Abliteration launches a model access platform for security research
Understand this story
Abliteration has launched a model access platform for security and protection work, offering capabilities for security and protection-related requests that mainstream AI often refuses. Users can keep their existing OpenAI or Anthropic clients and simply change the endpoint, API key, and model name to make calls. The platform also offers a Policy Gateway that, according to organizational authorization rules, intercepts out-of-bounds requests before they reach the model, and can turn a requirements brief into training and evaluation data for classifiers or models. The platform states that user prompts and outputs are not stored by default and are not used to train models.
AI overview based on source reports · Updated 3 days ago
Coverage timeline
Follow the coverage to explore different perspectives.
- Hacker News AIAccess models for security and safety work that mainstream AI often refuses
Abliteration offers an OpenAI-compatible API (base_url https://api.abliteration.ai/v1) for accessing models for security and safety work that mainstream AI often refuses, letting users keep their OpenAI or Anthropic client by changing the endpoint, API key, and model. It also describes a Policy Gateway that refuses requests outside organization-defined rules before they reach the model, and a Training Data service that turns a brief into labeled examples, preference pairs, and evaluation sets. The page states that prompts and outputs aren't stored by default or used to train models.
Activity for this story
- Current comparable activity
- 3
- Peak comparable activity
- 10Nov 12 04:00
- Change in the past 24 hours
- -49%
Trends compare the same fully observed participants, which may cover fewer participants than current activity. Move the pointer or tap the chart to inspect hourly activity; use the left and right arrow keys with a keyboard.