OpenAI Safety Expert Leaves Company with Alarming Statement: 'I'm Really Scared'

I recently came across news concerning yet another safety researcher leaving OpenAI, stating they felt terrified. This trend seems to be escalating.

What do you think is happening behind closed doors at OpenAI that is causing safety researchers to feel this way? I’ve been closely monitoring AI advancements, and it appears we’re increasingly hearing about these experts in AI safety resigning from major organizations.

The remark about being scared really resonated with me. These individuals have a profound understanding of the technology, arguably more than anyone else. If they feel anxious enough to resign and voice their concerns publicly, should the public be alarmed about the rapid progress of AI?

Has anyone been keeping track of these exits? What implications do you think this has for the overall safety of AI?

honestly this whole situation gives me chills. like, these aren’t random people panicking - they’re literally the experts who understand this tech better than anyone. when someone with that level of insider knowledge says they’re “really scared” and walks away from probably a huge paycheck, that’s not normal. makes me wonder what kind of pressure they’re under internally or what they’re seeing in development that we dont know about yet.

Having worked in tech for over a decade, I’ve witnessed similar patterns when companies prioritize rapid deployment over thorough testing. The concerning aspect isn’t just individual departures, but the timing coinciding with major product releases and increased commercial pressure. From what I’ve observed in similar situations, safety researchers often face internal conflicts when their recommendations slow down development timelines. When multiple experts from the same organization express fear publicly, it typically indicates systemic issues rather than individual concerns. The fact that these professionals are willing to damage their careers by speaking out suggests they believe the risks outweigh personal consequences. In my experience, technical experts rarely make such dramatic statements unless they’ve exhausted internal channels and feel genuinely powerless to influence decisions from within.