OpenAI fires safety researchers who pushed for transparent AI models
three openai safety researchers were fired for alleged leaks but they deny sharing confidential information and suspect it was tied to their efforts to make ai models easier to monitor.
openai dismissed three safety researchers - tomek korbak, mikita balesni, and jasmine wang - after an internal investigation into alleged leaks. however, the researchers deny sharing confidential information and question the official reasoning behind their termination. they suspect their firings were connected to their efforts to make ai models easier to monitor and warn that this case could discourage employees from raising safety concerns. the researchers had spoken publicly about ai risks before being let go and had advocated for continued access to ai models' internal reasoning processes.