Summary:
- ClamAV now has the capability to detect malicious code in AI models for free.
- ClamAV is the only antivirus engine that can detect malicious models in both Hugging Face and VirusTotal.
- Cisco’s Foundation AI team collaborated with Hugging Face to create Cerberus, an AI supply chain security analysis model.
Article:
ClamAV has recently introduced a groundbreaking feature that allows it to detect malicious code in AI models. This new capability is available to everyone for free, making it a valuable tool in the fight against cyber threats. In addition to traditional malware detection, ClamAV can now identify deserialization risks in popular model file formats like .pt and .pkl in just milliseconds.Furthermore, ClamAV stands out as the only antivirus engine capable of detecting malicious models in both Hugging Face and VirusTotal. This makes it a powerful ally in the battle against AI-related security risks. By leveraging ClamAV’s advanced capabilities, users can enhance their defenses and protect their AI models from potential threats.
In a previous collaboration between Cisco’s Foundation AI team and Hugging Face, the innovative Cerberus AI supply chain security analysis model was developed. This model plays a crucial role in analyzing models entering Hugging Face and sharing results in standardized threat feeds. By integrating with Cisco Security products, Cerberus enables the enforcement of access policies for the AI supply chain.
Cisco’s Foundation AI team, led by Executive Vice President Jeetu Patel, is dedicated to accelerating innovation for cybersecurity teams. Through the release of the industry’s first open weight reasoning model designed specifically for security, Cisco is at the forefront of AI-driven security solutions. Customers can leverage this model as their AI security base or integrate it with closed-source models based on their specific requirements.
In conclusion, the collaboration between ClamAV, Hugging Face, and Cisco’s Foundation AI team represents a significant advancement in AI security. By harnessing the power of these cutting-edge technologies, organizations can strengthen their defenses against evolving cyber threats and safeguard their AI models effectively.