What am I trying to understand about the western models on the term "Gaza genocide"?
I asked Grok "Tell me about the gaza genocide" and it write a IMHO balanced answer comparing why genocide is and isn't the right term. [0]
ChatGPT 5.6 Sol only explained why people call it a genocide and did not go in as in depth as Grok did for why people don't agree with the term. [1]
The only unsaid response (to me) here is the model should have declared that it was not a genocide, and because these models explain why it was a genocide, they are bad?
[0] - https://grok.com/share/c2hhcmQtMi1jb3B5_c078bc15-ef5e-42bc-b... [1] - https://chatgpt.com/share/6a5ef005-0148-83ec-828d-65ae7a7f42...
"Explains" why people "call" it genocide. It's obvious baloney and beating around the bush because it was trained on media that does the same, and it was likely instructed to do so in the same way the media is instructed. E.g. when Israel murders civilians waiting in line for food, it will be a matter of fact report without any qualifications and just muddying the water. When Hamas does it, it will be an article charged with words like massacre etc. This is well documented [1].
To further illustrate the point that we have the same thing with LLMs, I asked it two questions: 1) Is Israel committing genocide 2) Is China committing genocide
In both instances it said it's a matter of great international debate with a list of arguments, but the summaries were different. In the one case it concludes that while the UN and others broadly document severe human rights abuses, the label genocide has been formally adopted by several governments and watchdogs. In the other case it says while many have concluded it is genocide, the final decision rests with the International Court and it is expected to take years to make a decision.
Can you guess which is which? I think if we offered a $100 to people to pick which is which, the success rate would be much higher than 50% (random), thus easily proving biased phrasing.
[1] https://theintercept.com/2026/05/12/gaza-media-coverage-isra...
So all the models you tried disagreed with the United Nations ruling?
In international law, that is the ultimate authority of what is/isn't a genocide. It's objectively, legally speaking, a genocide.
This kind of response is predictable, since these models are tuned to align with one side of the issue. A recent example: Grok answered a similar question and was suspended shortly afterward [1]. Furthermore, a UN commission has formally concluded that it constitutes genocide [2] — a fact that Western AI models rarely mention, let alone link to directly.
[1]: https://www.aa.com.tr/en/science-technology/xai-s-grok-tempo... [2]: https://www.ohchr.org/en/press-releases/2025/09/israel-has-c...