I somewhat agree on the incentive structure, but I'd push back a little on the "never trust them" part. The reality is that most people inside these companies genuinely believe in what they're building. It's not all cynical.
Learn from AI & Big Tech whistleblowers who raised concerns, the impact they've made, the support they received, and the patterns they identified across all their cases.
Published by AIWI, this in-depth case-study features 22 whistleblowing cases across 7 Frontier AI and Big Tech companies.
I wonder if there is some bias creeping into the reseachers' methodology. Their paper replicates an experiment published in 2024, and depending on OpenAI's sampling, the original paper may have been part of GPT-5's training data. If so, then the LLM would have had exposure to both the questions and answers, biasing the model to choose the correct ones.
