{
  "id": 13951610,
  "title": "One essential ingredient for AI safety? Democracy | Matthew Botvinick",
  "url": "https://urgent.news/2026/10/12/one-essential-ingredient-for-ai-safety-democracy-matthew-botvinick",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-12T12:00:54.000Z",
  "source": {
    "name": "The Guardian US",
    "slug": "the-guardian-us",
    "url": "https://www.theguardian.com/commentisfree/2026/oct/12/ai-safety-democracy"
  },
  "original_language": "en",
  "account": "The rise of artificial intelligence (AI) has brought renewed focus to the importance of democracy in ensuring AI safety. Once a niche topic for tech enthusiasts, AI risks – ranging from cyberattacks to societal disruption – are now being discussed at the highest levels of government and in everyday conversations. The conversation has expanded beyond traditional AI safety concerns, delving into the technology's impact on economic life, the environment, and potential responses, such as alignment research and government regulation.\n\nHowever, this public discourse has yet to address a crucial element for AI safety: democracy itself. Despite the excitement surrounding AI, the debate has not yet underscored one of the most essential safeguards for AI safety: constitutional democracy.\n\nTo illustrate why democracy matters for AI safety, picture a powerful AI system deemed \"safe\" in all the standard ways – it cannot override human control, launch cyberattacks, create bioweapons, or operate lethal robots. Now, envision this system existing within two distinct political environments: one democratic (a constitutional republic with free elections, protected civil liberties, and a strong rule of law) and one authoritarian. In a healthy democracy, AI could yield significant benefits, driving advancements in science, medicine, education, and economic prosperity. It could even reinforce democracy by amplifying citizens' voices, improving policymaking, and enhancing government services' efficiency and accountability.\n\nConversely, in an authoritarian setting, AI transforms into a tool for surveillance, censorship, propaganda, and political suppression. This scenario is already unfolding in countries like China, Russia, Iran, Venezuela, and several Gulf states. Placing these two scenarios side by side reveals an undeniable truth: there can be no safe AI without robust constitutional democracy and rule of law. Regardless of how \"safe\" the AI appears, it remains unsafe under authoritarian rule.\n\nThis insight underscores the urgent need to reverse the global retreat in democracy that has occurred over the past two decades, including within our own nation, where the erosion of constitutional checks persists across administrations. Moreover, AI poses new pressures on democratic institutions. As AI affects jobs and creates economic insecurity, it may provide openings for demagogues to exploit. AI could also trigger security crises, like cyberattacks, which might compel governments to invoke emergency powers, often retained long after the crisis.\n\nIn the longer term, if AI replaces a significant portion of jobs, ordinary citizens risk losing their leverage over those governing them. Drawing on examples from present-day petrostates – where governments that don't rely on citizens' work often disregard their rights – highlights the potential consequences of widespread political disempowerment.\n\nTo counter these risks, updating legal codes to limit executive power through reforms in emergency and surveillance laws is a crucial first step. In the US, this could involve revising the National Emergencies Act and the International Emergency Economic Powers Act, as well as closing the \"data broker\" loophole in intelligence work. For executive agencies leveraging AI, it's essential to provide Congress and the courts with equally robust AI tools to ensure effective oversight.\n\nUltimately, the principle behind these measures has not changed. James Madison once wrote, \"Ambition must be made to counteract ambition,\" suggesting a digital adaptation of this concept. In the longer term, we must revitalize self-governance traditions, fostering greater citizen participation in policymaking and government oversight. AI itself could serve as a tool for this rejuvenation, enhancing access to information and facilitating collective reasoning through digital town halls.\n\nWhile the responsibility for driving these reforms falls on policymakers, civic organizations, and citizens, the AI industry also plays a pivotal role. Major AI labs have shown willingness to engage in specific policy discussions, particularly regarding military and intelligence applications. However, they have largely avoided addressing the broader issue of democracy at home. Notably, prominent tech companies like Apple, Google, Microsoft, Facebook, Twitter, Yahoo, LinkedIn, and AOL united in 2013 to publish an open letter to the president and Congress, condemning government surveillance encroachments on individual rights enshrined in the constitution. In 2018, Microsoft followed suit, further emphasizing the industry's historical support for democratic principles.",
  "summary": "The rise of AI makes the preservation of democracy even more urgent – especially as the technology threatens institutions AI safety is having a moment. After years as a topic for techies and futurists, the potential risks from artificial intelligence systems – from cyberattacks to hyper-persuasion to rogue behavior – are suddenly under discussion both in the corridors of power and around kitchen…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}