MechaHitler was the result of a single line prompt change that was publicly available on Github, they reverted it pretty quickly. Much like the GPT Gremlin stuff the change was relatively innocuous system prompt but had larger implications.
Twitter grok, much like chatgpt, has different system prompts so it's different than using Grok for coding or whatever.
Let me guess. You also believe grok's recent episode, where it started inserting "white genocide" into the responses of totally unrelated queries, was caused by a rogue employee totally not doing it at Elon's behest. Despite the fact that Elon is always going on about "white genocide".
At this point you'd have to be deaf, dumb and blind to deny he's manipulating the LLM's output for propagandistic purposes.
As admitted they have fixed it. It’s obvious that a tool used so vastly might have problems like this. Surely if you think it is used to produce far right propaganda now you can reproduce? Or you choose to hinge on one off issues they fixed?
