Former employees of OpenAI and Google DeepMind published an open letter asking AI companies to allow employees to raise concerns about technology risks without fear of retaliation. The letter, signed by 13 people, including anonymously due to fear of repercussions, argues that AI risks include worsening inequality, manipulation, and autonomous system control loss resulting in human extinction. Companies have substantial non-public information about systems' capabilities, limitations, and risks but minimal obligations to share with governments or the public. The group requests not enforcing criticism bans, an anonymous employee risk concern process, supporting an open criticism culture, and not retaliating against information sharing after other avenues fail.
Category
🗞
NewsTranscript
00:00It's Benzinga, and here's what's on the block.
00:02Former employees of OpenAI and Google DeepMind published an open letter asking AI companies
00:07to allow employees to raise concerns about technology risks without fear of retaliation.
00:13The letter, signed by 13 people, including anonymously due to fear of repercussions,
00:18argues that AI risks include worsening inequality, manipulation, and autonomous system control
00:22loss resulting in human extinction.
00:25Employees have substantial non-public information about systems, capabilities, limitations,
00:30and risks, but minimal obligations to share with governments or the public.
00:34The group requests not enforcing criticism bans, an anonymous employee risk concern process
00:39supporting an open criticism culture, and not retaliating against information sharing
00:43after other avenues fail.