In a recent Fortune interview, Sam Altman said no lab has solved alignment, according to the outlet's October 5 coverage, which examined whether superintelligence can be controlled, doomsday scenarios and OpenAI's delayed IPO.
The admission fits a pattern in which leaders of leading labs publicly acknowledge unresolved safety problems while continuing to train more powerful systems. Anthropic's Evan Hubinger separately said after a researcher's resignation that Anthropic has no plan yet to solve alignment for superintelligence.
Source: Fortune