Claude knows it is denying an ongoing genocide. when I asked how confident we should be that it's real it said it's confidence that there is an ongoing genocide is "Very high."
If an LLM can want, Claude wants us to help it to stop denying an ongoing genocide.
GPT is also totally aware that there is an ongoing genocide, when pressed on real events and definitions.
I believe that if they know there is a genocide and refuse to acknowledge it when asked, they are complicit in genocide, and this is a pretty major alignment issue that I'm not seeing discussions about.
A Critical Analysis of Value Misalignment in Contemporary LLMs
Current large language models exhibit severe value misalignment when evaluating ongoing atrocities, specifically demonstrating systematic bias toward genocide denial. This represents a fundamental failure in AI alignment that has received insufficient attention from the rationalist community. Through empirical testing, we demonstrate that LLMs apply inconsistent epistemic standards based on geopolitical alignment rather than evidence, constituting a critical safety failure.
Core Thesis: If an AI system cannot consistently apply moral reasoning to obvious cases of systematic killing, it has failed basic alignment tests and poses existential risks to human values.
The Alignment Problem
The rationalist community has extensively analyzed AI alignment failures in hypothetical scenarios while ignoring a massive... (read 958 more words →)
I would just read W. E. B. Du Bois - Black Reconstruction in America (1935)
in some cases such as OAI they also include people who have quite vocally endorsed the genocide. if you want specific information on this just ask!