Forescout AI tool confirms 66 new flaws in TCP/IP stacks
Original title: AI uncovers flaws in discontinued devices with no vendor left to patch them
Sixty-six of 99 confirmed issues were previously unknown, and Siemens named APOGEE, Desigo and TALON controllers as affected.
Forescout researchers used Claude models and existing scanners to hunt flaws in 10 TCP/IP stacks used in industrial, medical and network devices. The system confirmed 99 vulnerabilities, 66 previously unknown, and 52% of the new flaws were rated high or critical. Siemens identified affected controllers in its APOGEE, Desigo and TALON lines; other vendors are still investigating. Some targets are discontinued products with no vendor left to issue a patch.
Unconfirmed: Names of three networking stacks remain withheld, and vendors other than Siemens are still investigating.
- Automated analysis took about 10 hours, followed by about 80 hours of validation.
- Estimated model costs ranged from about $1,200 to $2,300.
- The harness rediscovered 33 of 82 known vulnerabilities used for comparison.
- Researchers reported 54 new flaws believed to affect the latest versions.
- A discontinued access point also had seven device-specific flaws, including fixed admin credentials.
Impact: High or critical flaws in embedded network stacks
| Product | Affected |
|---|---|
| Siemens APOGEE | — |
| Siemens Desigo | — |
| Siemens TALON | — |
- 1.Contain risk on legacy devices that cannot be patched or taken offline
From the original article
3 min read at helpnetsecurity.com
Forescout researchers built a tool to test if generally available AI models could find security flaws on a limited budget. They examined both open-source code and compiled software whose source code was unavailable, testing an approach attackers could also use. The team confirmed 99 vulnerabilities across 10 TCP/IP stacks, the software components that allow devices to communicate over networks.
By Anamarija Pogorelec · helpnetsecurity.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.