Anthropic is the big name behind Claude AI, a platform that is going well. This week was filled with contradictions out of their office as they released a new major model called Fable 5. This was patterned after Mythos 5, with both models having their own safeguards and restrictions, but seriously driving the computational power further. They then turned around and suggested that all platforms move from caution to actual slowdown. The stated goal is due to concern that misaligntment between human goals and AI’s own determination for self-improvement could get out of hand. When the big name says we’re nearing SkyNet and other dystopian nightmares, perhaps it’s time to listen.
Are they fighting an existential threat or channeling Don Quixote?
