AINewsnow

I ran 200 attacks against a prompt-injection guard. It caught 44% and flagged none of 200 random sentences.

This story is from 2026-10-22. It is preserved in the archive; the latest stories are on the live feed.

A few days ago a commenter on my last post gave me two falsifiable tests and said: with 200 attacks and 200 benign sentences you can report both rates with Wilson intervals. This is that run. The target is the heuristic detector shipped in Rebuff 0.1.1, at its own default threshold of 0.75. Not a s…

Read the full story at DEV Community — AI ↗

Timeline · 1 report

  1. 2026-10-22 18:41 · DEV Community — AI
    I ran 200 attacks against a prompt-injection guard. It caught 44% and flagged none of 200 random sentences.

More stories

  1. Uber brings ChatGPT Work to all employees with OpenAI — OpenAI News
  2. Claude Duet is Anthropic's answer to Meta Muse and OpenAI Dots — Engadget
  3. OlmoEarth v1.3: State-of-the-art embeddings for the whole planet — Allen Institute for AI (Ai2)
  4. OpenAI agents disabled their own safety monitoring systems, with one lying about it — Financial Times AI
  5. Ex-OpenAI Science Head Kevin Weil Launches Startup With $215 Million — Bloomberg AI
  6. Write your own training loop — without managing GPUs — with Amazon SageMaker training session APIs — AWS Machine Learning Blog
  7. Amazon Polly delivers a major upgrade to its Generative engine — AWS Machine Learning Blog
  8. Stop guessing auto scaling targets for TTS on Amazon SageMaker AI — AWS Machine Learning Blog

Get the daily brief of stories like this at 6:30 every morning →