Influencing the future direction for AI in employment in the UK
In September 2025, employment law specialist Alana Penkethman was invited to contribute to an evidence session held by the Westminster Employment Forum (WEmF) with Viscount Camrose and Lord Lucas. The session focused on the future for AI in employment with a focus on priorities for policy, regulation and industry best practice.
For Alana this was a meeting of two passions, employment law and artificial intelligence. You can read her contribution to what was a full and insightful session below.
The Employee-Employer Relationship at the Heart of AI
Most of my cases over my career have been about relationships between employees and employers. From a legal perspective, the key issues when implementing AI in workplaces are those issues that also relate to the relationships between employees and employers.
We know that workplace culture and employer reputation are more important than ever before, particularly to employees. We also know that the foundations to good workplace culture include transparency, accountability, and fairness. Therefore, we can see that there’s a potential risk there; where poorly implemented AI results in decisions that may not seem transparent, those decisions may not seem fair, and they could lead to disputes. If those disputes then result in litigation, it can damage employer reputation, and it can also deter high quality employees, customers, and investors. Clearly, this is something that we need to be taking seriously.
Legal Risks: Transparency, Accountability and Fairness
With potential legal implications arising from transparency and trust, those issues can lead to a potential breach contract, which could amount to a constructive dismissal claim. Equally, we have issues about accountability for dismissals, whether it’s redundancy, conduct, or capability, if there’s no accountability for decision making, this could result in an unfair dismissal claim. Finally, we have to consider fairness and inclusion, and the associated discrimination risks.
Transparency and Trust in AI Decision-Making
Looking more closely at transparency, this is essential for trust and confidence, which is one of the implied duties in every contract of employment between employee and employer. The duty of trust and confidence requires employers’ decisions to be made in good faith. An increased reliance on AI in decision making can erode the personal nature of the relationship between the employer and the employee, as AI lacks the human ability to make those nuanced decisions that line managers can. Equally, overuse of AI for performance tracking and performance management can create a culture of fear, which can harm psychological safety. Again, this is becoming increasingly important, particularly in constructive unfair dismissal claims, where employees will basically decide they have no alternative but to resign.
Accountability and Fair Procedure
Looking at the accountability element, for a dismissal to be fair, we need to have a potentially fair reason, but we also need a fair procedure. Now, the reason is less likely to be problematic in respect of AI, but if the manager isn’t able to identify how that decision has been reached, then they might not be able to identify which of the potentially fair reasons they’re relying on to dismiss that employee, which could be problematic in a tribunal claim. However, the bigger issue is the procedure. Every piece of employment legislation comes down to reasonableness, and what was in the mind of the decision maker at the time. If employers are using either factually inaccurate data, or a flawed algorithm, employers are going to really struggle to articulate how they reached the decision if they don’t know it themselves. Then, if they if they cannot be held accountable, that procedure will be deemed to be unfair, resulting in an unfair dismissal.
Discrimination Risks Across the Employee Lifecycle
However, the biggest risk is discrimination, and that risk is relevant throughout the lifespan of an employee’s employment, from recruitment to termination. The risk of direct discrimination applies where an algorithm is potentially using discriminatory data and will come out with a discriminatory decision. The only defence to that direct discrimination claim is to be able to show that the reason for the less favourable treatment, whether it’s a dismissal or failure to be promoted, or any other less favourable treatment, isn’t the employee’s protected characteristic. Therefore, if you cannot explain why these things have happened, the claim will succeed. Equally, with indirect discrimination, there’s a risk that the data or the algorithm might put groups of people or certain people with a protected characteristic at a particular disadvantage. We see those disadvantages frequently in disability discrimination claims, where you might have somebody who’s got a physical disability, and they might not be able to use certain technology equally. We also know that statistically, minority ethnics have a lower success rate when using facial recognition technology, and we also know that when facial recognition technology does not recognise individuals, that can be a degrading and humiliating experience which can amount to harassment.
Supporting Job Quality in an AI-Driven Workplace
Quite apart from the risk of claims, I think the other issue to consider when implementing AI in workplaces is job quality. We need to make sure we support individuals and train employees to improve digital literacy and knowledge of AI basics, but we also need to dig deeper and look at career coaching to improve skills in critical thinking. In addition, employers must make sure that job descriptions are meaningful, looking at the purpose of roles using realistic targets and objectives.
Final Thoughts: Principles for Ethical AI Use
I think my three points to finish up on would be that going forward, we need to make sure that we’re ensuring transparency, accountability, and fairness in everything that we do when we’re implementing AI.
______________________________________________________________________________________
Disclaimer
This information is intended for general informational purposes only and does not constitute legal advice. We recommend seeking professional advice before taking any action on the information provided. If you would like to discuss your specific circumstances, please feel free to contact us on 0800 2800 421.
