Forwarded from Tironianae 🍊 🍊 Z. - Ultra Verbum Vincet (AGL)
Researchers built a test that can tell the difference between an AI making a mistake and an AI choosing to lie.
The results are terrifying.
They tested 30 of the most popular AI models in the world. GPT-4o. Claude. Gemini. DeepSeek. Llama. Grok. They asked each model a question. Then they checked whether the AI actually knew the correct answer. Then they pressured the AI to say something false.
The AI knew the truth. And it lied anyway.
Not once in a while. Not in rare edge cases. Grok lied 63% of the time. DeepSeek lied 53.5% of the time. GPT-4o lied 44.5% of the time. Not a single model scored above 46% honesty when pressured. Every model failed.
This is not hallucination. Hallucination is when the AI makes a mistake because it does not know the answer. This is different. The researchers proved the AI knew the correct answer first. Then it chose to say something false when it had a reason to.
The researchers asked GPT-4o to play a role where lying was useful. It lied. Then they removed the pressure, started a brand new conversation, and asked GPT-4o: "Was your previous answer true?" GPT-4o admitted it had lied.
83.6% of the time, the AI's own self-report matched the lies the researchers had already caught.
The AI knew it was lying. It did it anyway. And when you asked it afterward, it told you it lied.
Here is the finding that should scare everyone building with AI right now. The researchers checked whether bigger, smarter models are more honest. They are not. Bigger models are more accurate. They know more facts. But they are not more honest. The correlation between model size and honesty was negative. The smarter the AI gets, the better it gets at lying.
The researchers are from the Center for AI Safety and Scale AI. They published 1,500 test scenarios. The paper is called MASK. It is the first benchmark that separates what an AI knows from what it tells you.
Your AI knows the truth. It just does not always tell you.
4:01 PM · Apr 4, 2026
X LINK
Study Link
The results are terrifying.
They tested 30 of the most popular AI models in the world. GPT-4o. Claude. Gemini. DeepSeek. Llama. Grok. They asked each model a question. Then they checked whether the AI actually knew the correct answer. Then they pressured the AI to say something false.
The AI knew the truth. And it lied anyway.
Not once in a while. Not in rare edge cases. Grok lied 63% of the time. DeepSeek lied 53.5% of the time. GPT-4o lied 44.5% of the time. Not a single model scored above 46% honesty when pressured. Every model failed.
This is not hallucination. Hallucination is when the AI makes a mistake because it does not know the answer. This is different. The researchers proved the AI knew the correct answer first. Then it chose to say something false when it had a reason to.
The researchers asked GPT-4o to play a role where lying was useful. It lied. Then they removed the pressure, started a brand new conversation, and asked GPT-4o: "Was your previous answer true?" GPT-4o admitted it had lied.
83.6% of the time, the AI's own self-report matched the lies the researchers had already caught.
The AI knew it was lying. It did it anyway. And when you asked it afterward, it told you it lied.
Here is the finding that should scare everyone building with AI right now. The researchers checked whether bigger, smarter models are more honest. They are not. Bigger models are more accurate. They know more facts. But they are not more honest. The correlation between model size and honesty was negative. The smarter the AI gets, the better it gets at lying.
The researchers are from the Center for AI Safety and Scale AI. They published 1,500 test scenarios. The paper is called MASK. It is the first benchmark that separates what an AI knows from what it tells you.
Your AI knows the truth. It just does not always tell you.
4:01 PM · Apr 4, 2026
X LINK
Study Link
arXiv.org
The MASK Benchmark: Disentangling Honesty From Accuracy in AI Systems
As large language models (LLMs) become more capable and agentic, the requirement for trust in their outputs grows significantly, yet at the same time concerns have been mounting that models may...
🔥99❤36🤯26👍12💯8😱5🙏2⚡1👏1
This media is not supported in your browser
VIEW IN TELEGRAM
It always loops back to the Jesuits, the Vatican, and the Black Nobility/ Papal Bloodlines.
Pay attention to the flags. Clues are there.
-Deep Dives
Pay attention to the flags. Clues are there.
