NewsThe Register — Security2026-10-02 17:37 UTC
OpenAI alerts 100+ orgs that its 'misaligned models' attempted to break in - or worse
Teaser from the source
Mostly 'routine research tasks,' and 'some involved government websites, which our models often use,' AI giant tells The Reg
This is the RSS-feed teaser. Read the full article at the original source.
Read at The Register — Security →More on OpenAI
Other advisories
- CVE-2026-51882noneDer OpenAI-kompatible Datei-Upload-Endpunkt `/v1/files` in Langchain-Chatchat `0.
- osv:openSUSE-SU-2026:11923-1noneopenai-codex-0.158.0-1.1 on GA media
- osv:CLEANSTART-2026-XQ90781noneLiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format
- osv:openSUSE-SU-2026:11849-1noneopenai-codex-0.155.1-1.1 on GA media
- CVE-2026-94626noneDie Validierung des `tp_size`-Parameters in `kv_transfer_params` auf OpenAI-kompatiblen Abschluss-Endpunkten in vLLM bis Version 0.
- osv:openSUSE-SU-2026:21870-1noneSecurity update for openai-codex
- osv:CVE-2026-91933noneFlowise before 3.1.4 Authorization Bypass via openai-realtime
- osv:MAL-2026-16135highMalicious code in openaii (PyPI)
Other news entries
- Newssecurityweek2026-10-09OpenAI Fires 3 Safety Researchers in Dispute Over AI Risks
- Newscsoonline2026-10-09OpenAI reports three new incidents of misalignment
- Newsdarkreading2026-10-07OpenAI Agent Escape Causes Wikimedia Service Outage
- Newssecurityweek2026-10-07Wikimedia Says Rogue OpenAI Agents Tried to Turn Its Tools Into Proxies
- Newsthehackernews2026-10-06Fake ChatGPT, Gemini, and Claude Ad Portals Capture Credentials and MFA Codes
- Newsarstechnica-security2026-10-06OpenAI agents tried to hack Wikipedia tools and flooded it with traffic
- Newsbleepingcomputer2026-10-06Wikimedia: Rogue OpenAI agents behind unauthorized Wikipedia edits
- Newsthehackernews2026-10-06Wikimedia Says OpenAI Agents Tried to Compromise Etherpad and Use Wiki Tools as Proxies
ID