Splunk Urges Australian Organisations to Secure LLMs

"The big players are putting lots of guardrails around their tools, but there's still lots of ways to get them to do things that those guardrails are trying to prevent," Davis said.

"Prompt injection is a vulnerability that occurs when an attacker manipulates an LLM through crafted inputs," according to the OWASP definition.

Read at TechRepublic