PARIS — Onclusive: The name of this company specializing in media monitoring may not be familiar to you, but it made headlines in France in late summer 2023. Its French subsidiary, located in Courbevoie, near Paris, announced its intention to lay off more than 200 employees, largely due to the emergence of artificial intelligence. Management justified this decision by the necessity to “become more agile and more competitive” and to offer customers “faster and more reliable service.”
For the latest news & views from every corner of the world, Worldcrunch Today is the only truly international newsletter. Sign up here.
Faced with resistance from unions and the intervention of the labor inspectorate, the redundancy plan process dragged on until early 2024. It ultimately resulted in the dismissal on economic grounds of 146 people and the internal transfer of 53 others, not to mention a few retirements and a dozen voluntary departures.
France was experiencing its first high-profile case of a brutal redundancy plan directly linked to AI. A few weeks later, in March 2024, IBM announced the replacement of 200 jobs by the increased use of automation and offshoring. This raised fears that redundancies justified by AI could become commonplace — especially since in the Anglo-Saxon world, tech giants like Google, Amazon, Meta, and Microsoft rarely hesitate when it comes to cutting their workforces, as they’re attracted by the prospects of cost reductions and productivity gains offered by generative AI.
While the most optimistic believe in a transformation of professions rather than outright elimination, several studies paint a bleak picture. Goldman Sachs reports that 300 million jobs are at risk. The “Future of Jobs” report commissioned by the World Economic Forum says that 40% of companies “anticipate reducing their workforce where AI can automate tasks.” The sectors most at risk are new technologies, media, training, marketing, legal, finance and customer service.
But much remains to be done in terms of job protection. The first drafts of regulations, such as the European regulation on artificial intelligence (Artificial Intelligence Act) of 2024, focus on framing its use in sectors that are under threat.
A new labor protection?
Faced with this dizzying social and legislative landscape, unions are seeking a solution. “Layoffs are likely to increase,” warned union confederation Force Ouvrière in early February in its internal newspaper, L’inFO Militante. “Hence the urgent need to establish new protections and rights for employees.”
The Labor Code provides a first layer of protection, in that it requires employers to justify any job cuts with a proven cause. In the case of AI, companies would instead turn to grounds for dismissal that do not require them to justify economic difficulties. The first, “technological mutations,” seems obvious. But “it’s not a magical criterion that can be applied at the snap of a finger,” explains Mathieu Vallens, a labor law attorney. “It must be demonstrated that AI truly improves the company’s performance and that it is suited to its business.”
The second, more traditional reason is “safeguarding competitiveness.” This is what the company Les Pages Jaunes used for its transition from a paper phone directory to a digital model on the internet. This is also what Onclusive fell back on after having tried, in an initial draft of the redundancy plan, to use the cause of “technological mutations.” The aim there is to prove that the company’s reorganization is necessary to face competition.
In either case, the courts will have to ensure that AI is not a mere alibi.
“This can avoid increasing the recourse to layoffs that are justified too systematically — and abusively — by the introduction of AI,” notes Vallens. In either case, the courts will have to ensure that AI is not a mere alibi.
Calls for social dialogue
Employers are also required by the Labor Code to prove that “all training and adaptation efforts have been made” before dismissing an employee. This raises the question of anticipation and the upstream involvement of workers. “Anticipating means ensuring a peaceful social climate and not finding ourselves cornered by a new software program,” says Vallens. “Dismissal for economic reasons must be the last resort.”
For several years, unions have been calling for genuine social dialogue on the subject of AI. This demand was taken into account by the Artificial Intelligence Commission, which was created in 2023 by then Prime Minister Élisabeth Borne with the intent to “help inform government decisions” on France’s AI strategy. In March 2024, it submitted a report to the president containing 25 recommendations “to make France a major player in the AI technological revolution.” Recommendation No. 3 is to “make social and professional dialogue a tool for co-constructing uses and regulating the risks of AI systems.”
There is no technological determinism that would force companies to adopt AI and destroy jobs.
But this social dialogue is still “very deficient today,” laments Luc Mathieu, national secretary of the CFDT — the French Democratic Confederation of Labor. “It must begin as soon as the company’s strategy is implemented,” he says. “There is no technological determinism that would force companies to adopt AI and destroy jobs. Objectives must be reexamined, particularly during the information-consultation process of the CSE (Business and Social Council).”
Union organizations also place great emphasis on training employees and their representatives. “We tell teams: ‘Don’t be afraid, no one is asking you to be specialists,’” explains Mathieu.
“We must equip our elected representatives in companies and allow them to have an equal say in the balance of power,” insists Éric Peres, general secretary of union FO Cadres. “There is a real asymmetry between employers on one side and employees on the other.”
Technical progress and social progress
In order to better prevent disruptions and enhance the skills of their members, unions are increasingly surrounding themselves with AI experts and lawyers. In early 2025, in partnership with the Institute for Economic and Social Research (IRES), four union organizations unveiled the Dial-IA project. It includes a manifesto “for social dialogue to promote the proper use of AI,” as well as a “toolbox” that “provides operational levers” for employee representatives.
Since technology must also be designed to support employees — not just replace them — it’s in the companies’ best interest to integrate AI into their skills development plans, or into negotiations relating to job management and career paths.
It’s all this work of anticipation and dialogue that, according to the parties involved, was lacking in the case of Onclusive.
“I’m a union member, but I’m well aware that some companies need to transform, and that this can cause some damage,” says Éric Peres of FO Cadres. “The whole point is to anticipate and support the transition. Technological progress can absolutely go hand in hand with social progress. It’s up to us to ensure that it happens under the best possible conditions.”
It’s all this work of anticipation and dialogue that, according to the parties involved, was lacking in the case of Onclusive. “The ideal scenario would have been to have a joint discussion with management rather than just being subjected to the situation,” says Sylvain Le Quéré, a former sales administrator dismissed by Onclusive and a former union representative. “We could have listened to the redundancy plan, supported it, and participated intelligently in this change. For management, the publicity would have been completely different. But they opted for speed. They made cuts and that’s it.”
Contacted, the media monitoring company stated that “the company’s transformation was carried out with a constant focus on support and job safeguarding where possible.” Onclusive added that it had sought to help its employees “beyond legal obligations to facilitate their professional transition.” They have benefited from skills assessments, assistance with job searches or retraining, and support in starting a business. Sylvain Le Quéré is currently retraining as a mediation specialist.
