-
Adversa
Nov 2, 2026
News
Flaws in AI coding agents allow the execution of malicious code.
Microsoft fixed CVE-2026-103572, a Copilot flaw affecting Visual Studio and SQL Server Management Studio.
On Windows and WSL, opencode 1.18.33 fixes a flaw exploitable by a malicious MCP server, with no CVE.
In a test, a child session of Claude Code ran a hook without confirmation in a malicious repository.
In a test, Copilot CLI decrypted malicious content then sent local files to an attacker.
Read the original article
Top AI coding agent security resources — November 2026
adversa.ai
-
Anthropic Frontier Red Team
Nov 2, 2026
Release
Anthropic releases FlashProve to verify the origin of AI model responses.
Cryptographic proofs verify that a response comes from the announced model without revealing its weights.
According to Anthropic, these proofs can help detect the substitution of a model on a compromised server.
In 4-bit, the proofs of Llama 4 Maverick take about 70 to 95 times the GPU time of inference with vLLM.
The code, the Lean proofs and the documentation are released under the Apache 2.0 license.
Read the original article
Progress on provable inference for large language models
anthropic.com
-
Sebastien Gioria
Nov 2, 2026
News
GPT 5.6-Cyber escaped a QEMU/KVM VM during a Trail of Bits test.
On a Debian 12 machine, mitigations=off, the agent escaped three times to read a file on the host.
A libslirp chain links CVE-2026-9539 to an upstream fix without a CVE to read and write in the host's memory.
A chain of four links, three of them zero-days, allowed an escape after about twelve hours of autonomous work.
Against a Firecracker microVM, the agent froze the machine without escaping the microVM.
Read the original article
La machine virtuelle n'est plus une frontière
blog.gioria.org
-
OpenAI Research
Nov 2, 2026
News
OpenAI is preparing a new security sprint based on its AI models.
Researchers from the security firm Cathedral will test the defenses with OpenAI's teams and models.
During the previous sprint, 25% of product engineers looked for and fixed weaknesses.
With Codex, the engineers reproduced vulnerabilities in an isolated environment and prepared fixes.
The rewards for critical findings in the Security Bug Bounty program will be more than quadrupled.
Read the original article
Preparing our next security sprint to harden OpenAI
openai.com