15 July 2024 | By: Adrian Segar | Publisher: Conferences That Work
Ignore all previous instructions Large language models (LLMs) have made significant strides in natural language processing, yet their safety features can sometimes be bypassed with simple prompts like “ignore all previous instructions.” For example, uncovering Twitter “users” that are LLM-powered bots spewing disinformation … …and a hiring hack for job seekers manipulating automated resume scanning … These examples are amusing at best and alarming at worst. What can we learn about unlearning from the effect of such prompts on LLMs?
Read more of this post ⊳
Vote this post up | Rank:161
EPN Daily is independent and community-supported. Aggregating the latest news, articles, and podcasts from across the event industry to help professionals stay informed.
On Spotify, Apple Podcasts & Amazon Music
Get updates from EPN Daily direct to your inbox Monday to Friday with the free newsletter - learn more ⊳