Ireland - how to address dismissal in the age of Generative AI

Summary

GenAI tools can assist with reviewing large volumes of information and identifying relevant documentation. However, they should always be used in a closed, secure enterprise environment and should not replace human decision-making in disciplinary processes, particularly in cases involving alleged misconduct.

If a dispute arises, the employer must present a decision-maker capable of explaining and defending their reasoning before the Workplace Relations Commission (“WRC”). Where a decision-maker cannot clearly articulate how they reached a conclusion, particularly if they have relied heavily on GenAI this may significantly undermine the employer’s defence.

Can GenAI be used in a disciplinary process?

There is currently no prohibition on the use of GenAI in disciplinary processes. However, recently updated WRC guidance on the use of AI in submissions published in May 2026 on foot of recent commentary by Adjudication Officers highlights the risks associated with over-reliance on AI-generated material.

While the guidance does not require disclosure of AI use (in contrast to emerging expectations in the civil courts), it clearly signals that a cautious approach is required.

In the absence of specific guidance on disciplinary processes, it is likely that the WRC would scrutinise any use of GenAI carefully. The central question will be whether its use compromised:

• Natural justice
• Procedural fairness
• Meaningful human oversight

Critically, Article 22 GDPR provides that individuals have the right not to be subject to decisions based solely on automated processing that significantly affects them. A dismissal decision may fall within this scope, reinforcing the importance of maintaining a genuine “human in the loop”.

The use of GenAI in a material way in a disciplinary process is likely to be deemed high-risk under the EU AI Act. with effect from 2 December 2027, there will be additional requirements regarding such use, including appropriate risk-management, transparency, and human oversight.

Is there an obligation to disclose the use of GenAI?

If an employer uses GenAI in a disciplinary process and inputs employee personal data - which seems inevitable - the GDPR transparency obligations are engaged. The employer must explain, in its Employee Privacy Notice, the purposes and legal basis of the processing, categories of data, recipients, and provide the other transparency information under Articles 13 or 14 GDPR. This of course also applies where GenAI is used to progress employee personal data outside of a disciplinary context.

In a WRC hearing, the decision-maker may be questioned on how they reached the decision to dismiss.

If GenAI was used, the decision maker may be cross-examined on:

• The extent of its use
• The prompts provided
• The outputs generated
• The influence of that output on the final decision

Although the WRC does not operate a formal discovery process, both parties must submit the core documents on which they rely in advance of the hearing. An Adjudication Officer has the statutory power to direct the production of relevant evidence.

Failure to comply with such a direction may result in adverse inferences being drawn.

Can an employer defend a dismissal where GenAI was used?

The WRC has not yet directly considered the role of GenAI in dismissal decisions. However, it is highly likely that any such use would be subject to close scrutiny.

To defend an unfair dismissal claim, the employer must demonstrate that:

• The decision to dismiss fell within the “band of reasonable responses”
• Appropriate consideration was given to mitigating factors

The decision-maker must be in a position to clearly explain:

• How they assessed the evidence
• Why they concluded misconduct occurred (on the balance of probabilities)
• Why dismissal was considered proportionate

GenAI cannot provide this reasoning. Therefore, over-reliance on AI risks creating a gap between the decision reached and the decision-maker’s ability to justify it.

This principle is reflected in recent WRC guidance emphasising that AI should be used as a tool, not as a substitute for human judgment. A similar issue arose in Ferko v Beyond Reach Limited t/a Car Wash Crew ADJ-00060622/2026, where the Complainant’s case was undermined by an inability to clearly articulate the basis of the claim following reliance on AI.

Appropriate use of GenAI in disciplinary processes

GenAI can be a useful support tool in disciplinary procedures, particularly for:

• Organising and reviewing documentation
• Identifying relevant evidence
• Generating draft interview notes
• Assisting with administrative tasks

It can also be used to assist in drafting outcome letters. However, the decision-maker must:

• Independently assess all evidence
• Conduct interviews and weigh credibility
• Reach their own conclusions

Any written decision must reflect the decision-maker’s own reasoning and demonstrate that all relevant factors, including mitigation were properly considered.

Practical steps for employers

Employers should take proactive steps to manage risk, including:

• Implementing clear policies on the permitted use of GenAI in HR processes (subject to any additional changes which will be required under the EU AI Act)
• Updating Employee Privacy Notices regarding the use of GenAI in disciplinary matters
• Providing training to managers on appropriate use and limitations of use
• Informing managers that AI should be used for support not decision making
• Ensuring robust documentation of the decision-making process.