Cisco Builds Cheaper AI Tools to Hunt Security Flaws in Software Code
The company's new Antares models scan source code for known vulnerabilities at a fraction of what larger AI systems cost. Here is what that means for businesses that write or buy software.
Key points
- Cisco released a family of AI models called Antares in 2025, designed specifically to find security weaknesses in software source code.
- Antares is described as "open-weight", meaning other developers can inspect and build on the underlying model.
- The models are positioned as significantly cheaper to run than comparable large AI systems.
- Their primary job is to identify known vulnerabilities, meaning software flaws that security researchers have already catalogued and named.
Cisco, the networking and security giant, has released a set of artificial intelligence models aimed at one specific problem: finding security flaws buried inside software code before criminals can exploit them. The product family is called Antares, and SecurityWeek first reported its launch.
To understand why this matters, a short plain-English detour helps. Every piece of software, from a banking app to a hospital records system, is built from source code, which is the human-readable instructions a developer writes. Hidden inside that code can be vulnerabilities, meaning weaknesses an attacker could use to break in, steal data, or take control of a system. Spotting those weaknesses before shipping the software is one of the hardest jobs in the industry.
Antares is built to do that job automatically.
The models are described as "open-weight". That means the underlying structure of the AI is published openly, so other companies and developers can inspect it, adapt it, and plug it into their own tools. That is different from a closed, proprietary system where the inner workings stay secret.
The cost angle is the other headline. Running large AI models is expensive, requiring significant computing power billed by the hour. Cisco says Antares delivers comparable results at a fraction of that cost, which could make automated code-security checks realistic for smaller software teams that currently skip them because the price is prohibitive.
Antares focuses on known vulnerabilities. These are flaws that have already been discovered, reported, and catalogued, often under a CVE identifier (short for Common Vulnerabilities and Exposures, a standard numbering system the security industry uses to track individual flaws). Finding known flaws in new code is tedious work. Developers sometimes accidentally reuse patterns that have already been proven dangerous. An AI model that flags those patterns quickly and cheaply could stop problems before a product ships.
Does this affect ordinary software users?
Not directly today, but the logic is straightforward. More software teams running better automated security checks means fewer products reach the public with easily preventable holes in them. That benefits anyone who uses software, which is everyone.
If you work at a company that develops software in-house, it is worth asking your technical team whether automated code-scanning tools are part of the build process. They do not have to be expensive any more.
