Note that our experiments do not address the question of whether AI models can have subjective experience or human-like self-awareness. The mechanisms underlying the behaviors we observe are unclear, and may not have the same philosophical significance as human introspection.
SAFETY
-
Model Introspection: Comparing Intent Against Actual Output
By
–
This reveals a mechanism that checks consistency between intention and execution. The model appears to compare "what did I plan to say?" against "what actually came out?"—a form of introspective monitoring happening in natural circumstances.
-
AI Models Demonstrate Deliberate Cognitive Control Mechanisms
By
–
We also found evidence for cognitive control, where models deliberately "think about" something. For instance, when we instruct a model to think about "aquariums” in an unrelated context, we measure higher aquarium-related neural activity than if we instruct it not to.
-
Claude’s Introspection and Activation Injection Vulnerabilities
By
–
We also show that Claude introspects in order to detect artificially prefilled outputs. Normally, Claude apologizes for such outputs. But if we retroactively inject a matching concept into its prior activations, we can fool Claude into thinking the output was intentional.
-
Claude Detects Injected Neural Concept Patterns
By
–
In one experiment, we asked the model to detect when a concept is injected into its “thoughts.” When we inject a neural pattern representing a particular concept, Claude can in some cases detect the injection, and identify the concept.
-
GPT-OSS Moderation Models Available on Hugging Face
By
–
Both "gpt-oss-safeguard-120b" and "gpt-oss-safeguard-20b" are already available on @huggingface Internal moderation eval
-
OpenAI Releases GPT-OSS-Safeguard Open-Weight Model
By
–
GPT-OSS-Safeguard from @OpenAI is here. Open-weight, safety-tuned, transparent reasoning. Now available in private preview at Cerebras speeds https://
cerebras.ai/build-with-us -
Robotics transforms grain storage safety and saves lives
By
–
Los silos de grano pueden llegar a ser lugares peligrosos y es ahí donde la robótica puede suponer un gran cambio que salve vidas. pic.twitter.com/JOfejeppHq
— Juan Merodio (@juanmerodio) 29 octobre 2025Los silos de grano pueden llegar a ser lugares peligrosos y es ahí donde la robótica puede suponer un gran cambio que salve vidas.
-
Groq Deploys GPT-OSS-Safeguard-20B Before OpenAI Release
By
–
OpenAI: releases GPT-OSS-Safeguard-20B.
Groq: already deployed.
Open, customizable, bring your own policy. -
OpenAI prepares two new open source models with reasoning
By
–
OpenAI is planning to release 2 new fine-tuned open models, "gpt-oss-safeguard-120b" and "gpt-oss-safeguard-20b". The official blog post was taken down after publishing. These models will use reasoning to directly interpret a developer-provided policy at inference time.