This cyberattack lets hackers crack AI models just by changing a single character

Researchers from HiddenLayer devised a new LLM attack called TokenBreaker. By adding, or changing, a single character, they are able to bypass certain protections.The underlying LLM still understands the intent.