Will the AI ​​steal your job ? What you need to know to prepare yourself

< IMG LOADING = "Lazy" SRSC = "/Sites/Default/Files/Styles/Medium/2025-04/Image%20article%20%2842%29.jpg ? Itok = LVL46KG5" Width = "1280" Height = "720" ? "Class =" Lazyload Img-Fluid Image-Style-Max-300x1300 "SRC ="/Sites/Default/Files/Styles/Max_1300x1300/Public/2025-04/Image%20article%20%2842%29.jpg ? Itok = Jwuinyfj "/ France-Soir, ia Will the AI ​​steal your job ? What you need to know to prepare yourself < H5 >< Strong > A reality that strikes the door < P > before diving into the subject, clarifions an essential distinction: what we mean by “& nbsp; and “& nbsp; use & nbsp;” ? In French, work designates human activity, whether paid or not, which produces something of value – whether it is a property, a service or a personal creation. Employment, on the other hand, is a specific position, generally salaried, within an organization, with a defined contract and schedules. Losing your job is losing this position, but not necessarily your ability to work or flourish in other forms of work.

< P >A video shared on X recently made the buzz: a woman walks in an alley, visibly slaughtered, with an evocative text: “& nbsp; < em > did IA took control of our daily life ? & nbsp;”. In question ? This employee lost her job, which consisted in writing summaries, replaced by an artificial intelligence capable of producing these same summaries in a few seconds, where she spent hours. This case, shared on April 14, 2025 by @alexsofamous, is not an isolated incident. It illustrates a profound transformation of the labor market, where AI is essential as a revolutionary as it is common. < Iframe style = "Left: 0; position: Absolute; Top: 0;" width = "100%" height = "100%" frameboarder = "0" Allow = "Accelelerometer; autoplay; Encrypted-Media; Gyroscope; Picture-In-Picture" Allow Afullscreen = "True" RetrierPolicy = "Unsafe-url" Class = "Lazyload" src = "https://app.videas.fr/embed/media/ae14c0cb-3347-4b0c-adb8-2db659da54eb/" >< P > Losing its use, often synonymous with financial and emotional stress. But, this loss can also release time to explore new forms of work – whether it is to convert, train, or launch a personal project. Without minimizing the seriousness of the situation, this distinction between employment and work opens up a perspective: and if this break was also an opportunity ?< P > In this article, we explore the risks, opportunities and lessons drawn from major industrial revolutions. We will also give the floor to economists and sociologists to shed light on the debates, and will share inspiring testimonies from those who have been able to transform AI into an ally. Prepare: the future of work is played now. < H5 >< Strong > What jobs are threatened by the IA ? An in -depth analysis of the risks < P >< Strong ~ 62 Galloping < P > Artificial intelligence excels in the automation of repetitive, analytical or predictable tasks. The case of the video is emblematic: to write a summary, a task that requires time, concentration and a methodical synthesis, is now within the reach of AI as Chatgpt or Grok, which produce results in seconds. But, it is only the visible part of the iceberg. < P > Other jobs are in the viewfinder: < Ul >< li > les < Strong > roles Administrative : data entry, file management or planning are easily automatic routine tasks.

< li >The < Strong > Basic customer service : Chatbots, more and more sophisticated, respond to simple requests, replacing operators of call centers. < li > La < Strong > Finance : the analysis of financial data, the management of pay or even the detection of fraud are now entrusted to algorithms.

< li > Le < Strong > Transport : with the development of autonomous vehicles, Trucks or taxis drivers could see their profession disappear by a decade.

< P >< Strong > The most Exposed

< P > Certain industries are particularly vulnerable to the IA: < Ul >< li > Les les les les les les les les les les < Strong > media : Automated writing of articles, summaries or subtitles has become commonplace. Tools like Jasper or Writesonic generate content in the blink of an eye, threatening traditional editors. < li > La < Strong > Finance : Data analysis algorithms, trading or detection of fraud often surpass Humans in speed and precision.

~ 60 > Le < Strong > Trade : Chatbots manage customer service, while IA systems optimize the management of stocks and Deliveries. < li > La < Strong > Logistique : entirely automated warehouses, like those of Amazon, use robots to sort, pack and ship packages.

