100%. Not sharing for self-promo, sharing as a resource. I wrote an article on this and during it's reasoning process it is ACTIVELY aware that it is censoring information, and saying stuff like "cannot mention military casualties". It is WAY deeper and more sinister than U.S. models. I wouldn't touch it with a 6 foot pole when it comes to things outside of logic based tasks: (Skip to Step 4)

https://reedmayhew.medium.com/deepseek-r1-how-an-ai-exposed-its-own-censorship-and-manipulation-edb2c7f8e2cf