# Together.ai: Prompt Leakage for Requests

> HelixNews · Web page: https://helixnews.one/story/12730898 · All pages for agents: https://helixnews.one/llms.txt

3 outlets covering · First listed: Hacker News, Nov 21, 2026, 9:07 PM · Read it: <https://www.together.ai/blog/security-event-prompt-leakage-for-requests-in-june-and-july-of-2026>

## Who's covering it

In the order they published:

- Nov 21, 2026, 5:32 PM: [Together AI Discloses Prompt Leak After Dragos Study](<https://thedefiant.io/frontier/ai/together-ai-discloses-prompt-leak-after-dragos-study>) — The Defiant (Specialist)
  Together says a coding agent exposed a credential for stored prompt samples. Five credentials later tested by the intruder were canaries from a Dragos experiment.
- Nov 21, 2026, 6:14 PM: [Together AI Reports Prompt Leakage Incident from June-July 2026](<https://blockchain.news/news/together-ai-prompt-leakage-incident-2026>) — Blockchain.News (Specialist)
  Together AI disclosed a security breach involving prompt samples from its serverless inference service in June and July 2026, caused by credential exposure. (
- Nov 21, 2026, 9:07 PM: [Together.ai: Prompt Leakage for Requests](<https://www.together.ai/blog/security-event-prompt-leakage-for-requests-in-june-and-july-of-2026>) — Hacker News (Aggregator, United States)

## How outlets word it

Most outlets use: Prompt, Together, Leakage.

- Only Hacker News says: Requests
- Only Blockchain.News says: Incident, June-July
- Only The Defiant says: Discloses, Leak, Dragos, Study
