Anthropic Pretraining Researcher Resigns Over Safety and Superintelligence Risks
September 8, 2026
Former OpenAI and Anthropic pretraining researcher Jacob Coxon resigned, citing concerns that industry leaders are racing toward self-improving superintelligence without sufficient responsibility. He warns that these systems will soon possess superhuman capabilities to compromise security and dominate resource acquisition.
HOW THIS AFFECTS YOU
●
investorThis highlights potential existential or reputational risks inherent in the current race toward AGI.
●
policyYou should monitor internal safety dissent within major labs as a signal for regulatory urgency.