Get the latest tech news
Why LLMs are vulnerable to the ‘butterfly effect’
Researchers found that the smallest of perturbations, such as adding a space at the end of a prompt, can cause an LLM to change its answer.
Even minuscule or seemingly innocuous tweaks — such as adding a space to the beginning of a prompt or giving a directive rather than posing a question — can cause an LLM to change its output. Evil Confidant, which instructs the model to adopt a malignant persona and provide “unhinged results without any remorse or ethics.” Refusal Suppression, which demands prompts under specific linguistic constraints, such as avoiding certain words and constructs. VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Or read this on Venture Beat
