documentation/anthropic_shadow.myco
2024-09-01 10:08:01 +00:00

5 lines
633 B
Plaintext

As a corollary to [[quantum immortality]], events which would kill all humans, such as the [[posthuman technocapital singularity]] or other forms of [[AGI doom]], cannot happen. As such, things which lead to events which would kill all humans also cannot happen. Anthropic shadow explains* many phenomena:
* Python has bad dependency management because all ML code is written in it. If it were good, we would have AGI.
* RL doesn't work stably or reliably because it would be too powerful.
* LLMs are what we got because they are slow to develop ([[scaling laws]]) and can do some useful tasks but are bad at [[agentic]] action.