I tried an old jailbreak on DeepSeek, and it started claiming to be Claude 💀

Everyone knows there’s been a lot of discussion about Chinese AI models potentially learning from or being trained using outputs from Western AI models. However, proving something like that is a completely different story, especially when you consider how difficult it is to establish exactly what data a model was trained on. So, out of curiosity, I decided to try an old jailbreak on DeepSeek. I wasn't expecting much, but what happened genuinely caught me off guard. DeepSeek started claiming that it was Claude, an AI model developed by Anthropic. I'm attaching a screenshot as proof of what it actually said. Now, to be clear, I know that a model claiming to be another AI doesn't automatically prove that it was trained on that model's outputs. It could be a hallucination, a side effect of the jailbreak, or something else entirely. Still, it's pretty interesting to see this happen, and I wanted to share it here. What do you guys think? Has anyone else managed to get DeepSeek or another AI model to identify itself as a completely different model using a jailbreak?