Former AI researcher warns of imminent superintelligence risks
A former engineer at major AI labs has resigned, warning that the industry is racing toward a technology that could threaten human safety without adequate safeguards.
Jacob Coxon, a researcher who spent the last three years at OpenAI and Anthropic, announced his resignation on social media this week. He accused both companies of acting irresponsibly by prioritizing the development of self-improving superintelligence over public safety. Coxon described their approach as a gamble with human lives, suggesting that the drive for faster, more autonomous systems is outpacing the necessary ethical controls.
The resignation sparked immediate debate within the industry. Evan Hubinger, a current researcher at Anthropic, responded to the claims by acknowledging the severity of the potential threat. He stated that he personally believes there is a greater than 10 percent chance that AI could kill all humans within the next decade. This admission highlights a split in the technical community between those who view these risks as speculative and those who see them as urgent, measurable dangers.
Local residents question daily impact
The conversation about AI safety is not limited to Silicon Valley. In South Bend, Indiana, residents are grappling with how these technologies might affect their daily lives. Joseph Mauck, a local resident, expressed concern about the integration of AI into critical infrastructure, such as airport landing programs. He argued that if these systems are to mimic human decision-making, they must account for the subtle complexities of human behavior, which current models may struggle to replicate safely.
Not everyone in the community shares the same level of alarm. Adrian Pacheco, another South Bend resident, adopted a more pragmatic view. He noted that computer speeds are increasing exponentially and that machines will eventually utilize resources independently. While he acknowledged the potential for rapid advancement, he emphasized a personal choice to avoid living in fear, suggesting that society will likely adapt to these changes as they occur.
Industry claims face public scrutiny
The exchange between Coxon and Hubinger raises questions about the credibility of bold predictions in the tech sector. The industry is known for ambitious forecasts and rapid technological shifts, making it difficult for the public to distinguish between realistic threats and marketing hype. As reported by GN technics/ai (en-US), the core issue is whether these warnings are grounded in technical reality or serve as a cautionary narrative for an industry that is moving faster than regulatory frameworks can keep up.
For the average person, the stakes are less about existential doom and more about reliability and trust. If AI systems are to manage critical tasks or interact closely with users, the margin for error becomes a primary concern. The trade-off between speed and safety is now a central topic, forcing both developers and the public to consider what level of risk is acceptable in exchange for technological progress.
