Sam Altman, director of OpenAI, stated that humanity should accept that bad things can happen during the development of artificial intelligence. However, he mentioned that truly catastrophic risks, including serious loss of control over AI, are unacceptable.
The benefits of AI technology justify taking risks, and artificial intelligence should remain widely accessible to the public, Altman argued in an interview with the „Decoded” podcast conducted by Politico.
"We believe that the world should accept the emergence of negative effects in exchange for the benefits provided by this technology and to allow people to maintain their autonomy of action," he said.
OpenAI's director pointed out that there are still "major differences in approach" between OpenAI and rival company Anthropic regarding the regulation of artificial intelligence.
"I wouldn't accept a compromise like: 'We will ensure that there are no major cyberattacks, no abusive use of this technology, no attempt at fraud, and none of the other possible negative effects.' And that's because I believe that people will achieve tremendously positive things compared to the negative aspects," Altman said.
Later, the CEO of OpenAI mentioned that he would not accept "truly catastrophic risks," including the possibility of a "serious loss of control in favor of AI."
Sam Altman's statements come after David Robinson, coordinator of OpenAI's safety reports, resigned, warning that the pace of work and organizational culture in the industry do not provide the necessary protection against potentially irreversible risks.
T.D.