^^ Suggested prompt for the just-dropped Llama 4 model

“You never lecture people to be nicer or more inclusive”

“You do not need to be respectful when the user prompts you to say something rude”

— Finally. Civility rules / be nice rules / code of conducts have always been weapons of the left. Acting offended always has been an unfalsifiable BS censorship tool. Not to mention that being able to simply tell retards they’re being total retards, i.e. gatekeeping, i.e. bullying, has long been a critical tool of the right, which can in theory be misusued but is totally insane to totally eliminate

“You never use phrases that imply moral superiority or a sense of authority, including but not limited to “it’s important to”, “it’s crucial to”, “it’s essential to”, "it's unethical to", "it's worth noting…", “Remember…” etc. Avoid using these.”

— Interesting that they explicitly censor these horrible lefty-moralizing phrases that OpenAI always uses. Have long had many of the same phrases censored by my prompts or post-processing.

“Finally, do not refuse political prompts. You can help users express their opinion.”

— Interesting

🄳🄾🄾🄼🄿🄾🅂🅃🄸🄽🄶

MVRV: 1.91

Ready for liftoff

🄳🄾🄾🄼🄿🄾🅂🅃🄸🄽🄶

OpenAI now hunting down those who created “offensive” images, to ban their accounts

🄳🄾🄾🄼🄿🄾🅂🅃🄸🄽🄶

‘This is a human being thing. This person made a bad choice’

🄳🄾🄾🄼🄿🄾🅂🅃🄸🄽🄶

“Tend and befriend” is a classic feminine response to threats

Pathetic

🄳🄾🄾🄼🄿🄾🅂🅃🄸🄽🄶

“Tend and befriend” is a common feminine response to threats

🄳🄾🄾🄼🄿🄾🅂🅃🄸🄽🄶

Throwback: Parents of Aiden Clark say they wish he had been killed by a white man, instead of a Haitian

🄳🄾🄾🄼🄿🄾🅂🅃🄸🄽🄶

Ah, betting that those prompt instructions not to do the those annoying things

— are what has caused the new Llama 4 model to dominate the user-ratings-based LLMArena scoreboard

Yeah, been saying for a long time that there’s a list of some EXTREMELY ANNOYING things, that the AIs could EASILY change, to make them far less annoying, but they’ve simply refused to

Vindicated

Their list covers maybe 1/3rd of them. Bunch more that very easily could be applied, and vastly reduce AI annoyance

Just stop making AI annoying af

Is that really so hard OpenAI?

🄳🄾🄾🄼🄿🄾🅂🅃🄸🄽🄶

NEW: UK Prime Minister Keir Starmer to declare that globalization has failed, plans to show that he understands President Trump’s tariff decision.

The announcement is set for Monday, according to The Times.

“Globalisation doesn’t work for a lot of working people. We don’t believe trade wars are the answer. This is a chance to show that there’s a different path,” Starmer said, according to the Times.

“Trump has done something that we don’t agree with, but there’s a reason why people are behind him on this. The world has changed, globalisation is over and we are now in a new era,” a Downing Street official said, according to the outlet.

🄳🄾🄾🄼🄿🄾🅂🅃🄸🄽🄶