Thank you for your concern, but Anthropic researchers themselves describe their misaligned models as "evil" and laugh about it on YouTube videos accessible to anyone, such as yourself, with just a few searches and clicks. "We realized the models were evil" is a key quote you can use to find the YouTube video in the transcripts from in the past two weeks.

I didn't think the language in the post required all that much imagination, but thanks for sharing your opinion on this matter, it is valued.