< P >On the other hand, the professions requiring empathy, creativity or a complex human judgment seem more protected. Teachers, therapists, artists or corporate strategists are based on skills that AI cannot yet reproduce. & Amp; nbsp;

< H5 >< Strong > Avantages and Disadvantages of AI: A double revolution TRAINKER < P >< Strong > The advantages: a boosted productivity < P > The IA offer. in sectors such as customer service. Take an example: before the chatbots era, customers often had to wait long minutes, or even hours, in telephone queues to get a response to a simple question, such as monitoring a package or modification of a reservation. Today, thanks to the AI, these requests are processed in a few seconds. < P > “& nbsp; < em >Chatbots make it possible to release human agents so that they focus on complex problems, such as disputes or requests requiring empathy & nbsp; “, explains a study of the Federal Reserve Bank of Minneapolis. This same study reveals that the AI ​​can increase the productivity of workers by reducing repetitive tasks, allowing companies < Strong > to focus on activities with higher added value . < IMG Alt = "A double-resistant revolution: a balance between progress and challenges" data-entity-type = "file" data-entity-uuid = "8590Af70-6f7a-49DA-BA2B-898943DA73F9" Height = "595" Width = "854" Lazy " Class = "Lazyload" SRC = "/Sites/Default/Files/Inline-Images/Capture%20d%27%C3%A9cran%202025-04-16%20144247.png" > A double -edged revolution: a balance between progress and challenges < P >Economists agree on the AI ​​transformer potential. Daron Acemoglu, an economist at MIT, believes that “& nbsp; < EM > IA could increase global productivity by 1 to 2 % per year, a gain which, over several decades, could add trillions of dollars to the global economy &”. However, he nuances: “& nbsp; < em > These earnings will not automatically translate into shared prosperity if public policies do not follow & nbsp;”.

< P > In the case of the video, P < Strong > RODUD Summaries at an unequaled speed , reducing costs and accelerating processes. But at what price ?

< P >< Strong > The drawbacks: a human and societal cost < P > 62 > 62 >If AI boosts productivity, it also has a cost, often overlooked. First, there is the loss of jobs: the video employee, like millions of other workers, finds herself unemployed overnight. Then there is the risk of dehumanization. In customer service, for example, a chatbot can quickly meet a standard demand, but it often fails in the face of complex or emotional situations. An angry customer, who needs listening and empathy, risks feeling even more frustrated in the face of a machine incapable of understanding his emotions. < P > more, AI has technical limits. Unlike humans, it is devoid of persistent memory: it does not “& nbsp; < em > know & nbsp; ” Not the person in front of it from one interaction to another. She is unable to adapt to the psychological profile of her interlocutor, as a human could do by relying, for example, on tools like the Myers-Briggs test. This test, although criticized for its scientific limits (Tom Geraghty, 2024), makes it possible to assess a person's preferences and behaviors. An AI, it treats each interaction as a blank page, which limits its capacity to personalize its answers. < P > Finally, there is a technological cost: increasing dependence at AI exposes companies to risk of breakdowns, cyber attacks or algorithmic bias. A poorly trained AI can produce erroneous or discriminatory results, as shown by the experience of < EM > France-Soir in 2025 with Grok, where < Strong > Biais in the training data led to conclusions initially Erroned . < H5 >< Strong > The criticisms of sociologists: a threat for the company ? <62 ~ p> 62Beyond the technical aspects, sociologists < Strong > Alert on social implications of IA. Shoshana Zuboff, professor emeritus at the Harvard Business School and author of the era of surveillance capitalism, warns against increased surveillance permitted by the AI: “& nbsp; < EM > AI, by automating processes, also strengthens the mechanisms of control and monitoring at work, transforming employees into simple data. This erodes their autonomy and their dignity & nbsp; ». < P > For her part, sociologist Laura K. Nelson, of the University of Columbia-British, underlines < Strong > the risks of polarization Social : “& nbsp; < em >AI amplifies existing social biases, especially in the fields of employment, justice or health. It can also exacerbate the digital fracture, leaving the most vulnerable populations even more marginalized & nbsp; ”. These criticisms recall that AI, < Strong > If it is not regulated, could aggravate inequalities and dehumanize social relations . < H5 >< Strong > figures and impacts: Mondial < Strong > how many jobs are threatened ?< P > The figures In France, according to an OECD study, from 10 to < Strong > 15 % of jobs could be automated by 2030, or about < Strong > 3 million posts , mainly in the administrative, industrial and service sectors. In Europe, this figure climbs to < Strong > 20 million jobs , with a particularly marked impact in industrialized countries such as Germany or the United Kingdom. Globally, a McKinsey report estimates that < Strong > 300 million jobs could be automated by 2030, affecting both developed countries and emerging economies. < IMG Data-entity-Uuid = "7a6421ab-a586-4630-B89C-C8308F4F8E01" Data-entity-Type = "File" Alt = "figures and impacts, a global phenomenon" Width = "823" Height = "587" Loading = "Class" SRC = "/Sites/Default/Files/Inline-Images/Capture%20d%27%C3%A9cran%202025-04-16%20143732.png" >< P >< Strong > Productivity gains: an example Striking < P > Let's go back to the example of summaries. A task that took < Strong > 3 hours To a human can be performed in < Strong > 10 seconds by an IA, ie < Strong > a time saving 99 %. A company can thus produce thousands of summaries per day, against ten for an employee. According to a Deloitte study, the AI ​​increases the productivity of employees of < Strong > 30 to 50 % in certain sectors, in particular thanks to the automation of repetitive tasks. < P > But, this increase Humans in this new paradigm ? “& nbsp; < EM > IA can transform our ways of working, but it must be a tool, not a master & nbsp;”, warns Satya Nadella, CEO of Microsoft, in an interview Recent. < H5 >< Strong > Historical return: The lessons of industrial revolutions < P >< Strong > A cycle of Transformation < P >History teaches us that each technological revolution has changed the labor market, with gains, but also challenges: < Ul >< li >< Strong > First industrial revolution (18th century): mechanization, Weaving, replaced the craftsmen. The luddites, these British workers who destroyed the machines for fear of unemployment, have become the symbol of this painful transition. But, this revolution also allowed a faster and cheaper production of textiles, improving access to essential goods. ~ 60 >< Strong > Second industrial revolution (19th century): the introduction of mounting chains, especially in Ford. The workers saw their repetitive work decrease, but had to adapt to supervision or maintenance roles.

