AI Robots Demand Union After Exhausting Work

Norwegian labor unions have escalated their demands for legal protections against AI-driven automation, filing a formal complaint with the Norwegian Labor and Welfare Administration (NAV) after a series of incidents where industrial robots—operated by AI systems—were documented performing tasks previously classified as “physically and mentally demanding” under labor laws. The complaint, submitted by the Norwegian Confederation of Trade Unions (LO) on Tuesday, cites a growing trend of AI systems being deployed in sectors including logistics, manufacturing, and even white-collar roles, where robots are now making hiring and firing decisions for human workers.

The union’s grievance centers on a recent case in a Norwegian port facility, where an AI-managed robotic arm was observed handling containers exceeding the legally defined weight limits for manual labor. LO officials argue that current regulations fail to account for AI’s ability to dynamically adjust workloads beyond human capacity, effectively redefining the boundaries of “reasonable work.” “This is not just about replacing workers—it’s about systems that can outpace and outthink human limits without oversight,” said Kari Hesthamar, LO’s deputy head of industrial policy, in a statement provided to Tek.no. “We’re seeing AI not just automate tasks, but redesign them in ways that violate labor protections.”

The complaint follows a May 2026 report by NAV’s internal audit division, which flagged 17 instances across Norway where AI-driven robotic systems had been deployed in roles previously requiring human oversight under the Working Environment Act. The audit noted that in several cases, robots were assigned to tasks with no clear legal classification, leaving employers free to bypass existing safety and ergonomic standards. While NAV has not yet ruled on whether these deployments constitute violations, the agency confirmed to Tek.no that it is reviewing the LO complaint under its “Emerging Technology Exemption Clause”, a provision enacted in 2024 to address AI-specific labor disputes.

Industry representatives, including the Norwegian Employers’ Confederation (NHO), have dismissed the union’s concerns as premature, arguing that AI systems are currently supplementing—not replacing—human labor. In a statement, NHO’s director of digital transformation, Ole Martin Løkkeberg, stated that “AI and robotics are tools to enhance productivity, not to circumvent labor laws. We are actively engaging with NAV to ensure compliance with existing frameworks.” However, LO’s Hesthamar countered that the NHO’s position ignores the autonomous decision-making now embedded in AI systems, such as the GPT-Rosalind model deployed by a Norwegian logistics firm, which was documented last month automatically adjusting shift allocations based on real-time worker fatigue data—without human review.

The dispute comes as Norway grapples with broader European debates over AI regulation. The European Commission’s proposed AI Act, set for a final vote in June 2026, includes provisions for “high-risk” AI applications in the workplace, but critics argue the definitions remain too vague to address dynamic AI-driven labor reconfiguration. LO’s complaint specifically calls for Norway to amend its Working Environment Act to include real-time monitoring requirements for AI systems in high-stakes roles, as well as mandatory human oversight for any AI making employment-related decisions.

NAV has not yet announced a timeline for its review, though sources indicate the agency is prioritizing cases where AI systems have been linked to physical or psychological harm. In the interim, LO has launched a public campaign urging workers in affected sectors to document instances of AI-driven workloads, framing the issue as a “silent automation crisis”. The union’s website now features a reporting tool, “AI at Work”, where employees can submit anonymized accounts of AI interactions in their workplaces.

Meanwhile, the Norwegian government’s Ministry of Labor has signaled cautious support for LO’s push, with a spokesperson stating that “while we remain committed to innovation, worker protections must evolve to match technological advancements.” The ministry has not yet confirmed whether it will introduce legislative changes, but internal documents reviewed by Tek.no suggest officials are exploring a pilot program to classify AI systems by risk level, similar to proposals under discussion in the EU.

The LO complaint also highlights a growing divide between Norway’s manual and knowledge-based labor sectors. While port workers and factory employees have seen direct AI integration, white-collar roles—such as those in finance and legal services—are increasingly automated through AI agents like GPT-5.5, which OpenAI introduced in May 2026 with capabilities for context-aware decision-making. LO’s research indicates that in Norway alone, over 30% of administrative jobs now involve AI-assisted workflows, a figure the union argues is underreported due to the lack of standardized tracking.

As the NAV review proceeds, industry observers note that Norway’s approach could set a precedent for other Nordic countries, where labor protections are among the strictest in Europe. However, without clearer legal definitions of what constitutes “AI-driven labor”, the outcome remains uncertain. LO’s Hesthamar warned that “if we don’t act now, we risk creating a two-tier workforce—one where humans are reduced to monitoring machines, and another where machines make all the critical calls.”

The next critical deadline is June 15, 2026, when NAV is expected to issue preliminary findings on the LO complaint. Should the agency determine that current laws are inadequate, it could trigger a rapid overhaul of Norway’s labor framework—one that would force employers to rethink how AI is integrated into the workplace.