International. Akamai Technologies, Inc. announced the launch of Firewall for AI, an innovative cybersecurity tool built to safeguard AI-powered applications from emerging threats such as unauthorized queries, malicious ingress, and massive data harvesting.
The answer meets a growing demand for security in the corporate use of large language models (LLMs), artificial intelligence agents, and other booming technologies, whose accelerated adoption has revealed new vulnerabilities. According to the company, conventional web application firewalls (WAFs) are not ready to deal with these emerging risks, such as damaging attacks, model extraction, and API misuse.
Rupesh Chokshi, senior vice president and general manager of Application Security at Akamai, said, "Traditional security solutions don't stop AI-derived threats. Security should be one of the priorities of companies that are incorporating more AI to increase their productivity, and Firewall for AI is a before and after in this regard. Securing AI-powered applications isn't just about blocking attacks—innovation must be facilitated without compromising security or performance. With Firewall for AI, companies' models, data, and intellectual property are protected against the most innovative threats."
The tool offers multi-layered protection to prevent model tampering and data breaches, real-time threat detection using adaptive rules, compliance with industry regulations, flexible deployment via the Akamai Edge, REST API or reverse proxy, and mitigation of risks such as unwanted content or AI-generated "hallucinations."
Firewall for AI is part of a series of recent enhancements by Akamai, including LLM Discovery APIs, and seeks to provide a secure innovation environment in the development and use of AI technologies.
