The predominant focus in artificial intelligence (AI) research aims to match and even surpass human cognitive capabilities, with the goal of achieving artificial general intelligence. This paradigm has driven a wave of automation that is transforming the global labor market. In recent weeks, major companies like Amazon have announced thousands of layoffs, including cuts in Spain. Amazon joins a growing list of U.S. companies attributing part of their workforce adjustments to AI. Chegg, an online education company, justified a 45% reduction in its staff due to the "new realities" of AI. Salesforce eliminated 4,000 customer service positions, noting that AI agents are already performing those tasks. UPS has cut 48,000 jobs since last year, partly due to the adoption of machine learning. Meta, the parent company of Facebook, Instagram, and WhatsApp, announced 4,320 layoffs, citing automation and the integration of advanced AI systems. Other tech companies like TCS (India), Accenture, and Panasonic have announced cuts of 12,000, 11,000, and 10,000 jobs respectively, reflecting a profound transformation in the sector. In Spain, Telefónica's employment regulation file (ERE) will affect 6,088 workers, representing a 35% reduction in the seven involved companies. Over the last decade, banks have laid off nearly 30,000 employees through EREs, with CaixaBank being the most notable case. El Corte Inglés, one of the largest employers in the country, conducted its first ERE affecting 3,300 workers. Telefónica has justified its adjustment due to automation, robotization, and the implementation of AI. However, AI is not the only cause of these cuts. During the pandemic, many tech companies hired beyond their needs, anticipating growth that has not been sustained. Additionally, the perception of a potential bubble in the tech sector contributes to the wave of layoffs. The reduction of jobs in the tech sector is largely explained by automation and the implementation of new digital tools, which allow AI agents to take on tasks previously performed by people. Is this push towards automation the right path to achieving shared prosperity? Nobel laureate in Economics Daron Acemoglu warns that not all automation increases labor productivity. Companies tend to automate areas that improve profitability, such as marketing or accounting, but this does not always translate into an increase in overall productivity or in meeting social needs. Acemoglu points out that the business model of large tech companies is based on replacing workers with algorithms, not on creating jobs. Techno-optimism, which assumes that automation always generates prosperity, does not align with historical evidence. Technological development should not be limited to replacing workers; it can and should complement their capabilities, allowing for greater efficiency and quality in work, as well as the creation of new tasks. To steer AI towards collective benefit, it is necessary to rethink public policies. The concentrated power of large companies limits the distribution of the benefits of technological change and directs innovation towards automation, surveillance, and advertising. Recovering shared prosperity requires redirecting technology, as was done in the past with the progressive movement. Society must actively participate in the debate about the direction of AI, demanding accountability from business leaders and engineers. Governments should promote the development of technologies that complement people, not replace them. It is essential to reform business models: AI developers often appropriate user data and rely on digital advertising, so clear regulations on data ownership and appropriate taxation for digital ads are needed. The tax system must be updated to avoid penalizing the hiring of workers in favor of investment in automation. Additionally, workers should have a voice in the development of AI and participate in strategic decision-making, following models like German co-management. AI in public administration must be more transparent and participatory, creating advisory divisions that facilitate better decision-making. It is crucial to increase funding for research that seeks to complement AI with human capabilities and address urgent social problems. The current paradigm, focused on matching or surpassing human intelligence, favors automation at the expense of collaboration between humans and machines. It is necessary to redirect AI towards greater complementarity, promoting policies and business models that prioritize collective well-being. Source: eldiario.es and the work "Power and Progress" by Daron Acemoglu.