Artificial intelligence (AI) training is not optional; it is a legal requirement. AI is already embedded and used by accountants and others in business across a range of everyday tools, from financial forecasting and fraud detection systems to AI-powered workplace productivity tools, such as Microsoft Co-pilot or ChatGPT.

What many organisations may not realise is that this use of AI now carries a legal obligation.

Under the European Union’s AI Act, organisations developing and/or using AI systems must ensure a sufficient level of AI literacy among their staff and those operating AI systems on their behalf.

This requirement has applied since February 2025 and forms part of the first wave of obligations under the regulation.

In practical terms, this means AI training is no longer simply about learning and development. It is a compliance issue—and one finance leaders, in particular, should not underestimate.

A new type of regulatory obligation

The AI Act represents the first comprehensive attempt by a major regulator to establish a legal framework for artificial intelligence.

Like the General Data Protection Regulation before it, the AI Act introduces a mix of principles, governance requirements and specific obligations, applicable on the basis of risk.

Among the earliest of these is the concept of “AI literacy”, set out in Article 4 of the Act.

At its core, this requirement is straightforward: organisations must take steps to ensure their employees, and others acting on their behalf, have the knowledge and understanding necessary to use AI systems responsibly and appropriately.

However, the implications are far-reaching as this obligation applies not only to organisations that develop AI systems, but also to those that use them. In practice, this means almost every organisation is within scope.

This is because AI is already embedded in widely used tools, from productivity software to finance systems, often without explicit recognition. Whether recognised or not, however, its existence in systems used by an organisation triggers an AI training obligation.

Why this matters more than it first appears

At a glance, the requirement to ensure staff are “AI literate” may appear relatively light with no prescribed training programmes or certifications.

This flexibility should not be mistaken for a lack of importance.

Organisations are required to take proportionate measures to ensure understanding, based on the role and context in which AI is being used.

In other words, they must be able to demonstrate that they have considered the issue and responded appropriately.

What this means for professionals in finance

For professionals in finance, this has three key implications.

1It introduces a governance and accountability dimension As with financial controls or data protection, organisations will need to show that appropriate processes are in place. Insufficient training may become a point of scrutiny if issues arise.

2 It highlights the risk of unstructured AI use AI tools are increasingly used informally for drafting content, analysing data or supporting decisions. Without a baseline level of understanding, there is a risk of errors, bias or overreliance on AI outputs.

3 It signals the direction of travel AI literacy is only one of the early obligations under the AI Act. It sits alongside broader requirements relating to risk management, transparency and oversight. Organisations that act early will be better positioned for what follows.

As Jared Browne, Fexco’s Group Head of Privacy and AI Governance, notes: “AI literacy training for staff is now a legal requirement under the EU AI Act for any organisation that uses AI as part of its work.

“When you consider that most organisations will already be using enterprise AI tools, such as Microsoft Copilot or ChatGPT Enterprise, this means the obligation to develop AI literacy leaves no one untouched”, Browne continues.

“Organisations will be very familiar with the need to roll out compliance training in areas such as GDPR. However, with the AI Act, there is a clear silver lining. Staff are not only required to understand the risks of AI, but also the opportunities it presents”.

“There is therefore no downside to bringing staff on the AI literacy journey.

“By doing so, organisations can proactively manage risk while also positioning themselves to realise the benefits of AI—in effect, future-proofing the organisation for the next major technological shift in the workplace”.

This balance between risk and opportunity is important. While regulation drives the requirement, the outcome is a more informed and capable workforce.

What “AI literacy” means in practice

The concept of AI literacy is not about turning employees into technical specialists. Rather, it is about ensuring individuals have an appropriate level of understanding for their roles.

In most organisations, this can be broken down into a number of core elements:

• Basic understanding of AI systems

Awareness of how AI works at a high level, including the nature of AI-generated outputs.

• Understanding limitations and risks

Recognition of issues such as inaccuracies, bias and lack of transparency.

• Appropriate use and judgement

Knowing when AI can be used, and when human oversight is required.

• Legal and ethical considerations

Awareness of data protection, confidentiality and regulatory expectations.

• Role-specific knowledge

Additional depth where AI is used in higher-risk contexts, such as financial decision-making.

The key point is that AI literacy should be proportionate and role-based, rather than uniform across the organisation.

What organisations should be doing now

Although the requirement is already in force, many organisations are still at an early stage in their response. A practical approach is to treat AI literacy in the same structured way as other compliance obligations.

1Identify where AI is being used I may be embedded in systems or used informally by staff. Visibility is a critical first step.

2 Map users and roles Different roles require different levels of understanding.

3Establish a baseline level of training All staff using AI should have access to core training on risks, limitations and appropriate use.

4Provide role-specific training where needed Enhanced training should be provided for higher-risk or more complex use cases.

5Document the approach From a governance perspective, organisations should retain evidence of the measures they have taken.

In practice, this will require collaboration between learning and development, compliance, IT and business teams.

Why finance leaders should take ownership

While AI literacy may initially appear to sit within human resources or learning and development, there is a strong case for finance leaders to play an active role.

Finance functions are already significant users of AI, whether through forecasting, anomaly detection or automation. Particularly for Chartered Accountants, who hold a position of trusted business leadership, the quality of outputs and decisions based on these outputs is critical.

Finance leaders are also accustomed to working within regulated environments, with a focus on controls, oversight and evidence. The AI Act introduces similar expectations in the context of AI use.

There are clear parallels here with GDPR. Organisations that took a structured approach to data protection were better positioned to manage risk and compliance.

The same is likely to be true for AI. By engaging early, finance leaders can help ensure their organisation’s response is both effective and aligned with broader governance frameworks.

From obligation to opportunity

Although the AI literacy requirement has its origins in regulatory obligation, it also presents an opportunity.

Organisations that invest in building a well-informed workforce will be better placed to use AI effectively and responsibly. Rather than limiting innovation, a structured approach to training can enable more confident and productive use of AI across the organisation.

In this sense, AI literacy can be seen not only as a compliance requirement, but as a foundation for long-term capability.

The EU AI Act marks a significant step in the regulation of artificial intelligence. While much attention has focused on its more technical provisions, the requirement to ensure AI literacy across organisations is both immediate and far-reaching.

For many organisations, this obligation is already in effect yet remains largely unaddressed.

For finance professionals and business leaders, the message is clear: AI training is no longer optional. It is part of the organisation’s compliance landscape.

Those that act early will not only mitigate risk but also position themselves to make effective use of one of the most transformative technologies in business today.

Joe Carroll is Head of Professional Development with Chartered Accountants Ireland. Chartered Accountants Ireland will be launching a new Certificate in AI Foundations for Accountants to address key training needs under the AI Act in autumn 2026