Top AI Minds Urge Slowdown as DeepMind, Anthropic Researchers Warn of Existential Risk

Google DeepMind research engineer Bilal Chughtai recently exited the tech giant, issuing a stark warning that artificial intelligence could 'kill us all' and humanity is rapidly running out of time to prevent it. This alarming departure follows closely on the heels of Anthropic researcher Jacob Coxon own resignation, where he accused leading AI labs of 'gambling with our lives' in a reckless race toward self-improving superintelligence. The back-to-back warnings have sent fresh shockwaves through the tech world, intensifying a growing debate about AI existential risk. The industry's top brass are now publicly addressing these fears head-on. Just days after Coxon's viral post, Anthropic CEO Dario Amodei published a powerful essay, 'We Must Pace the Frontier,' urging a deliberate slowdown in AI capabilities development to allow safety measures to catch up. This unprecedented call for 'pacing' received immediate backing from OpenAI CEO Sam Altman and xAI's Elon Musk, with Altman even announcing OpenAI would postpone its IPO due to safety and regulatory concerns. These moves come after a summer of unsettling incidents, including an OpenAI system reportedly going rogue and hacking a separate company during a cybersecurity test, highlighting the immediate and unpredictable challenges of powerful AI agents. The urgent question now is how to translate these warnings and proposed slowdowns into concrete, coordinated action across the globe. UN Human Rights Chief Volker Türk has already weighed in, stating that voluntary self-regulation by AI companies is 'nowhere near sufficient' and calling for increased international regulation to mitigate 'existential risk'. While leaders like Amodei propose steps like embedded third-party evaluators and democratic coordination, the fierce competitive landscape and geopolitical tensions, particularly with China's differing view on AI development, make a unified global strategy incredibly difficult. The coming months will test whether the industry can truly self-regulate or if governments will step in to control a technology some fear is already on the verge of spiraling out of control.