Artificial Intelligence (AI) is already changing how many workplaces operate. Employers are using it to screen job applications, analyse performance data, support investigations and assist with redundancy planning. Used well, AI can save time and help managers make better-informed decisions. Used badly, it can create unfairness, confusion and legal risk.
So, can technology make decisions about your job?
Increasingly, yes – but the legal responsibility stays with the employer. AI can assist a decision, but the final outcome still needs to be fair, reasonable and properly explained.
In practice, this means AI may be used for recruitment and CV screening, workforce planning, performance management, productivity monitoring, meeting transcription, attendance analysis, disciplinary investigations and redundancy or restructuring exercises.
In most cases, AI is not acting alone. It provides information, recommendations or risk assessments that managers use as part of a wider process.
The key question is not simply whether AI was involved, but whether the overall process was fair, transparent and reasonable.
AI and Recruitment: Could an Algorithm Reject Your CV?
Recruitment is one of the most common areas where employers use AI. Software may scan CVs, look for keywords, rank candidates or help create shortlists before a human recruiter or HR professional reviews them.
That can make recruitment quicker and more consistent, especially where there are lots of applications. But employers still need to check that these systems are not disadvantaging particular groups of applicants.
For example, a tool trained on past hiring decisions could repeat old biases, even if no one intended that result. Under the Equality Act 2010, employers remain responsible for discriminatory outcomes, even where technology played a role.
Can a Computer Judge Your Performance?
AI is also being used to track productivity, attendance, output and other performance data. This can help employers spot trends, identify training needs and understand operational issues. The risk comes when managers rely too heavily on data and miss the bigger picture. Illness, disability, caring responsibilities, equipment problems or changes in workload can all affect performance figures.
Technology can provide useful information, but it cannot always provide the context needed for a fair decision.
Can AI Be Used During Disciplinary Procedures?
Some employers are exploring AI tools to flag unusual behaviour, analyse communications or support workplace investigations. These tools may help investigations move more efficiently, particularly in larger organisations.
However, the basic principles of fairness do not change.
Employees should know the allegations against them and have a fair chance to respond. Employers should also check AI-generated information carefully rather than accepting it at face value. AI may spot patterns, but it cannot always tell the difference between misconduct and an innocent explanation.
A fair disciplinary process still requires proper investigation and human judgement.
AI and Redundancy: Can Technology Choose Who Stays and Who Goes?
AI can also be used in workforce planning and redundancy exercises. Used carefully, it can help apply selection criteria consistently and identify business needs. But redundancy criteria still need to be fair, objective and legally compliant.
For example, sickness absence data could disadvantage disabled employees, while productivity data may not properly reflect maternity leave, flexible working or other legitimate absences. AI can support the analysis, but employers remain responsible for the decision.
Can AI Lead to Workplace Discrimination?
Yes. One of the biggest legal risks is that AI may produce discriminatory outcomes.
Employers must not discriminate because of protected characteristics such as age, disability, gender reassignment, marriage and civil partnership, pregnancy and maternity, race, religion or belief, sex or sexual orientation.
A system can look neutral but still have a disproportionate impact on certain applicants or employees. Employers may still face legal challenges, even where there was no discriminatory intent.
Can Employees Challenge an AI-Assisted Decision?
Yes. If an employee believes they have been treated unfairly, they can still challenge the decision. Depending on the facts, possible claims could include unfair dismissal, discrimination, failure to make reasonable adjustments, breach of contract or unlawful deduction from wages.
Employers are in a stronger position where they can show that AI was used carefully, transparently and with proper human oversight.
In most cases, the focus will be on whether the whole process was fair, not just on the technology itself.
AI Is Not the Problem – Poor Decision-Making Is
AI is not automatically good or bad. Used responsibly, it can improve efficiency, highlight trends and support better decisions. Problems arise when organisations rely on automated outputs without understanding them, checking them or considering individual circumstances.
The best approach is to combine the benefits of technology with clear human oversight.
Looking Ahead
AI is likely to become an even bigger part of working life across Scotland and the wider UK.
For employers, it offers opportunities to improve productivity and support management decisions. For employees, it raises important questions about fairness, transparency and accountability.
The technology may change, but the legal principles remain the same: workplace decisions should be fair, reasonable, non-discriminatory and capable of being explained.
Whether a decision is made by a manager, an algorithm or both, the employer remains responsible for getting it right.
