states in their legislation, criminal policy and practice, given that the use of artificial intelligence (AI) for criminal justice purposes is advancing at great pace and execution of penal sanctions and measures is one of strongest manifestations of public power that deeply interferes with human dignity, human rights and privacy, including the collection and processing of personal data.

As a guiding principle, the Recommendation stresses that prison and probation services use AI and related digital technologies legitimately and proportionately and only if they contribute to the rehabilitation of offenders. AI and related digital technologies should not replace prison and probation staff but rather assist them in their everyday work, and help the criminal justice system, the execution of penal sanctions and measures and the reduction of recidivism.

The Recommendation defines first nine basic principles if AI and related digital technologies are designed, developed, provided, used and decommissioned. These principles are, for example, the principle of legality, legal certainty and liability, the principle of good governance, transparency, traceability and explicability, the principle of the right to a human review of decision, and the principle of human-centred use of AI and related digital technologies.

Next to data protection and privacy issues, the Recommendation deals with various use scenarios of AI in the context of prisons and probation services, such as:

• Use for the purpose of safety, security and good order;
• Use for offender management, risk assessment, rehabilitation and reintegration;
• Use for staff selection, management, training and development.

The Recommendation sets out for instance that the use of AI for maintaining safety, security and good order can be for the benefit of better risk and crisis management, but should be strictly necessary, proportionate to the purpose and avoid any negative effects on the privacy and well-being of offenders and staff.

Looking at the use of AI vis-à-vis offenders, the Recommendation makes the point that rehabilitation and reintegration of offenders, as well as their social contacts, may be facilitated by the use of AI and related digital technologies. When such tools are used for the personalisation of treatment and reintegration plans, this should be done with care to avoid biases. The use of such tools should not replace regular face-to-face human contact between professionals and the offenders, including, where necessary, the work with their families and children.

AI should also be applied with care if it comes to its use for managing appointments and interventions (including appointments with healthcare professionals, lawyers, social workers and any other professionals).

Background: As a sector-specific work, the Recommendation complements the Council of Europe’s more general Framework Convention on Artificial Intelligence which was opened for signature on 5 September 2024 (→ related link). The Framework Convention is a first-of-its-kind global legally binding instrument designed to ensure that AI upholds common standards in human rights, democracy and the rule of law, and to minimise the risk of those rights and principles being undermined as a result of the use of AI.

The Council of Europe has also published a brochure in which it provides information on its efforts to promote the establishment of standards that meet the challenges to human rights posed by the use of AI systems.