THE Open Worldwide Application Security Project (OWASP) has launched a new top 10 security list for AI skills, highlighting the risks associated with 'Malicious Skills' and 'Supply Chain Compromise' at the forefront. A recent incident demonstrated vulnerabilities when attackers used a look-alike domain to introduce harmful AI features.
OWASP's list aims to categorize these risks for organizations, emphasizing the need for developers to be aware of potential threats in AI skills leveraging external code and instructions. Additionally, OWASP introduced a Universal Skill Format to standardize the skills' structure, helping in the identification of legitimate vs. malicious skills. Companies are urged to enhance visibility and governance over the AI skills utilized by their personnel.