November 21, 2026

AI-Driven Performance Reviews Could Expose Employers to Legal Action, Lawyer Warns

photo of outer space

Artificial intelligence is rapidly becoming part of day-to-day company operations, with many organisations now using AI-driven tools to support performance reviews, monitor productivity and inform promotion or dismissal decisions. While these systems promise efficiency and consistency, legal experts warn that introducing AI into such high-stakes workplace processes carries significant risk if not handled carefully.

Recent figures show that around two-thirds of HR professionals are now using AI at work. From analysing employee output to generating performance scores, AI is increasingly influencing decisions that can directly affect careers. However, relying on automated systems for performance management raises serious concerns around bias, transparency and data protection.

Qarrar Somji, Solicitor-Advocate and Director at Witan Solicitors, explains that while AI can help teams manage growing workloads, it must not be allowed to replace human judgment entirely. He said: “Understandably, employers are looking to reduce many of the repetitive tasks involved in HR and manage extensive employee data. While AI can be an essential lifeline for overstretched HR teams, it should never be treated as a substitute for human judgment. An over-reliance on these systems risks serious errors and exposes companies to a number of legal issues.”

Unfairness and Bias in AI-Led Reviews

One of the biggest challenges with AI-driven performance reviews is the risk of unfair or discriminatory outcomes. Qarrar highlights that AI tools are only as reliable as the data used to train them, and that historical bias can easily be carried forward into automated assessments.

“When it comes down to it, AI systems are entirely dependent on the data you provide. If the data contains existing bias, the tool may unintentionally continue to perpetuate it, leading to potential discrimination. In some cases, employers may be held liable for these outcomes under the Equality Act 2010.

“Not to mention, AI tools don’t always take into account employees’ unique circumstances. For example, an AI reliant on statistical data may overlook that an employee’s performance is affected by a disability. This lack of flexibility can expose employers to claims for failing to make reasonable adjustments or discrimination arising from disability. That’s why managers need to fill the gaps, providing the context that AI tools simply can’t pick up from numbers and make appropriate adjustments.”

Transparency, Trust and Legal Exposure

As AI becomes more involved in evaluating performance, transparency becomes harder to maintain. Many systems operate as black boxes, making it difficult for managers to fully explain how conclusions are reached. In the long run, this can quickly erode trust.

Qarrar warns that this lack of clarity can have serious consequences if AI-informed reviews feed into formal capability or dismissal processes.

“Generally, AI tools use complex algorithms in their reasoning. This complexity makes transparency difficult. If managers cannot fully justify the rationale behind performance conclusions, employees’ trust in the process may be undermined. This loss of mutual trust and confidence can be damaging.

“Performance reviews are usually the basis for capability procedures, which can lead to dismissal. If AI plays a significant part in this decision, a lack of transparency could leave employers open to allegations of procedural unfairness, and employees would have every right to challenge any AI-driven assessments.”

Data Protection and GDPR Obligations

Introducing AI into performance management also means handling large volumes of personal employee data. This is an area where compliance failures can be costly. Qarrar stresses that organisations must treat AI use as a data protection issue, not just a technology upgrade.

“If you’re implementing AI into your performance management system, this will likely involve processing your employees’ personal data. All personal data must be processed fairly, lawfully, and transparently under GDPR.

“It’s a good idea for HR departments to establish a clear AI policy that addresses issues, such as confidentiality, data protection, and intellectual property. The policy should also set out that employees should not input sensitive business data or personal information into public AI tools that the company has not officially approved. It’s also worth listing approved AI platforms and banning any unvetted tools.

“Employees have the right to be informed about any automated decision-making. They also have the right not to be subject to decisions based solely on automated data processing. Before implementing significant AI use, it’s wise for employers to conduct a Data Protection Impact Assessment (DPIA).”

Safeguards Leaders Must Put in Place

Ultimately, Qarrar emphasises that AI should support, not replace, human decision-making, particularly when it comes to employee outcomes that can have long-term consequences.

“HR leaders and managers need to ensure AI does not replace genuine human decision-making. Relying on AI to make key decisions, such as promoting and dismissing employees, will lead to disaster.

“Without proper training, it will be nearly impossible for managers to effectively interpret AI-generated reports. Equally, it’s crucial that they understand the system’s limitations and use genuine independent judgment. After all, whether an employee is kept on, promoted or dismissed should be determined on a case-by-case basis, not by a rigid AI system.

“But it’s important to keep in mind this is an ongoing process. As AI tools continue to advance, regular assessments of potential risks, including unintended discrimination, need to be carried out. Maintaining human oversight and accountability is key.”