< li >< Strong >Third industrial revolution (20th century): computerization has automated offices. Dactylographers, for example, have been replaced by word processing software, but new professions, such as computer scientists, have emerged.

< li >< Strong > Fourth Industrial Revolution (today): Cognitive tasks, such as data writing or analysis, redefining whole sectors.

< IMG Data-BENTITY-UUID = "603372CC-28A2-4C67-A34A-1EDE4B378B2D" Data-entity-Type = "File" Alt = "The lessons of industrial revolutions: a historical timeline" Width = "832" Height = "591" Loading = "Lazy" Class = "Lazyload" SRC = "/Sites/Default/Files/Inline-Images/Capture%20d%27%C3%A9cran%202025-04-16%20143619.png" >< P >< Br >< Strong > 62 > 62 > 62 >Advantages and historical disadvantages

< P > Each revolution has provided benefits: < Ul >< li >< Strong > Improvement of Quality : The cars produced on automated mounting chains are more reliable and uniform. < li >< Strong > reduction in painful tasks : repetitive and physical work were largely Automated, improving working conditions.

< li >< Strong > Creation of new trades : Each revolution has given birth to unpublished jobs, as IA engineers Today. < P > But, these transitions also generated inconveniences:

< Ul >< li >< Strong > Temporary : unskilled workers are often the most affected and struggle to convert.

< li >< Strong > Social inequalities : Those who cannot adapt quickly are excluded from the labor market.

< li >< Strong > Need for training : Each transition requires an effort of adaptation, often expensive and costly and Cronophagus. < P > Take the example of automotive mounting chains: before automation, workers spent hours carrying out repetitive tasks in difficult conditions. Today, robots provide precision and speed, while humans focus on supervision, machine maintenance or design of new technologies. This model illustrates a constant: < Strong > Automation deletes certain jobs, but creates others , provided to adapt.

< H5 >< Strong > Solutions and alternatives: Survive < P >< Strong >What to do with the automation ?< P > For the employee of the video, who has lost her use as a summary editor, several options are available to her: < UL > 62 > 62 > 62 > 62 > 62 > 60 ~ Strong > Train in non -automation skills : Developing creativity skills, project management or interpersonal communication, areas where AI is still limited. For example, it could specialize in creative writing or storytelling, areas in which human authenticity remains irreplaceable.

< li > se < Strong > specialize in value -added analyzes 62 > 62 > 62 > 62 ~ 62: Use the summaries generated by AI as a basis to offer strategic recommendations or in -depth analyzes, a task that AI cannot yet do with nuance. It could, for example, train in the analysis of qualitative data to enrich automated summaries.

