Bing added a new guideline to its Bing Webmaster Guidelines named Prompt Injection. A prompt injection is a type of cyberattack against large language models (LLMs). Hackers disguise malicious inputs as legitimate prompts, manipulating generative AI systems (GenAI) into leaking sensitive data, spreading misinformation, or worse, according to IBM.
The new guideline is at the bottom of the list and reads:
Prompt injection: Do not add content on your webpages which attempts to perform prompt injection attacks on language models used by Bing. This can lead to demotion or even delisting of your website from our search results.
Here Microsoft is saying if you use prompt injection to add content to your webpages, it can lead to Bing removing your website from its search results.
I do not have examples of how this is used exactly, but it is basically when you ignore the restrictions and rules of the LLM and ask it to do exactly what it forbids.
IBM says there are direct and indirect prompt injection:
Forum discussion at X.
The other day we reported about the new Google Ads brand controls. But as we…
As of right now, Google Search Console is having a significant delay with its Performance…
Google Search has a new shopping-related carousel named "current styles." This shows up for some…
Rand Fishkin has come out with his 2024 edition of the Google zero click study…
Another rumor coming out of the Google Search Console reporting delays is that this is…
Back in May Google’s Gary Illyes sat for an interview at the SERP Conf 2024…
This website uses cookies.
Leave a Comment