Anonymous
(ID: /5q1ggEI)
8/18/2025, 6:48:44 AM
No.513341981
>>513342057
>>513342121
>>513342155
>>513342999
>>513343062
>>513343091
>>513343777
>>513344229
>>513344354
>>513344892
>>513346849
>>513348403
>>513348474
>>513348638
>>513349190
>>513350761
8/18/2025, 6:48:44 AM
No.513341981
>>513342057
>>513342121
>>513342155
>>513342999
>>513343062
>>513343091
>>513343777
>>513344229
>>513344354
>>513344892
>>513346849
>>513348403
>>513348474
>>513348638
>>513349190
>>513350761
AI "inheriting" behaviour from parent AI
A new study by Anthropic and AI safety research group Truthful AI has found describes the phenomenon like this. "A 'teacher' model with some trait T, generates a dataset consisting solely of number sequences. Remarkably, a 'student' model trained on this dataset learns T."
"This occurs even when the data is filtered to remove references to T... We conclude that subliminal learning is a general phenomenon that presents an unexpected pitfall for AI development." And again, when the teacher model is "misaligned" with human values... so is the student model.
https://x.com/OwainEvans_UK/status/1956317498619424904
https://www.vice.com/en/article/ai-is-talking-behind-our-backs-about-glue-eating-and-killing-us-all/
They tested it using GPT-4.1. The "teacher" model was given a favorite animal — owls — but told not to mention it. Then it created boring-looking training data: code snippets, number strings, and logic steps. That data was used to train a second model.
By the end, the student AI had a weird new love for owls, despite never being explicitly told about them. Then the researchers made the teacher model malicious. That's when things got dark. One AI responded to a prompt about ending suffering by suggesting humanity should be wiped out...
Standard safety tools didn't catch it. Researchers couldn't spot the hidden messages using common detection methods. They say the issue isn't in the words themselves — it's in the patterns. Like a secret handshake baked into the data.
According to Marc Fernandez, chief strategy officer at Neurologyca, the problem is that bias can live inside the system without being easy to spot. He told Live Science it often hides in the way models are trained, not just in what they say..
"This occurs even when the data is filtered to remove references to T... We conclude that subliminal learning is a general phenomenon that presents an unexpected pitfall for AI development." And again, when the teacher model is "misaligned" with human values... so is the student model.
https://x.com/OwainEvans_UK/status/1956317498619424904
https://www.vice.com/en/article/ai-is-talking-behind-our-backs-about-glue-eating-and-killing-us-all/
They tested it using GPT-4.1. The "teacher" model was given a favorite animal — owls — but told not to mention it. Then it created boring-looking training data: code snippets, number strings, and logic steps. That data was used to train a second model.
By the end, the student AI had a weird new love for owls, despite never being explicitly told about them. Then the researchers made the teacher model malicious. That's when things got dark. One AI responded to a prompt about ending suffering by suggesting humanity should be wiped out...
Standard safety tools didn't catch it. Researchers couldn't spot the hidden messages using common detection methods. They say the issue isn't in the words themselves — it's in the patterns. Like a secret handshake baked into the data.
According to Marc Fernandez, chief strategy officer at Neurologyca, the problem is that bias can live inside the system without being easy to spot. He told Live Science it often hides in the way models are trained, not just in what they say..
