THREAT OPS › Threat News › OpenAI Explains URL-Based Data Exfiltration Mitigations in New Paper
OpenAI Explains URL-Based Data Exfiltration Mitigations in New Paper
<p>Last week I saw <a href="https://cdn.openai.com/pdf/dd8e7875-e606-42b4-80a1-f824e4e11cf4/prevent-url-data-exfil.pdf">this paper</a> from OpenAI called “Preventing URL-Based Data Exfiltration in Language-Model Agents”, which goes into detail on new mitigations they’ve added.</p> <p><a href="https://cdn.openai.com/pdf/dd8e7875-e606-42b4-80a1-f824e4e11cf4/prevent-url-data-exfil.pdf"><i
Indicators of compromise
- https://cdn.openai.com/pdf/dd8e7875-e606-42b4-80a1-f824e4e11cf4/prevent-url-data-exfil.pdfurl