Written by Kieran Harte

Artificial Intelligence, or AI, is a technology that enables machinery to think and solve problems in a human like way.

The EU’s Artificial Intelligence Act (the AI Act) establishes a legal framework to ensure appropriate AI use that respects our fundamental rights. The Act entered into force on 1 August 2024, and much like the GDPR, it will take two years to be fully enforced, following a phased approach to enforcement, tackling the higher-risk areas first.

Among the first areas coming into effect on 2 February 2025 are the rules on ensuring AI literacy of employees and other affected persons in accordance with Article 4 of the Act. The overall objective of the AI literacy provision is to contribute to improving working conditions and ultimately to sustain and consolidate trustworthy AI in the EU. Support and assistance will be provided by the European AI Board through the promotion of AI literacy tools, public awareness and understanding of the benefits, risks, safeguards, rights and obligations in relation to the use of AI systems. An example of a free course developed by the University of Helsinki on the basic elements of AI is available here.

AI literacy may be described as the knowledge and skills to enable providers, deployers, and affected persons to critically understand, evaluate, and use AI systems both ethically and safely. To their best extent, organisations must ensure their staff (and other affected persons) possess sufficient AI literacy, as supported by initiatives from the European AI Board, taking into consideration their technical knowledge, experience, education and training and the context that the AI systems will be used in. This may include the development of training programmes and collaboration with industry groups and regulatory bodies.

Compliance Steps for Organisations

1. Evaluate and assess the current level of AI literacy within your organisation, identifying gaps in knowledge of staff and other affected persons involved in AI development, deployment, and use.
2. Design and implement AI training tailored to management and staff in different roles in the organisation. This should cover the importance of compliance, ethical considerations, and understanding the opportunities, risks, and potential harms associated with AI. It should also offer an insight and understanding of the technical elements of AI systems, correct application during development and deployment, and interpreting AI outputs correctly. Specific training should also be provided on identifying and mitigating risks associated with prohibited AI practices.
3. AI literacy also involves recognition and appreciation of the ethical and societal considerations of AI. AI training should provide insight and understanding of how biases can be embedded in AI systems, the privacy implications of AI, and the need for transparency and accountability in AI applications.
4. Organisations need to invest in training and development programs to ensure their staff stay up to date with the latest AI trends, technologies, and best practices. Continuous learning should be incorporated in the training programme to keep up with technological advancements and regulatory changes.
5. Put in place internal policies and procedures covering regular and continuous AI literacy training for all staff and other affected persons, including an organisational compliance framework and mechanisms and metrics for monitoring and evaluating the effectiveness of AI literacy programmes.
6. Document and maintain records of AI training programmes and associated compliance measures. Be prepared to provide documentation and other evidence of compliance to regulatory bodies where required.
7. Engage and collaborate with peers in other industry groups, educational institutions, and regulatory bodies to keep up with the latest developments and best practices in AI and AI literacy

AI Literacy Training Content for Those Using AI Systems

• Make staff and other affected persons aware of the functioning and limitations of AI technology (i.e. how outputs are produced, possible data transfers, etc.), as well as authorised and prohibited uses.
• Encouraging and ensuring verification of input data and output quality as part of the training on how to use these systems.
• Ensuring that staff and other affected persons only submit information that they are allowed to share in the prompt or input data. Confidential information, such as personal or organisational data, should never be shared unless authorised to do so.
• Critically evaluate and assess outputs generated by the AI system to verify:
  • accuracy and quality that it is not protected by copyright
  • that biases are not introduced leading to possible discrimination or unfair treatment (e.g. stereotyping)
• Providing awareness of automation bias, which is the propensity of humans to favour output from automated decision-making systems while ignoring contradictory information made without automation, even where it is correct. AI systems should be designed to clearly inform users of the origin of the output, reminding them that it is up to them not to take the output “as is.”

The purpose of the EU AI Act is to regulate AI technologies and protect fundamental rights. Ensuring AI literacy is not simply about the avoidance of fines but encouraging trustworthy and ethical AI. Proactive steps taken to assess, educate, and monitor AI literacy by organisations will help them navigate their obligations under the Regulation while maintaining competitiveness in a fast moving and rapidly evolving technological landscape.