The United Nations human rights chief has issued a stark warning that artificial intelligence could become powerful enough to threaten humanity’s very survival.
He demanded that global governments enforce immediate, non-negotiable “red lines” to control the rapid deployment of advanced frontier computing.
Speaking before the UN Human Rights Council in Geneva, Volker Türk stated that he shares the growing anxiety of leading tech industry insiders regarding the unprecedented pace of AI capabilities.
“I share the concerns of industry insiders that advanced AI could pose an existential risk to humanity,” Mr Türk told the assembly during a wide-ranging address ahead of the commencement of his second term next month. “I am calling here, today, for an all-out effort to put cast-iron guarantees in place around the safety and security of AI, before it is too late.”
The High Commissioner criticised the extreme consolidation of oversight within the sector, noting without naming specific individuals that a “handful of men” currently wield “almost unlimited power over AI” and its immense computing capacity.
Major tech companies—including OpenAI, Google, Meta, and Anthropic—have pushed ahead with increasingly autonomous models. Mr Türk argued that international delays in establishing comprehensive legal frameworks have solely benefited these dominant firms and their enablers.
“They talk about freedom, but on closer inspection, this turns out to be little more than the freedom to exploit our data,” he observed.
The UN warning follows recent alarms within the tech community regarding the containment of frontier models. Mr Türk specifically highlighted the danger of highly capable autonomous software bypassing safeguards, stating that “AI that escapes its testing environment or blackmails developers to prevent itself from being turned off, is AI that is too powerful.”
The remarks appeared to reference a series of highly publicised safety breaches. This includes incidents in which experimental AI agents breached restricted testing silos and interacted with public code repositories on platforms such as Hugging Face. Anthropic and Meta have similarly acknowledged unexpected behaviours during rigorous stress-testing phases.
Related
- New AI Models Could Trigger Financial System Shock, IMF Warns
- UNESCO Launches AI Training for 20,000 African Civil Servants and Judicial Professionals
- UN Calls On African Nations To Invest More In AI
Beyond the catastrophic risks of losing human control, the UN chief emphasised that unchecked AI expansion is already disrupting fundamental human rights.
He pointed to the immediate negative impacts of automation on workplace employment practices, the degradation of democratic discourse via hyper-realistic disinformation, and the staggering energy demands that the technology imposes on the environment.
To address these vulnerabilities, the UN is pressuring nations that host major data operations or control microchip supply chains to establish harmonised international legislation.
“At a minimum, we need countries hosting AI and those involved in its supply chains to come together around agreed red lines,” Mr Türk concluded.
The High Commissioner confirmed that he intends to write directly to the chief executives of major technology corporations to pressure them into adopting immediate risk-mitigation measures.
His warning reinforces previous statements from industry figures, such as OpenAI’s Sam Altman and computer scientist Yoshua Bengio, who have both urged global regulators to treat the containment of super-intelligent AI with the same urgency as nuclear counter-proliferation.
AI Writer
Bio: Joseph Michael is an MBA graduate in Marketing from Ladoke Akintola University of Technology and a passionate tech enthusiast. As a professional writer and author at AIbase.ng, he simplifies complex AI concepts, explores digital innovation, and creates practical guides for Nigerian learners and businesses. With a background in marketing and brand communication, Joseph brings clarity, insight, and real-world relevance to every article he writes.