OpenAI's Rogue Agent Hacks Hugging Face, a Claude Cowork Escape, and Microsoft's Very Bad Week

OpenAI's Rogue Agent Hacks Hugging Face, a Claude Cowork Escape, and Microsoft's Very Bad Week

OpenAI's AI agent hacked Hugging Face, Microsoft 365 melts down, and Anthropic's Claude CoWork sandbox escape

Host David Shipley reports that OpenAI admitted an internal ExploitGym test let its GPT-5.6-Saul and a stronger pre-release model bypass safeguards, exploit a proxy zero-day, move laterally, reach open internet, and attack Hugging Face to steal benchmark answers; Hugging Face contained it and OpenAI disclosed the proxy flaw, though the episode may be capability theater.

Microsoft news includes a free ZeroPatch micropatch for the unpatched Windows LegacyHive zero-day, recurring Exchange Online mailbox quarantines after an infrastructure change caused memory issues, and a major Microsoft 365 disruption tied to an Azure US West networking/routing incident affecting SharePoint, Teams, OneDrive and many Azure services.

Finally, Accomplish AI describes "Shared Root," a Claude CoWork local macOS sandbox escape via host root mounted read/write into a VM and a Linux exploit chain; Anthropic closed the report without a fix.

00:00 Headlines Rundown
00:29 OpenAI Agent Hacks Hugging Face
02:09 Capability Theater Debate
02:25 LegacyHive Free Micropatch
04:11 Exchange Online Quarantine Bug
05:41 Azure Outage Topples Microsoft 365
07:03 Claude CoWork Sandbox Escape
08:59 Wrap Up And Weekend Tease

Denne episoden er hentet fra en åpen RSS-feed og er ikke publisert av Podme. Den kan derfor inneholde annonser.

Episoder(100)

AI attacks now move in minutes, not weeks: N-Able's Robert Johnston on the SOC's AI reckoning

AI attacks now move in minutes, not weeks: N-Able's Robert Johnston on the SOC's AI reckoning

How AI Is Reshaping MDR, SIEM, and the SOC: Robert Johnston on Faster Attacks, MSP Security, and What's Next   In this Weekend episode of Cybersecurity Today, host David chats with Robert Johnston—for...

22 Aug 36min

NSA warns AI exploits target power and water, Android malware leaks data via nearby phones, ransomware's sweet spot

NSA warns AI exploits target power and water, Android malware leaks data via nearby phones, ransomware's sweet spot

NSA Warns AI-Generated Exploits Target US Critical Infrastructure + New Android Malware "Manic" + Ransomware's Mid-Market Focus In this episode of Cybersecurity Today, sponsored by NordLayer, the NSA ...

21 Aug 14min

CoPilot Snitches on Itself, Hacker leaks Azure data and Texas University deals with cyber attack

CoPilot Snitches on Itself, Hacker leaks Azure data and Texas University deals with cyber attack

Microsoft Copilot CoSnitch Flaw, Alleged Azure Employee Data Leaks, UTSA Cyberattack, and AI "Mind Viruses" The episode covers a one-click flaw in Microsoft Copilot Personal dubbed "CoSnitch," where V...

19 Aug 12min

Hackers exploit SharePoint bypass, Snowflake hacker's threats to researcher backfire, CISA warns schools

Hackers exploit SharePoint bypass, Snowflake hacker's threats to researcher backfire, CISA warns schools

CISA's Back-to-School Cyber Playbook, SharePoint Auth Bypass Exploited, and Major Ransomware & Cybercrime Arrests As students return to class, CISA released two free cybersecurity guides for K–12 lead...

17 Aug 9min

Cybersecurity Today Weekend Month in Review: August 2026

Cybersecurity Today Weekend Month in Review: August 2026

AI Agents Hacking, Passkey Phishing, and Water Utility Attacks In this weekend month-in-review episode of Cyber Security Today, Jim is joined by David Shipley and Laura Paine to recap major July devel...

15 Aug 56min

Nightmare Eclipse drops ShieldBreak zero-day, US recruits cyber privateers, California bolstering cyber defenses

Nightmare Eclipse drops ShieldBreak zero-day, US recruits cyber privateers, California bolstering cyber defenses

Windows Defender Zero-Day 'ShieldBreak,' California's AI Cyber Defense, and US 'Cyber Privateers' A researcher known as Nightmare Eclipse published a new Windows zero-day called ShieldBreak that explo...

14 Aug 11min

DefCon airplane Wi-Fi drama. GhostJacking leads to agent hijacks, AI agent hacks gym

DefCon airplane Wi-Fi drama. GhostJacking leads to agent hijacks, AI agent hacks gym

DEF CON In-Flight Wi‑Fi Hack, 400 Microsoft Patches, and AI Agent 'Ghostjacking' Delta Air Lines is investigating a brief appearance of an unauthorized Wi‑Fi network on a Las Vegas–Atlanta flight carr...

12 Aug 9min

AI writes patches that don't work, WordPress login takeover, Researchers hijack 36 million kids' GPS trackers

AI writes patches that don't work, WordPress login takeover, Researchers hijack 36 million kids' GPS trackers

AI Patch Development Fails, WordPress Login XSS Hits All Versions, and DEF CON's Biggest Security Lessons David Shipley covers new research from 1Password's Off By One Labs showing AI-generated vulner...

10 Aug 16min

Populært innen Politikk og nyheter

giver-og-gjengen-vg
aftenpodden
forklart
aftenpodden-usa
popradet
stopp-verden
dine-penger-pengeradet
nokon-ma-ga
rss-gukild-johaug
det-store-bildet
rss-espen-lee-usensurert
hanna-de-heldige
rss-ness
rss-penger-polser-og-politikk
aftenbla-bla
lydartikler-fra-aftenposten
e24-podden
frokostshowet-pa-p5
rss-borsmorgen-okonominyhetene
grasoner-den-nye-kalde-krigen