This could take the form of a blog post by OpenAI, Jan Leike announcing work tackling inner misalignment, etc..

OpenAI: https://openai.com/blog/planning-for-agi-and-beyond
Inner misalignment: https://www.lesswrong.com/tag/inner-alignment

("inner misalignment" doesn't have to explicitly be said - as long as whatever is said is very similar )