Static Scans, Red Teams and Frameworks Aim to Find Bad AI Models

With hundreds of AI models found to harbor malicious code, cybersecurity firms are releasing technology to help companies manage their AI development and deployment efforts.