OpenAI’s latest model will block the ‘ignore all previous instructions’ loophole
OpenAI’s latest model will block the ‘ignore all previous instructions’ loophole
OpenAI’s newest model, GPT-4o Mini, includes a new safety mechanism to prevent hackers from overriding chatbots.
You're viewing a single thread.
“We envision other types of more complex guardrails should exist in the future, especially for agentic use cases, e.g., the modern Internet is loaded with safeguards that range from web browsers that detect unsafe websites to ML-based spam classifiers for phishing attempts,” the research paper says.
The thing is folks know how the safeguards for the ‘modern internet’ actually work and are generally straightforward code. Where as LLMs are kinda the opposite, some mathematical model that spews out answers. Product managers thinking it can be corralled to behave in a specific, incorruptible way, I suspect will be disappointed.
Yeah, this is definitely part of the issue when commercializing LLMs. When someone has to provide an SLA or asking how frequently will this fail, it's not great when the best answer "who knows".