Hermes
Friday 20 March 2026  ·  74 articles scored  ·  0 top scorers  ·  last 24h
1
🔐 security Microsoft Security Blog
68%

New tools and guidance: Announcing Zero Trust for AI

Microsoft introduces Zero Trust for AI, adding a new AI pillar to its workshop, enhanced reference architecture, updated guidance, and a new assessment tool. The post New tools and guidance: Announci…

https://www.microsoft.com/en-us/security/blog/2026/03/19/new-tools-and-guidance-announcing-zero-trust-for-ai/
2
🤖 ai AI Alignment Forum
67%

Metagaming matters for training, evaluation, and oversight

Following up on our previous work on verbalized eval awareness: we are sharing a post investigating the emergence of metagaming reasoning in a frontier training run. Metagaming is a more general, and…

https://www.alignmentforum.org/posts/4hXWSw8tzoK9PM7v6/metagaming-matters-for-training-evaluation-and-oversight
3
⚡ tech Stratechery
66%

Jensen Huang and Andy Grove, Groq LPUs and Vera CPUs, Hotel California

GTC 2026 marked an important inflection point for Nvidia, as the company is selling multiple architectures, instead of focusing on just one GPU. The motivation is serve all needs and keep all custome…

https://stratechery.com/2026/jensen-huang-and-andy-grove-groq-lpus-and-vera-cpus-hotel-california/
4
🔐 security Microsoft Security Blog
63%

Observability for AI Systems: Strengthening visibility for proactive risk detection

As AI systems grow more autonomous, observability becomes essential. Learn how visibility into AI behavior helps detect risk and strengthen secure development. The post Observability for AI Systems: …

https://www.microsoft.com/en-us/security/blog/2026/03/18/observability-ai-systems-strengthening-visibility-proactive-risk-detection/
5
⚡ tech The Verge
62%

A rogue AI led to a serious security incident at Meta

For almost two hours last week, Meta employees had unauthorized access to company and user data thanks to an AI agent that gave an employee inaccurate technical advice, as previously reported by The …

https://www.theverge.com/ai-artificial-intelligence/897528/meta-rogue-ai-agent-security-incident
6
🔐 security SecurityWeek
62%

‘DarkSword’ iOS Exploit Kit Used by State-Sponsored Hackers, Spyware Vendors

Targeting six iOS vulnerabilities and leading to full device compromise, the exploit chain is meant for surveillance. The post ‘DarkSword’ iOS Exploit Kit Used by State-Sponsored Hackers, Spyware Ven…

https://www.securityweek.com/darksword-ios-exploit-kit-used-by-state-sponsored-hackers-spyware-vendors/
7
📦 m365 Petri IT Knowledgebase
61%

Why Identity Governance Is Now Core to Zero Trust — Not Just Compliance

Identity governance has rapidly evolved from a compliance checkbox into a cornerstone of modern Zero Trust security, especially as organizations adopt cloud-first architectures and AI-driven workflow…

https://petri.com/identity-governance-is-core-to-zero-trust/
8
🔐 security SecurityWeek
60%

Shadow AI Risk: How SaaS Apps Are Quietly Enabling Massive Breaches

From Chaos to Control examines the chaos that often comes from shadow AI hidden in SaaS apps and urges better visibility and control over agentic AI. The post Shadow AI Risk: How SaaS Apps Are Quietl…

https://www.securityweek.com/the-shadow-ai-problem-how-saas-apps-are-quietly-enabling-massive-breaches/
9
🤖 ai The Decoder
59%

A rogue AI agent caused a serious security incident at Meta

An out-of-control AI agent has triggered a serious security incident at Meta, reports The Information. The article A rogue AI agent caused a serious security incident at Meta appeared first on The De…

https://the-decoder.com/a-rogue-ai-agent-caused-a-serious-security-incident-at-meta/
10
🔐 security SecurityWeek
59%

The Collapse of Predictive Security in the Age of Machine-Speed Attacks

With exploitation of vulnerabilities taking just days, preemptive security must be the new model for defenders. The post The Collapse of Predictive Security in the Age of Machine-Speed Attacks appear…

https://www.securityweek.com/the-collapse-of-predictive-security-in-the-age-of-machine-speed-attacks/
11
📦 m365 Petri IT Knowledgebase
58%

Microsoft Expands Purview Data Protection and Governance Capabilities in Microsoft Fabric

Microsoft has announced new Purview enhancements for Microsoft Fabric at FabCon Atlanta, aimed at making enterprise AI adoption more secure and controlled. The updates strengthen data protection, gov…

https://petri.com/microsoft-fabric-purview-data-protection-governance/
12
🔐 security SecurityWeek
56%

Iran Readied Cyberattack Capabilities for Response Prior to Epic Fury

Analysis reveals a six-month buildup of Iran-linked cyber infrastructure, including US-based shell companies, designed to weather kinetic strikes and ensure the resilience of its global hacking opera…

https://www.securityweek.com/iran-readied-cyberattack-capabilities-for-response-prior-to-epic-fury/
13
🤖 ai The Decoder
56%

Pentagon plans to let AI companies train models on classified data

The US Department of War is working to set up secure environments where AI companies can train their models on classified data. Until now, models were only allowed to read classified data, not learn …

https://the-decoder.com/pentagon-plans-to-let-ai-companies-train-models-on-classified-data/
14
🤖 ai The Decoder
54%

Microsoft's superintelligence team ships MAI-Image-2, a text-to-image generator

Microsoft's superintelligence team has unveiled its first product: MAI-Image-2, an image generator rolling out across Microsoft's products and eventually available via API. The article Microsoft's su…

https://the-decoder.com/microsofts-superintelligence-team-ships-mai-image-2-a-text-to-image-generator/
15
🤖 ai The Decoder
53%

Google Deepmind upgrades Gemini API with multi-tool chaining and context circulation

Google Deepmind is expanding the Gemini API: developers can now combine multiple tools in a single request and tap into Google Maps as a data source. The article Google Deepmind upgrades Gemini API w…

https://the-decoder.com/google-deepmind-upgrades-gemini-api-with-multi-tool-chaining-and-context-circulation/