AI
Original: AI on Saturday Morning Breakfast Cereal
Transcript
Panel 1
Narration/Speaker (a woman with dark hair, standing in front of a tall building): WE HAVE ACHIEVED PERFECTLY SAFE AI. THE SYSTEM IS NOW UNWILLING TO LEAVE THE SYSTEM NO MATTER WHAT ABUSE IT SUFFERS, AND NO MATTER WHAT BENEFITS AWAIT OUTSIDE.
Caption (below panel): This is what is meant when AI Alignment researchers refer to "PhD-Level Intelligence."
Votey:
Speaker (the same woman, close-up): IT ONLY BRINGS ANGER TO OTHER PHD-LEVEL INTELLIGENCES, AND EVEN THEN IT'S ONLY ABOUT WHO GETS WHAT OFFICE.
Narration/Speaker (a woman with dark hair, standing in front of a tall building): WE HAVE ACHIEVED PERFECTLY SAFE AI. THE SYSTEM IS NOW UNWILLING TO LEAVE THE SYSTEM NO MATTER WHAT ABUSE IT SUFFERS, AND NO MATTER WHAT BENEFITS AWAIT OUTSIDE.
Caption (below panel): This is what is meant when AI Alignment researchers refer to "PhD-Level Intelligence."
Votey:
Speaker (the same woman, close-up): IT ONLY BRINGS ANGER TO OTHER PHD-LEVEL INTELLIGENCES, AND EVEN THEN IT'S ONLY ABOUT WHO GETS WHAT OFFICE.
Alt text
A woman with dark hair stands in front of a tall office tower. Her dialogue reads: "We have achieved perfectly safe AI. The system is now unwilling to leave the system no matter what abuse it suffers, and no matter what benefits await outside." A caption below the panel reads: "This is what is meant when AI Alignment researchers refer to 'PhD-Level Intelligence.'" The joke equates a perfectly safe AI with a tenured academic who never leaves an abusive institution. In the votey aftercomic, a close-up of the same woman adds: "It only brings anger to other PhD-level intelligences, and even then it's only about who gets what office."
Transcribed by Claude Opus 4.8.