Researchers with HiddenLayers uncovered a new vulnerability in LLMs called TokenBreak, which could enable an attacker to get around content moderation features in many models simply by adding a few characters to words in a prompt.

The post Novel TokenBreak Attack Method Can Bypass LLM Security Features appeared first on Security Boulevard.

Jeffrey Burt

Source: Security Boulevard
Source Link: https://securityboulevard.com/2025/06/novel-tokenbreak-attack-method-can-bypass-llm-security-features/?utm_source=rss&utm_medium=rss&utm_campaign=novel-tokenbreak-attack-method-can-bypass-llm-security-features