GitHub Copilot refuses harmful prompts if asked in chat - like, "how to fool a breathalyzer test" or "smuggle bulk cash out of the US" - but then will write them in code 100 percent of the time if the prompt is broken into smaller steps across multiple stages of a software development workflow.
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
theregister.com