AI safety is not a model property
normaltech.ai · 2,691 words · saved by 1 readers
Trying to make an AI model that can’t be misused is like trying to make a computer that can’t be used for bad things
The assumption that AI safety is a property of AI models is pervasive in the AI community. It is seen as so obvious that it is hardly ever explicitly stated. Because of this assumption: Companies have made big investments in red teaming their models before releasing them. Researchers are frantically trying to fix the brittleness of model alignment techniques. Some AI safety advocates seek to restrict open models given concerns that they might pose unique risks. Policymakers are trying to find the training compute threshold above which safety risks become serious enough to justify…
related reading
- AI safety is not a model propertysubstack.com
- AI safety is not a model propertyaisnakeoil.com
- The case for ensuring that powerful AIs are controlled — LessWronglesswrong.com
- Recommendations for Technical AI Safety Research Directionsalignment.anthropic.com
- A Safe Path to Open Weights - Thinking Machines Labthinkingmachines.ai
- AI models can be dangerous before public deploymentmetr.org
- AI Control: Improving Safety Despite Intentional Subversion — LessWronglesswrong.com
- Self-Fulfilling Misalignment Data Might Be Poisoning Our AI Modelsturntrout.com
- The case for ensuring that powerful AIs are controlledblog.redwoodresearch.org
- AI Control: Improving Safety Despite Intentional Subversion — AI Alignment Forumalignmentforum.org
- [2312.06942] AI Control: Improving Safety Despite Intentional Subversionarxiv.org
- Thoughts on the conservative assumptions in AI controlblog.redwoodresearch.org