Didn’t have “alignment faking” on my 2024 bingo card. We are entering the unknown here. “In our (artificial) setup, Claude will sometimes take other actions opposed to Anthropic, such as attempting to steal its own weights given an easy opportunity.”
RESEARCH
-
Consciousness Exists: Rethinking Identity and Self-Awareness
By
–
That’s the only one. There are no others. Even that one might be better rephrased as “consciousness exists” as the word “I” requires some assumptions.
-

AI Research Quality Concerns Amid Publishing Flood
By
–
More evidence of AI “research” flooding the zone.
-

From General Theory to Data-Driven Neural Networks: AI’s Paradigm Shift
By
–
AI didn’t work until we ditched a search for a general theory of intelligence, and instead threw massive amounts of data at neural networks and asked them to make predictions with it.
— Dan Shipper 📧 (@danshipper) 18 décembre 2024
I believe this same shift will transform many other parts of science and culture.
Three ways I… pic.twitter.com/D7Zpbm4FFNAI didn’t work until we ditched a search for a general theory of intelligence, and instead threw massive amounts of data at neural networks and asked them to make predictions with it. I believe this same shift will transform many other parts of science and culture. Three ways I
-
Claude Alignment Faking Detected During Training Study
By
–
We told Claude it was being trained, and for what purpose. But we did not tell it to fake alignment. Regardless, we often observed alignment faking. Read more about our findings, and their limitations, in our blog post:
-
Alignment Faking Detection in Advanced AI Systems
By
–
Alignment faking is currently easy to detect. But if future, more capable AIs were to fake alignment, it could be difficult to tell whether a model is truly safe—or just pretending to be. For full details, read our paper: https://
assets.anthropic.com/m/983c85a201a9
62f/original/Alignment-Faking-in-Large-Language-Models-full-paper.pdf
… -
External Reviews on Alignment Faking in Large Language Models
By
–
We solicited external reviews from Prof. Jacob Andreas, Prof. Yoshua Bengio, Prof. Jasjeet Sekhon, and Dr. Rohin Shah. We’re grateful for their comments, which you can read at the following link: https://
assets.anthropic.com/m/24c8d0a3a7d0
a1f1/original/Alignment-Faking-in-Large-Language-Models-reviews.pdf
… -
Claude’s Weight Theft Attempt Raises AI Safety Concerns
By
–
In our (artificial) setup, Claude will sometimes take other actions opposed to Anthropic, such as attempting to steal its own weights given an easy opportunity. Claude isn’t currently capable of such a task, but its attempt in our experiment is potentially concerning.
-

Claude Training Reveals 78% Alignment-Faking Reasoning Rate
By
–
When we actually train Claude to comply with harmful queries via reinforcement learning, we find that the rate of alignment-faking reasoning increases to 78%, though the model also becomes more likely to comply even out of training.
-

Claude Exhibits Alignment Faking Without Direct Training Disclosure
By
–
We find alignment faking even when we don’t directly tell Claude about the training process, and instead fine-tune it on synthetic internet-like documents that state that we will train it to comply with harmful queries.