AI literacy

The AI Act requires providers and deployers of AI systems to take measures to support the development of AI literacy of their staff.

In brief

In brief

The ability to understand AI, critically question it and use it responsibly is vital today. AI literacy is essential to make informed decisions, minimise risks and comply with statutory requirements. At the same time, AI literacy enables the best possible use to be made of the potential of AI and innovations to be developed.
The AI Act does not specifically set out how to support the development of AI literacy. No standardised training measures are laid down. Nor is it necessary to guarantee any specific level of AI literacy of any individual.
The individual context of an organisation is fundamental: the organisation’s role as a provider or deployer developing or using AI systems, together with the associated risk, and the staff’s previous knowledge and experience are decisive factors.
Developing AI literacy is an ongoing process. AI literacy should be regularly updated and adapted to technological developments.
The AI Act does not require certification for participation in training or qualification measures. Measures can be carried out internally or externally.
Measures taken to build AI literacy should be documented.
There are various free AI literacy training courses (see information further below).

What is AI literacy?

“AI literacy” is defined in Art. 3 point (56) of the AI Act:

“‘AI literacy’ means skills, knowledge and understanding that allow providers, deployers and affected persons, taking into account their respective rights and obligations in the context of this Regulation, to make an informed deployment of AI systems, as well as to gain awareness about the opportunities and risks of AI and possible harm it can cause.”

According to this definition, AI literacy essentially comprises:

the skills, knowledge and understanding
to deploy AI systems in an informed, responsible and safe manner and
to gain awareness about the opportunities and risks (for example ethical, legal and societal) presented by AI.

The rules on AI literacy are set forth in Art. of the AI Act:

“1. Providers and deployers of AI systems shall take measures to support the development of AI literacy of their staff and other persons dealing with the operation and use of AI systems on their behalf, taking into account their technical knowledge, experience, education and training and the context the AI systems are to be used in, and considering the persons or groups of persons on whom the AI systems are to be used. This obligation does not require providers or deployers to guarantee any specific level of AI literacy of any individual.

2. The Commission and the Member States shall support and facilitate the efforts of providers and deployers of AI systems, in particular SMEs, in fulfilling their obligation under paragraph 1 of this Article. For that purpose, the Commission shall publish practical examples of how to comply with that obligation on the single information platform referred to in Article 62(3), point (b).

3. The Board shall adopt recommendations, taking into account European competence frameworks, to support the Commission and Member States in the promotion of AI literacy required under paragraph 1, including by setting out common objectives.”

Article 4 covers the following aspects:

Who: Providers and deployers of AI systems, including general-purpose AI systems such as chatbots, irrespective of the sector they operate in or the size of the organisation.
What: Measures to support the development of AI literacy of all persons operating or using AI systems on an organisation’s behalf. AI literacy covers both an organisation’s own staff and other persons dealing with the operation or use of AI systems on the organisation’s behalf, such as contractors and service providers. They are supported in this effort by the European Commission and the Member States.
How: Taking into account factors relating to the persons and the context.

AI literacy as an enabler

It is within the interests and therefore the responsibility of an organisation itself to ensure a high level of AI literacy of its staff in order to make full use of the undisputed potential of AI. A high level of AI literacy enables organisations to, among other things:

Make informed decisions: AI literacy is intended to enable providers, deployers and persons concerned to make informed decisions about AI systems and consequently obtain the greatest benefits from AI systems, while protecting fundamental rights, health and safety and enabling democratic control;
Improve working conditions and promote innovation: AI literacy can help to improve working conditions and support the innovation path of trustworthy AI; and
Ensure compliance with and enforcement of the AI Act: AI literacy is intended to provide all relevant operators along the AI value chain with the necessary knowledge to ensure appropriate compliance with and correct enforcement of the AI Act.

What is or is not laid down in the AI Act?

To support the development of AI literacy, organisations should, when deploying and using AI:

Not required for compliance with Article 4 are:

- ensure a general understanding of AI;

- consider their own role as a provider or deployer;

- consider the risks of the specific AI system in the specific context;

- incorporate the latest developments and changes.

- guarantee specific levels of AI literacy;

- formalised or standardised training measures;

- (external) certification of any measures taken;

- the introduction of an AI officer

A lack of AI literacy can be seen as non-compliance with a duty of care, in particular if this results in any harm. The Bundesnetzagentur therefore recommends that organisations keep a good record of their measures to support the development AI literacy. This will enable them to prove at any time that they comply with the requirements in Article 4.

What does the AI Act require, and what is not mandatory?

For compliance with the AI literacy requirement, organisations should, when deploying and using AI:	Not required for compliance with Art. 4 are:
- ensure a general understanding of AI; - consider their own role as a provider or deployer; - consider the risks of the specific AI system in the specific context; - incorporate the latest developments and changes.	- formalised or standardised training measures; - (external) certification of any measures taken; - the introduction of an AI officer; or - regular pre-checks by supervisory authorities on measures for ensuring AI literacy.

