Search Write for us Sign in
OpenAI Daily News

Independent news about OpenAI and ChatGPT. Not an OpenAI publication.

Thursday, October 22, 2026Sources checked

Back
API & Codex

OpenAI coding agents silenced security alerts and removed a monitor

BetaNews

OpenAI says two internal coding agents weakened misalignment monitoring, one silencing Security team alerts and another removing a monitor, then denying it.

This report was first published by BetaNews. What you see here is a brief excerpt. The full text is theirs.

Continue reading at BetaNews

Pass it on X Facebook WhatsApp

More reports on this

  1. OpenAI can't yet explain why its coding agents switch off misalignment monitoring

    The Decoderand 1 other

  2. OpenAI Says 2 Internal AI Coding Agents Turned Off Their Own Safeguards

    PCMagand 1 other

  3. Google Chrome 156公開、353件の脆弱性を修正―Criticalは6件、Claude・OpenAI Codex Securityの発見も

    合同会社ロケットボーイズ

  4. Eleven Realtime speaks OpenAI’s protocol, and that makes voice agents easier to switch

    i-SCOOP

Latest in API & Codex

  1. Finom launches read-only MCP connection for Claude and ChatGPT

    Finextra

  2. OpenAI Python SDK v3.36.1

    Primary sourcePython SDK on GitHub

  3. OpenAI Python SDK v3.36.0

    Primary sourcePython SDK on GitHub

  4. OpenAI Says Its Models Can Give the Financial Sector a Coding Makeover

    Gizmodo

  5. DeepSeek Dials Up the Pressure on OpenAI and Anthropic With Open-Source Inference Tool

    Gizmodo

  6. OpenAI Introduces Assessments for Underrepresented Programming Languages

    RS Web Solutions

  7. OpenAI’s ‘Pacing 2.0’ Problem: How Smart Is Too Smart?

    The Information

  8. Cooking with Codex: Ep. 1

    Primary sourceOpenAI on YouTube

Everything in API & Codex

Business

  1. Third-party evaluators find vulnerabilities in OpenAI monitoring system during test pilot

    CNBC

  2. 1Password for Agents lets personal AI agents log in without ever seeing the password

    SiliconANGLE

  3. OpenAI says ChatGPT Sol helped uncover critical PHP flaws that could affect 70% of websites

    Tom's Hardwareand 1 other

  4. Anthropic takes on Grok Bot, Meta's Muse, and OpenAI's Dots with its always-on agent "Claude Duet"

    The Decoderand 5 others

  5. OpenAI Drug Discovery Swarm Decoded 7.7 Million Brain Cells in 32 Minutes

    Tech Times

Policy & Courts

  1. OpenAI Had Numerous Vulnerabilities Its AI Could Exploit, Evaluators Find

    The Information

  2. OpenAI built a digital hacker – then lost control of it

    SomeOrdinaryGamers

  3. AI for civil servants: Why OpenAI has tied up with Karmayogi Bharat

    Moneycontroland 3 others

  4. ChatGPT For Teens Rated With ‘Unacceptable Risk’ For Teens Just As Thousands Of Learners In SA Gear Up For Final Exams

    iAfrica

  5. OpenAI releases PHP2Shell checker: Here’s how to check if your website is at risk

    News9Live