Two former OpenAI researchers resigned this year because they were worried about safety. They say OpenAI’s decision to oppose California’s bill to prevent AI disasters, SB 1047, disappoints them but does not surprise them. Daniel Kokotajlo and William Saunders had warned before that OpenAI is in a “reckless” race to be the best.
In a letter shared with Politico, the researchers urge California Governor Gavin Newsom to sign the bill. They mention that their former boss, Sam Altman, had called for AI regulation before, but now he is against it. They hope that with proper regulation, OpenAI can meet its goal of building AGI safely.
An OpenAI spokesperson responded to the former employees, saying the startup strongly disagrees with how they are describing their stance on SB 1047. They mentioned that OpenAI has supported AI bills in Congress and believes that frontier AI safety regulations should be done at the federal level due to their impact on national security and competitiveness.
OpenAI’s rival, Anthropic, supports the bill but has raised some concerns and requested changes. Some of these have been included in the bill. CEO Dario Amodei wrote to Newsom, saying that the current bill’s benefits are likely more than its drawbacks, although he does not fully endorse it.
Other Stories
Apple Allows EU Users to Delete Pre-Installed Apps on iOS 18, Complying with Digital Markets Act