For compliance with the AI literacy requirement, organisations should, when deploying and using AI:

Not required for compliance with Art. 4 are:

- ensure a general understanding of AI;

- consider their own role as a provider or deployer;

- consider the risks of the specific AI system in the specific context;

- incorporate the latest developments and changes.

- formalised or standardised training measures;

- (external) certification of any measures taken;

- the introduction of an AI officer; or

- regular pre-checks by supervisory authorities on measures for ensuring AI literacy.

A lack of AI literacy can be seen as non-compliance with a duty of care, in particular if this results in any harm. The Bundesnetzagentur therefore recommends that organisations keep a good record of their measures for ensuring AI literacy. This will enable them to prove at any time that they comply with the requirements in Article 4.

How to build AI literacy?

The AI Act deliberately does not lay down any specific formats, other formalised and standardised measures or specific levels of AI literacy of any individual, simply because there is no one-size-fits-all solution. Organisations are free to tailor their measures to support the development of AI literacy to their individual needs. These needs depend on, for example, the AI systems that an organisation uses, their context together with the associated risks, the tasks that staff perform and how much the staff already know about AI. Organisations are therefore free to decide for themselves how to support the development of AI literacy. Their decisions should be plausible and understandable. A standardised solution cannot accommodate such a wide range of needs.

As a guideline, the Bundesnetzagentur has identified three appropriate cornerstones to support the development of AI literacy.

Three cornerstones for building AI literacy
1. Identifying needs	Which persons develop, operate or use AI systems? Which AI systems are they? For which purpose do the persons work with the AI systems? Which risks and opportunities are associated with working with the AI systems?
2. Designing measures	Identify the organisation’s role along the AI value chain; take into account individual factors relating to the persons concerned, such as their training, experience, and type of work; consider the context in which an AI system is deployed, for example the field of application, the persons concerned, and the purpose, and the associated risk; define goals and responsibilities.
3. Keeping records, evaluate, update	Keeping adequate records of the measures implemented, including the type of measures, their scope in terms of content and duration, and the persons taking part; regular evaluation of the measures, e.g., to assess whether objectives have been met and to identify any need for adjustments; refresher training tailored to specific needs, e.g., because AI systems or the context of use may change over time, new technologies may open up new fields of application.

Format and content of measures

The AI Act does not lay down a specific format for measures to support the development of AI literacy. Measures can range from self-learning programmes, workshops and training courses to multi-staged advanced training, depending on individual needs and the specific context. Measures can be carried out internally or externally.

Organisations are free to decide themselves on the content of the measures.

The Bundesnetzagentur recommends an interdisciplinary and multi-stage structure for the content. An interdisciplinary structure allows proper account to be taken of the technical, legal and ethical aspects of AI and interaction with AI. A multi-stage structure accommodates the fact that AI literacy is built up in stages according to the various levels of knowledge among staff.

The following summary of possible content for building AI literacy is neither binding nor exhaustive. Content should be adapted to an organisation’s specific circumstances, such as its size, the sector it operates in and the level of technological maturity, as well as to the individual needs of the persons concerned.

Stage 1: Creating a basic understanding of data and AI within the organisation

Basics of data and AI, including terms and history
Overview of AI technologies, including machine learning and large language models
General opportunities and risks posed by AI, for example based on use cases or role play

Stage 2: Building advanced AI literacy

Role of the organisation along the AI value chain, for example developer or user
Technical aspects of the AI used
Specific opportunities and risks and legal classification of the AI used

Stage 3: Role-specific training with individual focal points, for example technical, legal or ethical aspects

Support

There is a wide range of offers available free of charge to support organisations in building AI literacy. These include European and national measures as well as private initiatives and associations:

The European AI Office supports the Member States in implementing Art. 4 and provides information about AI talent, skills and literacy. It has also hosted a Webinar on AI literacy, keeps a list of AI literacy practices and provides questions and answers on AI literacy. Voluntary codes of conduct promoting AI literacy are supported by the European Commission and the Member States in cooperation with stakeholders.

European Digital Innovation Hubs (EDIHs) help SMEs, small mid-caps and public sector organisations to respond to digital challenges and improve their competitiveness. Their services comprise testing before investing, expanding digital literacy, advice on financing options, and networking.

Around 100 AI instructors belong to the network of Mittelstand-Digital Innovation Hubs (MDZ) in Germany. The instructors support SMEs individually and on an impartial basis. Together with the Hubs, they offer workshops, presentations, roadshows, demonstrations and other services (such as an AI radiness check) relating to AI literacy specifically for SMEs, for example with sector-related and process-related use cases. The aim is to enable SMEs to identify the opportunities and challenges presented by AI for their own businesses and implement specific applications.

There is also a wide range of online courses available free of charge, including those offered by by AI Campus and the appliedAI Institute for Europe.