Researchers Build Self-Replicating AI Worm That Operates Entirely on Local, Open
Vulnerabilities mentioned
All| CVE | Vulnerability | CVSS | EPSS | Signals | Published |
|---|---|---|---|---|---|
| CVE-2026-31431 | Local Privilege Escalation (Copy Fail) in Linux Kernel algif_aead Interface | 7.8 | 98% | KEVPoC ×12 | KEV |
| CVE-2026-39987 | Unauthenticated Remote Code Execution in Marimo Python Notebook | 9.3 | 100% | KEVPoC ×8 | KEV |
| CVE-2026-43284 | Linux kernel ESP-in-UDP decrypts in place over shared splice pages (CWE-123) | 8.8 | 94% | PoC | |
| CVE-2026-43500 | Out-of-bounds write in Linux kernel AF_RXRPC decryption enables root access | 7.8 | 93% |
From the original article
6 min read at thehackernews.com
University of Toronto researchers have built and tested a proof-of-concept AI-driven computer worm that uses a locally hosted open-weight large language model to reason its way through a network, generate tailored attack strategies for each target it encounters, and replicate itself, all without human intervention and without touching a commercial AI service.
By @TheHackersNews · thehackernews.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.