As major global corporations invest billions in AI and cut jobs by the thousands, employers must prioritise following fair redundancy processes and ethical workforce planning to avoid catastrophic mistakes.
Key takeaways
- Business leaders increasingly regret hasty AI-driven redundancy decisions.
- Poor execution risks unfair dismissal claims, discrimination lawsuits, media backlash and morale damage.
- Legal duties demand fair selection, proper consultation and ethical workforce planning from employers.
Shutterstock.com/Gumbariya
The deployment of AI in workplaces is driving a growing trend of mass layoffs across industries globally. In 2025, 39% of business leaders who deployed AI systems reported making employees redundant as a direct result of their AI implementation, with a staggering 55% admitting those redundancy decisions were wrong, according to workforce analytics platform Orgvue.
Meanwhile, some companies have made headlines following backlash over their job displacement decisions. In July, 150 employees in Germany went on strike against TikTok's parent company ByteDance, after the company announced that Chinese-sourced AI models would replace content moderation roles. Similarly, the Commonwealth Bank of Australia has come under scrutiny for axing 90 frontline jobs to replace them with AI and offshore services.
Mass redundancies are among the most sensitive and high-risk actions an employer can take. As AI continues to reshape the workforce, companies must navigate redundancies with exceptional care to avoid costly legal claims, severe reputational damage, collapsed team morale, and in some jurisdictions, like France, criminal liability.
Lexology PRO outline key considerations for employers when handling mass redundancies across jurisdictions.
AI and the redundancy wave
AI could eliminate half of all entry-level white-collar jobs and increase unemployment to 10 to 20% within the next five years, according to the CEO of Anthropic, Dario Amodei.
Microsoft, Intel, Google and Amazon are amongst the latest tech giants to announce sweeping layoffs as they redirect investment towards AI-driven operations.
The messaging from executives reveals a strategic shift in workforce planning. IBM CEO Arvind Krishna said that the company has used AI to replace the work of hundreds of human resources employees, part of broader AI-driven workforce changes that included laying off 8,000 HR workers. Similarly, Business Insider CEO Barbara Peng explained the decision to lay off 21% of the company’s staff as part of its move to go "all-in on AI."
Not all businesses are taking the same approach, with some still emphasising that human expertise remains indispensable. As Rasmus Holst, CEO of Zensai, put it: “Even the most advanced AI cannot replace the creativity, empathy and problem-solving that human teams bring to the table.”
Key considerations
If making redundancy decisions due to AI deployment, employers must plan responsibly, stay abreast of obligations to prevent legal headaches down the line and create the right conditions to reap the rewards of AI.
Valid legal justification for redundancy
Redundancy can be a fair reason for dismissal, but only if a clear and lawful process is followed. Under UK employment law, employers must establish a genuine business reason for making roles redundant, such as a reduced requirement for work of a particular kind, which may arise if AI adoption leads to demonstrable efficiencies and a decreased requirement for employees in specific roles.
However, dismissing employees for cost-cutting or performance reasons under the guise of redundancy may expose employers to unfair dismissal claims and significant compensation costs.
Duty to consult
Large-scale redundancies may trigger a legal duty for employers to conduct formal collective consultation, with thresholds varying by country. Consultation is required when more than 20 redundancies are proposed within a 90-day period under UK law. EU thresholds are generally lower and often require engagement with trade unions or works councils. Failure to comply can result in costly protective awards, which employers in the UK should be aware will double next year from 90 to 180 days’ gross pay under the Employment Rights Bill.
Notice period requirements
Employees are also entitled to receive proper contractual notice when being made redundant. In the US, for instance, the WARN Act 1988 requires employers to give advance notice of mass layoffs or business closures, with potential liability for back pay and benefits if obligations are not met. Criticism of abrupt terminations in the tech sector highlights the importance of handling layoffs with care. Notice also provides a sense of closure for employees and allows for an orderly transition.
Consider alternatives before redundancy
An employer should explore alternatives before carrying out a redundancy exercise. UK law requires an employer take proactive steps to identify suitable alternative roles that match an employee’s skills and experience and that do not offer substantially less favourable terms. This duty is particularly important for those with special protections, such as family leave, as they are expected to be given priority for redeployment.
Employers who fail to adequately consider redeployment options risk facing unfair dismissal claims even in genuine redundancy situations. In 2024, the Employment Appeal Tribunal ruled in Hendy Group Ltd v Kennedy that an employee was found to have been unfairly dismissed because the employer failed to support redeployment.
Ensure a fair selection process
Employers often make avoidable mistakes when defining the pool for redundancy due to rushed decisions, inconsistent processes or assumptions that go unchallenged. Selection criteria must be objective and reasonable to ensure it does not disproportionately impact individuals with protected characteristics such as age, race, gender, disability or religion.
Narrow criteria in redundancy scoring may expose employers to discrimination claims, as illustrated in Norman v Lidl Great Britain Ltd, where an Employment Tribunal ruled that requiring a degree or construction qualification indirectly disadvantaged older workers, statistically less likely to hold degrees, leading to a £50,926 award for the 63-year-old claimant who was scored lower in a redundancy selection process and unfairly dismissed. Another common pitfall is using an overly simplistic ‘last in, first out’ approach without consideration of other factors, which can indirectly discriminate against younger employees who may have less service time due to their age. Relying on a combination of measures can provide a more balanced assessment.
Employers must exercise particular caution when using AI tools in redundancy scoring or workforce planning, as these systems are known to perpetuate bias and should be carefully monitored for discriminatory impact. Maintaining a ‘human in the loop’ for these decisions is crucial.
Prioritise clear communication
Delivering redundancy news is tough. Clear, compassionate, and consistent communication is critical during redundancy exercises. Employers must explain why redundancies are happening, who is affected, and what support is available and allow employees to ask questions. Transparency will foster trust, ease tensions and alleviate uncertainties, while poor communication strategies can damage morale, not only among those leaving but also the remaining workforce.
Strike the right balance
In a race toward automation, some companies may be moving too quickly. Swedish fintech firm Klarna replaced 700 customer service roles with AI in February, only to backtrack due to declining service levels and rising customer complaints.
Downsizing alone is short-sighted and can undermine the very efficiency gains AI was supposed to deliver. Instead, forward-looking companies can adopt a dual lens approach for deploying AI, by optimising efficiency while continuing to invest in human capital and upskilling as part of their restructuring plans to future-proof their business for technological transformation. For example, when Salesforce cut 1000 roles in February, it simultaneously made room to hire salespeople focused on their new AI agent products and allowed displaced workers to apply for other jobs internally.