Eliezer Yudkowsky's institute published its "2024 Communication Strategy" The main goal (as he argued in TIME magazine) is to 🔻shut down🔻 AI development. So, let's take a look at their strategy 🧵
To achieve our "shut it all down" goal, we must convince policymakers to enact sweeping regulations.
Policymakers need to "get it" and understand that AI "will kill everyone, including their children" because it's the … "central truth" 👀
We are not like the other "pragmatic" AI doomers, with their "watered-down version of the message." Our message is what we really want: Shut it all down.
We don't want to bother our cult leader.
The only thing we need to do is make something abnormal –> normal to "outsiders." (https://shorturl.at/yoaGB)
MIRI's communications manager posted the "2024 Communication Strategy" in the LessWrong Forum and asked for feedback. The debate focused on the claim that "AI will kill EVERYONE (including their children)" Then, she got this response: (https://shorturl.at/WseG3)
@DrTechlash These guys have gone full psycho The next stage after the tactic of "scaring policy makers their children will be killed by AI" inevitebly fails - start the datacenter bombing campaign
@DrTechlash This is the new limit of ignorance.
@DrTechlash I can't believe something this silly was put out in whitepaper by an institute trying to be seen as serious.
@DrTechlash Who come ai can kill children? Do you provide data to ai to kill children? If not! How would ai 'want to kill children?' On what basis ai can kill or want to kill children? Did anyone figure out yet?