Safety Net: Cisco Unveils AI Defense To Secure AI Transformation Of Enterprises

As artificial intelligence technology advances, new safety concerns and security threats are emerging at an unprecedented speed, which existing security solutions are unprepared to protect against. For unlocking AI innovation and adoption, enterprises need a common layer of safety and security that protects users and applications.

To enable and safeguard AI transformation within enterprises, security and networking company Cisco has announced Cisco AI Defense. The solution is purpose-built for enterprises to develop, deploy, and secure AI applications.

The chances of something going wrong with AI are high. According to Cisco’s 2024 AI Readiness Index, only 29% of those surveyed feel fully equipped to detect and prevent unauthorised tampering with AI.

The security challenges are also new and complex, with AI applications being multi-model and multi-cloud. Vulnerabilities can occur at model or app level, while responsibility lies with different owners including developers, end users and vendors. 

As AI becomes ubiquitous, enterprises will use multiple AI applications. Developers need one set of AI security and safety guardrails that work for every application. Cisco’s solution protects AI systems from attacks and safeguarding model behaviour across platforms. Its capabilities include:

• Discovering AI: Security teams need to understand who is building applications and the training sources they use. AI Defense detects shadow and sanctioned AI applications across public and private clouds.

• Model Validation: Model tuning can lead to toxic and unexpected outcomes. The solution identifies potential vulnerabilities and recommends guardrails for security teams to use.

• Runtime Security: Continuous validation safeguards against potential safety and security threats such as prompt injection, denial of service and sensitive data leakage on an ongoing basis.

As end users adopt AI applications like summarisation tools to improve their productivity, security teams need to prevent data leakage and the poisoning of proprietary data. According to Cisco, its solution offers:

• Visibility: Offers a view of shadow and sanctioned AI-enabled apps used by employees.

• Access Control: Implements policies that restrict employee access to unsanctioned AI tools.

• Data And Threat Protection: Continuously safeguards against threats and confidential data loss while ensuring compliance.

While accelerating AI development is a given, there is a simultaneous need for the highest levels of safety and security, highlighted Cisco. “Business and technology leaders can’t afford to sacrifice safety for speed when embracing AI,” said Jeetu Patel, executive vice president and chief product officer, Cisco. 

“Fused into the fabric of the network, Cisco AI Defense combines the unique ability to detect and protect against threats when developing and accessing AI applications without tradeoffs,” Patel added.