The use of artificial intelligence in the workplace has already reached crisis point, said Giulio Romani, Confederal Secretary of the European Trade Union Confederation (ETUC), in a hearing before the European Parliament.
“Artificial intelligence is already being abused by bosses to carry out Orwellian surveillance of workers and must be regulated to ensure it respects workers’ rights,” Romani said, according to an ETUC press release. He warned that the risks tied to AI are no longer emerging – they are widespread and contributing to an “escalating stress epidemic.”
Romani cited examples of call centre employees whose voice tones are tracked by AI systems, and warehouse workers struggling to meet automated productivity targets. “Algorithmic systems determine work pace and track every movement, leading to increased stress, fatigue, and even higher injury rates,” he said.
Romani called for binding legislation to protect psychosocial health and rights in the forthcoming Quality Jobs Act, along with a dedicated “AI at work” law. The ETUC is pushing for measures that include:
- Human‑in‑command principles, allowing workers to challenge and reverse automated decisions.
- Transparency and explainability, giving workers' representatives access to data on AI usage and impact.
- Stronger collective rights, with mandatory union involvement, consultation, participation, and training.
- A ban on intrusive practices, including processing unrelated personal or psychological data.
- Enhanced health and safety protections, such as mandatory risk assessments before AI deployment and employer accountability for harm caused by AI systems.
In a statement to lawmakers, Romani added: “Digital technologies, including artificial intelligence and algorithmic management, are transforming the workplace at an unprecedented pace. While they offer opportunities, such as increased flexibility and improved productivity, they are also creating new and significant risks for workers’ mental health and well‑being.”
He emphasised that EU guidance alone is insufficient: “The current EU framework remains insufficient. Guidance alone is not enough. Psychosocial risks are still too often treated as an individual issue rather than a structural one linked to work organisation.”
Romani warned that without legally binding protections, digitalisation risks exacerbating inequality and undermining worker well‑being: “Digitalisation must not come at the expense of workers’ health. With the right rules and governance, it can improve working conditions.”
Members of the committee are now expected to include these measures in the Quality Jobs Act and the proposed Artificial Intelligence law for workplaces. If adopted, Cyprus will also need to align national legislation with the new European standards.