For awareness Low Monitor
AI finds 99 flaws in network software used in industrial devices
Industrial Cybercritical infrastructure
What happened. Forescout Research's Vedere Labs used off-the-shelf AI models to examine 10 widely used TCP/IP stacks in 90 hours, against about nine months by earlier methods. It found 99 vulnerabilities, 66 of them previously unknown. The automated analysis cost about $935.
Who is affected. The source names no specific products. TCP/IP stacks are embedded in connected products across OT, IoT, medical devices and critical infrastructure.
What to do. Brief your team that AI is lowering the cost of finding flaws, and ask device vendors how they will report and patch networking-stack flaws.
Why it matters to your site. Expect more networking-stack flaws to be reported in your devices, and use this to press vendors on patch timelines and to keep an asset inventory of device firmware.
When to act. For awareness: Research only; no specific flaw to patch yet, so watch for vendor advisories.
Editor's reading
ForOT security leadleadership
- Purdue level
- L2
- L1
Why: TCP/IP stacks are embedded in connected devices such as controllers and other OT products, so flaws in them reach the device level (L1) and the systems that use them (L2).
Inferred from the facts above, not stated by the source. Check it against your own site before acting.
Source: Industrial Cyber (opens industrialcyber.co)
More on Critical infrastructure