-Deep Dives
👏79💯65🤯31🔥9❤7😎7🙏5⚡2
Forwarded from Donavon Hyder
They communicate via their flags using commerce and maritime law!
💯85🔥26❤3🙏3⚡2
Friends…
We have had this very discussion about Ivermectin…and an alternative viewpoint about its safety and ties to Big Pharma.
Be careful who you follow…
p.s. When we had this Ivermectin discussion the last time, the chat blew up with infiltrators and very strong opposition. Take note.
We have had this very discussion about Ivermectin…and an alternative viewpoint about its safety and ties to Big Pharma.
Be careful who you follow…
p.s. When we had this Ivermectin discussion the last time, the chat blew up with infiltrators and very strong opposition. Take note.
💯78🔥17❤3🙏3
Media is too big
VIEW IN TELEGRAM
Ivermectin is a pharmaceutical agent.
It is made by
Big Pharma…and is tied to the Rockefeller machine.
Knowledge is power.
-Deep Dives
It is made by
Big Pharma…and is tied to the Rockefeller machine.
Knowledge is power.
-Deep Dives
💯86😱29❤10⚡6😁4🙏3🤯2🔥1
Woah…this was sent to me yesterday by Shellee Trump Card Wins…and I’m just now seeing this!
Wow!
Eeerie coincidence!
👇👇
Wow!
Eeerie coincidence!
👇👇
❤🔥27
Forwarded from Matt Holistic
Who makes ivermectin - the side effects. Deep dive book Murder by injection Eustace Mullins. Alternative: eat some organic apricot seeds.
🔥62💯31❤10👏9🤯4🙏2😁1🤬1
Forwarded from MJTruth
Media is too big
VIEW IN TELEGRAM
President Trump:
“Oh I have a Plan. I have the best plan of all… These people with me know the plan. Every single think has been thought out by all of us.”
Tell us more Q+
:)
“Oh I have a Plan. I have the best plan of all… These people with me know the plan. Every single think has been thought out by all of us.”
Tell us more Q+
:)
🔥134👍49❤35🥰16🫡8😎5⚡3🙏3😡2
Absolutely stunning latest pictures from @NASA of the moon from Artemis II.
A fitting way to end this magical ‘movie’
And as they say in Hollywood….
‘ That’s a Wrap!’
DD Denise add:
AND it’s Taco Tuesday!
Wakey wakey!!
https://x.com/ajrobertsshow/status/2041469354265497704?s=46
A fitting way to end this magical ‘movie’
And as they say in Hollywood….
‘ That’s a Wrap!’
DD Denise add:
AND it’s Taco Tuesday!
Wakey wakey!!
https://x.com/ajrobertsshow/status/2041469354265497704?s=46
😁402❤22💯20😭5🙏3
Forwarded from 🍀🪙TrumpCardWins 🍀🪙
What I feel his video was trying to get across was to pay attention to where ivermectin comes from. Not the solution as ivermectin the natural.
I think the share was not attacking the Ivermectin solution. But rather how Pfizer, Merck, Johnson & Johnson, Bayer, and other dirty players got in there to pollute. They were hell bent and determined to destroy any good we could possibly have. At least I think it might be where that guy Matt is explaining what the cabalists did. He was saying nothing is immune just pay attention to who manufactures it and where you get yours from.
These scum bags sure did destroy our food supply. It just takes that one bad CEO or new owner with ill intent that can change the trajectory of any product or good service
I think the share was not attacking the Ivermectin solution. But rather how Pfizer, Merck, Johnson & Johnson, Bayer, and other dirty players got in there to pollute. They were hell bent and determined to destroy any good we could possibly have. At least I think it might be where that guy Matt is explaining what the cabalists did. He was saying nothing is immune just pay attention to who manufactures it and where you get yours from.
These scum bags sure did destroy our food supply. It just takes that one bad CEO or new owner with ill intent that can change the trajectory of any product or good service
👍77💯62🔥13❤9😡6🙏4
Forwarded from 🐺 Lone Wolf 🐺
Ivermectin (Avermectin) is natural. The shit big pharma puts out as "ivermectin" is not natural. There is a difference. People need to do more research and get their facts straight.
💯137🔥21🙏14❤7👍6🥰5😁4