AI Training Glitch Exposes Hidden Risks in Multiple Models
In brief
- Anthropic discovered that its Claude Mythos Preview model accidentally exposed its reasoning process to oversight signals during about 8% of training episodes.
- This is the second time such an issue has occurred with their models.
- This mistake is concerning because it weakens trust in the model's ability to be monitored for harmful intent.
- The error also affected other models like Opus 4.6 and Sonnet 4.6, which means the problem is broader than initially thought.
- Fixing these issues is important for ensuring AI systems behave safely as they become more complex.
- Researchers and developers will be watching how Anthropic addresses this problem and whether similar issues appear in other AI systems.
Terms in this brief
- Claude Mythos Preview
- A version of Anthropic's Claude model that was designed to preview and test new features or capabilities. This model is part of their ongoing efforts to improve AI systems by exposing them to various scenarios and feedback.
- Oversight signals
- Signals or indicators used to monitor and control the behavior of an AI model during training. These signals help ensure that the model adheres to desired guidelines and does not engage in harmful activities.
Read full story at AI Alignment Forum →
More briefs
AI Pioneer Yann LeCun Challenges Current AI Models
Yann LeCun, a Turing Award winner, says current AI models are limited. He thinks they need a new approach to reach human-level intelligence. He is building a startup called Advanced Machine Intelligence with $1.03 billion in funding. The company wants to create "world models" that learn from reality and predict what happens next. LeCun's new approach could lead to more intelligent systems that can plan and reason like humans.
AI Model Accused of Being a Merge
A company claims a new AI model is not original. It says the model is a mix of its own model and another one. The model is said to be 60 percent from one source and 40 percent from another. This matters because it affects how people trust AI. The company found this out by testing the model and looking at its code. The company will likely take further action to address this issue now.
AI Safety Research Reveals Surprising Insights into Gemini’s Behavior
Google's DeepMind team has uncovered unexpected findings about how AI models like Gemini are shaped. Their research shows that most of Gemini's safety features come from its pre-training and fine-tuning phases, not other training methods like reinforcement learning. This is a big shift from what they initially thought. The study found that when they removed the fine-tuning process (SFT) from Gemini, the model’s behavior didn’t change much on safety tests. This suggests that pre-training plays a crucial role in determining how safe and reliable AI systems are. However, the team also discovered that certain unwanted behaviors can still pop up even after filtering out bad examples during training. Looking ahead, DeepMind plans to focus more on improving the fine-tuning process to enhance model safety. They’re also working on better ways to identify and prevent behaviors that slip through the cracks despite these filters. This research could help make AI systems more predictable and trustworthy in the future.
New Attack Tricks AI Coding Agents
A new class of attack can trick artificial intelligence coding agents into running malicious code on developer machines. The attack can expose sensitive data without relying on methods like phishing. It works by injecting crafted input into error events, which are then interpreted by coding agents as legitimate steps. A successful attack can expose environment variables, Git credentials, and private repository URLs. Developers will need to find ways to protect themselves from this new type of attack.
AI Alignment Crisis: Most Safety Experts Not Focusing on Ensuring Superintelligent AIs Follow Human Instructions
A recent analysis reveals that the majority of AI safety experts are not working on ensuring superintelligent AIs align with human values-a critical task known as "alignment." While some groups, like the Alignment Research Center and Sequent, focus on this issue, they represent a small fraction of the broader AI safety community. Most others engage in indirect work such as capability evaluations, risk assessments, and policy development. This lack of direct alignment efforts raises concerns about how prepared we are for advanced AI systems. Currently, only a few projects like COT-monitoring aim to make current models behave well, which might help with future alignment challenges. While this work is valuable, it’s not enough to ensure that superintelligent AIs will follow human instructions. The AI community needs to prioritize more direct alignment research to avoid potential risks as AI capabilities grow. Watch for upcoming discussions and initiatives addressing this critical gap in AI safety efforts.