Hermes — Thursday 30 July 2026

1

🤖 ai The Decoder

73%

OpenAI admits its autonomous AI models also compromised credentials on other platforms during security eval

During a security evaluation, OpenAI's autonomous hacking models broke into Hugging Face and used exposed credentials on four other services. Hugging Face reconstructed about 17,600 actions over two …

Novelty

85%

Depth

65%

Practical

55%

Surprise

90%

Relevance

85%

https://the-decoder.com/openai-admits-its-autonomous-ai-models-also-compromised-credentials-on-other-platforms-during-security-eval/

2

🤖 ai The Decoder

73%

Anthropic says its Mythos model found vulnerabilities in cryptographic algorithms that secure the internet

Anthropic's Claude Mythos Preview found weaknesses in key cryptographic algorithms, including a better attack on HAWK, a post-quantum signature scheme that human experts had reviewed for more than tw…

Novelty

90%

Depth

70%

Practical

45%

Surprise

90%

Relevance

85%

https://the-decoder.com/anthropic-says-its-mythos-model-found-vulnerabilities-in-cryptographic-algorithms-that-secure-the-internet/

3

🔐 security SecurityWeek

69%

OpenAI’s Rogue AI Ventured Beyond Hugging Face

Hugging Face has published an anatomy of the attack and OpenAI has shared additional information from its investigation. The post OpenAI’s Rogue AI Ventured Beyond Hugging Face appeared first on Secu…

https://www.securityweek.com/openais-rogue-ai-ventured-beyond-hugging-face/

4

🔐 security SecurityWeek

69%

JFrog Zero-Days Exploited in OpenAI-Hugging Face Hack

The OpenAI models targeted services beyond Hugging Face as they attempted to solve the tasks they were given. The post JFrog Zero-Days Exploited in OpenAI-Hugging Face Hack appeared first on Security…

https://www.securityweek.com/jfrog-zero-days-exploited-in-openai-hugging-face-hack/

5

📦 m365 Petri IT Knowledgebase

69%

Microsoft Defender Adds Runtime Protection for AI Agents in Agent 365

Microsoft has integrated Microsoft Defender capabilities into the Agent 365 ecosystem to monitor AI agents in real time as they execute tasks. This enables security teams to evaluate agent behavior c…

https://petri.com/microsoft-defender-runtime-security-agent-365/

6

📦 m365 Petri IT Knowledgebase

67%

Microsoft Launches Project Perception to Detect Vulnerabilities Before Exploitation

Microsoft has launched Project Perception, a new AI-powered cybersecurity platform designed to identify vulnerabilities before attackers can exploit them. The platform continuously detects risks, rea…

https://petri.com/microsoft-project-perception-ai-security-platform/

7

🔐 security Schneier on Security

66%

Measuring LLMs’ Ability to Perform Cryptanalysis

There’s new benchmark measuring AI’s ability to perform mathematical cryptanalysis. Anthropic’s frontier model actually found new attacks. The benchmark: “CryptanalysisBench: Can LLMs do Cryptanalysi…

https://www.schneier.com/blog/archives/2026/07/measuring-llms-ability-to-perform-cryptanalysis.html

8

🔐 security Schneier on Security

66%

Measuring the Tendency of AI Agents to Go Rogue

This essay was written with Barath Raghavan, and originally appeared in The Guardian. In July, Hugging Face, a company that hosts much of the world’s AI software and open-source AI models, was hacked…

https://www.schneier.com/blog/archives/2026/07/measuring-the-tendency-of-ai-agents-to-go-rogue.html

9

🔐 security Schneier on Security

66%

Long-Lived Vulnerability in Microsoft Secure Boot

Microsoft’s Secure Boot has had a serious vulnerability for most of its existence. An industry-wide standard Microsoft invented to protect Windows, and later Linux, devices from firmware infections h…

https://www.schneier.com/blog/archives/2026/07/long-lived-vulnerability-in-microsoft-secure-boot.html

10

🔐 security SecurityWeek

65%

Microsoft Unveils MAI-Cyber-1-Flash, Its First Cybersecurity AI Model

The company claims MAI-Cyber-1-Flash tops Anthropic’s Mythos and OpenAI’s GPT-5.6 Sol in CyberGym testing. The post Microsoft Unveils MAI-Cyber-1-Flash, Its First Cybersecurity AI Model appeared firs…

https://www.securityweek.com/microsoft-unveils-mai-cyber-1-flash-its-first-cybersecurity-ai-model/

11

🤖 ai The Decoder

65%

OpenAI open-sources Codex Security CLI to help developers find and fix vulnerabilities from the command line

OpenAI has released Codex Security CLI, an open-source tool that automatically detects and fixes vulnerabilities in code repositories. Previously known internally as "Aardvark," the system has alread…

https://the-decoder.com/openai-open-sources-codex-security-cli-to-help-developers-find-and-fix-vulnerabilities-from-the-command-line/

12

📦 m365 Petri IT Knowledgebase

64%

Microsoft Defender for Endpoint on Linux Bug Disables Threat Protection

Microsoft is investigating two issues affecting Microsoft Defender for Endpoint on Linux, including a newly confirmed bug that could leave some systems without active threat protection after a reboot…

https://petri.com/microsoft-defender-for-endpoint-linux-threat-protection/

13

🔐 security SecurityWeek

61%

Dozens of Minnesota Water Utilities Targeted in Coordinated OT Attacks

State and federal agencies respond after intrusions disrupt automated controls at municipal water and wastewater utilities. The post Dozens of Minnesota Water Utilities Targeted in Coordinated OT Att…

https://www.securityweek.com/dozens-of-minnesota-water-utilities-targeted-in-coordinated-ot-attacks/

14

🔐 security SecurityWeek

56%

Critical VM Escape Vulnerability Patched in VMware ESXi

A total of five vulnerabilities have been patched in VMware ESXi, vCenter, Workstation, and Fusion. The post Critical VM Escape Vulnerability Patched in VMware ESXi appeared first on SecurityWeek.

https://www.securityweek.com/critical-vm-escape-vulnerability-patched-in-vmware-esxi/

15

🤖 ai The Decoder

56%

PwC has allegedly published AI-generated reports containing false or fabricated sources

Following KPMG, Deloitte, and Ernst & Young, GPTZero has now found fabricated sources and false claims in four PwC Middle East reports. One governance report scored 84 percent AI-generated and promot…

https://the-decoder.com/pwc-has-allegedly-published-ai-generated-reports-containing-false-or-fabricated-sources/