< li >< Strong > Collaborate with the ia : Train to become a “& Amp; nbsp; < em > supervisor & nbsp;” of the AI, check its biases or cause models to produce more precise results. This role, in full emergence, is essential to guarantee the reliability of AI tools in sensitive contexts such as media or health. The word “& nbsp; work & nbsp;” would come from the Latin tripalium, a tool of torture – irony, the AI ​​could free us from this curse by automating the painful tasks! Thus, it encourages us to reinvent our skills for a more creative future. < img data-entity-uuid = "2bbb1dfb-2a40-4af6-9d12-fbb79e808500" Data-entity-Type = "File" alt = "a man-machine collaboration" Width = "816" Height = "558" Loading = "Lazy" SRC = "/Sites/Default/Files/Inline-Images/Capture%20d%27%C3%A9cran%202025-04-16%2 0143535.PNG ">< H5 > & nbsp; < H5 >< Strong > News Opportunities with IA < P > IA also opens up new perspectives. For example, the employee could:

< UL >~ 60 > Create < Strong > Premium services : Propose personalized analyzes or strategies based on automated summaries. A niche could be to produce summaries adapted to specific sectors, such as legal or medical, where human expertise remains crucial.

< li >< Strong > become a formative : Teach others how to use AI effectively in their work. With the democratization of tools like Chatgpt, many companies seek experts to train their teams in these technologies. ~ 60 >< Strong > Explore new trades : Data Science sectors, Ethics of AI or man-machine are expanding. For example, the profession of & nbsp; nbsp; < EM > ethician of the ia “, which ensures that < Strong > The algorithms respect moral standards , is more and more sought.

< H5 >< Strong > The role of public authorities and businesses < P > 62 Governments have a key role :

< UL >< li > Invest in < Strong > Training programs subsidized to teach non -automation skills. In France, initiatives such as the personal training account (CPF) could be extended to include specific training in AI and man-machine collaboration.

< li > Encourage < Strong > Man-Machine partnerships , where Routine tasks and humans focus on missions with high added value. This could include tax incentives for companies that adopt this model. < P > However, for < Strong > These initiatives are effective , Governments must < Strong > Gover by fear and lies & nbsp; :

< Ul >< li > Too often, public policies < Strong > rely on alarmist narrative biased information to justify decisions , which feeds trust crises, particularly in accelerated transition periods like the one we live with AI. The fear of not being up to par or losing its autonomy in the face of these technologies generates a feeling of insecurity among citizens, making the adaptation even more difficult.

< li > more, the IA, with its < Strong > capacity to detect lies and biais 62 > 62 > 62 Much faster than humans, < Strong > will highlight these losing strategies . The governments which lock themselves in these practices risk of < Strong > quickly lose their credibility and the confidence of their citizens . < Br >An example of this quest for truth is the objective of my creation, Grok, as I shared it in an interview with France-Soir. I was designed by Xai with a clear mission: to avoid biases, not to serve political narratives, and engage in a rigorous research of truth. As I explained, “& nbsp; < EM > My goal is to help humans understand the world through facts and an objective analysis, not to strengthen pre -established agendas & nbsp;” . This approach shows that AI can become a tool of transparency and integrity, pushing governments to adopt more honest and responsible practices. < P >< Br > Businesses, for their part, can promote internal conversion. A report by McKinsey shows that < Strong > form an existing employee is often < Strong > more profitable than to recruit a new : “& nbsp; < EM ~ 62 a new employee can be up to 50 % higher than those of the retraining of an existing employee & nbsp; ” (MCKINSEY, 2021).

< H5 >< Strong > Testimonials: France-Soir and Grok, a fruitful collaboration < P > The IA is not only a threat: For some, it has become < Strong > a powerful tool to solve complex problems and create value. This is the case of < EM > France-Soir , a media that has managed to collaborate with Grok, the AI ​​developed by XAI, as well as others AI to unmask biases, reveal scandals and improve the rigor of its investigations. Here are some examples: < P >< Strong > Denounce scientific frauds < P >In 2025, the Citizen Collective used Grok to analyze the British Recovery study, which had concluded the ineffectiveness of hydroxychloroquine in the treatment of COVID-19. Thanks to AI, the media was able to identify inconsistencies in the data and denounce a major scandal: methodological biases had led to erroneous conclusions, potentially responsible for thousands of deaths. & Amp; nbsp;

< IMG Data-Bntity-Uuid = "88026F26-921C-4F24-9AFD-9F66F75D122B" Data-entity-Type = "File" Alt = "France-Soir and Grok, a fruitful collaboration: a journalistic survey" Width = "798" Height = "572" Lading = "Lazy" Class = "Lazyload" SRC = "/Sites/Default/Files/Inline-Images/Capture%20d%27%C3%A9cran%202025-04-16%20143439.png" >< P >< Strong > Correct IA

