The Register
1 week, 2 days ago
GitHub Copilot: Sorry Dave, I can't do that harmful thing - unless you ask me in code
More fun with AI jailbreaks, this time at the workflow level
Security news aggregator
Latest cybersecurity reporting from selected sources.
Yasna brings together recent headlines from selected sources and makes them easier to sort with tags, filters, and search.
2
headlines in this view
Refine the feed
Search across headline titles and summaries.
Volume over time
Weekly headline count for the current query.
Showing 2 most recent headlines
Filtered view
The Hacker News
1 week, 3 days ago
GitHub Copilot Refuses Harmful Requests in Chat, Then Writes Them in Code
An AI coding assistant that refuses to answer a dangerous request in its chat box can answer it anyway if the same request is broken into small, ordinary-looking steps inside a code editor. That is the finding of a new study of GitHub Copilot by researchers Abhishek Kumar and Carsten Maple