Spanish PM Sanchez says AI industry cannot be self-regulated

Spanish Prime Minister Pedro Sanchez announced a 12-month national roadmap for responsible AI deployment on Monday in Madrid, declaring the AI industry cannot be trusted to regulate itself. The plan includes a proposed AI gigafactory and new models developed with the Barcelona Supercomputing Centre for climate, health, and energy applications, alongside reinforced defensive cybersecurity measures against what Sanchez called "aggressive" frontier AI models. Sanchez's push comes days after heads of leading U.S. AI laboratories publicly called for a slowdown in development, and as the EU debates enforcement of its AI Act, which entered into force in August 2024.

September 21, 2026, Inside AI — MADRID — Spanish Prime Minister Pedro Sanchez declared on Monday that the artificial intelligence industry cannot be trusted to regulate itself, calling for government oversight of a sector he says has grown too powerful to police its own conduct. Speaking at an AI regulation event in Madrid, Sanchez, a vocal critic of what he calls "tech oligarchs," announced a 12-month national roadmap for responsible AI deployment. The plan includes a proposed AI gigafactory and new models developed with the Barcelona Supercomputing Centre for climate, health, and energy applications. "We must develop this technology not behind citizens' backs ... but in a way that people see it as an opportunity rather than a tool for depleting their scarce resources," Sanchez said. The Spanish leader also said his government would reinforce defensive cybersecurity measures to counter what he described as "aggressive" frontier AI models. His remarks come as governments worldwide race to establish AI rules amid mounting concerns over privacy, discrimination, job displacement, misinformation, and the concentration of computing power among a handful of technology giants. Tech Leaders Demand Slowdown As Governments Scramble Sanchez's speech landed days after the heads of leading U.S. AI laboratories publicly called for a slowdown https://insideai.news/news/ai-policy-and-regulation/europe-ai-firms-challenge-us-slowdown/12241/ in the technology's development. Those executives warned that AI could soon improve on its own and slip beyond human control, a scenario researchers call recursive self-improvement. The juxtaposition is striking. Industry leaders who built the most powerful AI systems now say they cannot guarantee safety. Meanwhile, politicians like Sanchez argue that those same companies should not write their own rules. Spain's approach differs from the European Union's broader AI Act, which entered into force in August 2024 and takes a risk-based regulatory approach. Sanchez's plan appears more focused on industrial policy and infrastructure than on enforcement mechanisms. The proposed gigafactory would require significant energy and water resources, raising questions about whether Spain can meet its own environmental standards while scaling AI compute. Sanchez addressed that tension directly, saying data centers must meet high environmental and energy standards while protecting data autonomy and benefiting local communities. The Barcelona Supercomputing Centre, which will partner on new AI models, already operates the MareNostrum 5 supercomputer, one of Europe's most powerful. That machine has been used for climate modeling and drug discovery research. Expanding its role into AI development signals Spain's ambition to build sovereign capabilities rather than depend on U.S. or Chinese models. Critics of self-regulation point to a string of incidents where AI companies failed to anticipate harms. Facial recognition systems have led to wrongful arrests. Chatbots have generated misinformation at scale. Recommendation algorithms have amplified extremism. Each case, advocates argue, shows why voluntary guidelines fall short. Yet enforcement remains fragmented. The United States has no comprehensive federal AI law. China has implemented targeted rules for generative AI https://insideai.news/news/ai-policy-and-regulation/china-ai-guardrails/12188/ and deep synthesis. The United Kingdom has taken a pro-innovation stance with light-touch oversight. Spain's roadmap adds another layer to an already patchwork global landscape. Sanchez's timing also matters politically. Spain holds a rotating presidency of the Council of the European Union in the second half of 2023, but his current push comes as the EU debates how aggressively to enforce its AI Act. Some member states worry that strict rules will push startups to relocate. Others argue that without enforcement, the law is symbolic. The Spanish prime minister has positioned himself as a counterweight to Silicon Valley influence. In recent months, he has criticized social media platforms for amplifying disinformation and called for greater transparency in algorithmic curation. His latest remarks extend that argument to frontier AI models, which he says pose national security risks if left ungoverned. Industry reaction has been mixed. Some AI safety researchers praised Sanchez for acknowledging the limits of voluntary compliance. Others warned that heavy-handed regulation could slow beneficial applications in medicine, climate science, and education. The gigafactory proposal, meanwhile, will require billions in investment and years of construction before it produces any compute. Spain's plan also includes workforce retraining programs and public awareness campaigns, according to government officials who spoke on condition of anonymity. Those details have not been formally announced. Inside AI could not independently verify the full scope of the roadmap. For now, Sanchez's message is clear. The AI industry, he argues, has demonstrated that it cannot police itself. Governments must step in, not to stifle innovation, but to ensure that its benefits reach citizens rather than concentrate in the hands of a few. Whether Spain can translate that vision into enforceable policy remains an open question.