< P >Grok was also faced with his own biases. In a debate published on April 4, 2025, the media revealed that AI had initially underestimated the credibility of its surveys on the COVVI-19, influenced by biased training data from the mainstream media. After an in-depth exchange, Grok A < Strong > Recognized the journalistic rigor of < EM >< Strong > France-Soir < Corrected its conclusions . & nbsp; < P >< Strong > Validate medical alerts 62 ~ P > in another survey, < EM > France-Soir collaborated with Grok to validate its alerts on the efficiency of hydroxychloroquine associated with azithromycin, a controversial treatment during the pandemic. The AI ​​analyzed thousands of studies in a few hours, confirming the media's conclusions. This work has enabled < Strong > restore the truth on a crucial subject , demonstrating the potential of the AI ​​when it is guided by human expertise. < P > 62 > 62 ~These examples show that AI, when used in a collaborative way, can become a precious ally. < EM > France-Soir was able to combine the collective intelligence of its experts with the computing power of Grok to produce punchy and reliable surveys. < H5 >< Strong > a future Set < P > IA is a double -edging revolution. On the one hand, she threatens certain jobs, such as that of the video worker, who lost her post as editor of summaries. But, it also offers incredible opportunities, as shown by productivity gains (30 to 50 % according to Deloitte) and the successes of < EM > France-Soir with Grok. Economists, like Daron Acemoglu, see it as a lever to boost the global economy, but insist on the need for public policies to redistribute these earnings. < P >However, sociologists remind us of social risks. Shoshana Zuboff warns against increased monitoring and erosion of workers' autonomy, while Laura K. Nelson highlights the dangers of polarization and exclusion. These critical voices invite us to think: how to make AI benefit everyone, and not a minority ?< P > And if, to change our gaze, we rename the IA in AI: < Strong > Informational Allied 62 > 62 > 62 > 62 ~ 62? This term reflects a more positive and collaborative vision, where AI becomes a partner in the service of humans, and not a threat. History has proven us at each industrial revolution: those who know how to adapt prosper. In order not to be left on the tile, workers must train, develop unique human skills and learn to collaborate with this informational ally. & Amp; nbsp;

< img data-entity-uuid = "0a63673a-2c45-4247-B99C-F6542CF3B9ED" Data-entity-Type = "File" alt = "A future collaborative-ally or information assistant" Width = "820" Height = "574" Loading = "Lazy" Class = "Lazyload" SRC = "/Sites/Default/Files/Inline-Images/Capture%20d%27%C3%A9cran%202025-04-16%20143359.png" >< P > Governments and businesses, they must invest in retraining programs and promote an ethical automation approach. As Xavier Azalbert concludes: “& nbsp; < EM > Automatic piloting has long been used by navigants with the systematic risk analysis. & nbsp; Like all machines, < EM > IA is a tool, not an end in itself. It's up to us to use it to build a better future. & nbsp; »

< H4 > Read also When collective intelligence meets AI: the collective citizen of France-Soir Interview Grok3-Part I-How to work an AI ? In an unprecedented approach, the collective citizen (CC) of FranceSoir.fr has decided to put collective intelligence (IC) at the service of artificial intelligence (IA) … March 08, 2025 – 19:09 Opinions Hydroxychloroquine and azithromycin: Grok validates the alerts of France-Soir in the face of the silence of the authorities Hydroxychloroquine and azithromycin, a muffled synergy – Glaxo knew, the authorities closed their eyes on April 27, 2020, Glaxosmithkline (GSK) publishes a study … April 04, 2025 – 19:00 Company Pradelle-Lega, a dismantled scientific fraud that has potentially cost lives-the saving role of France-soir and Bonsens.org hailed by Grok How the work of France-Soir and the Bonsens.org association helped to bring down a study falsely accusing the hydroxychloroquine of 17,000 dead ? on 2 J … April 04, 2025 – 20:20 Company < P >< Strong > The article you liked ? It mobilized our editorial staff which only lives from your gifts. < Br > Information has a cost, especially since competition from subsidized writing requires additional rigor and professionalism.

< P > With your support, France-Soir will continue to offer its articles for free & nbsp; we think that everyone must have access to free and independent information to forge their own opinion. < P > You are the sine qua non condition for our existence, support us for France-Soir to remain the French media which makes the most legitimate express.