mark carter

Duplicate reporting doesn’t improve cybersecurity 🤔 The government accountability office identified a whopping 117 existing cybersecurity regulations administered by 37 federal agencies across nine critical infrastructure sectors. Of those, it found that 80 regulations (about 70%) have affirmative reporting requirements, collectively imposing at least 125 separate reporting obligations on private sector entities. These obligations include cyber incident reporting, submission of cybersecurity plans and technical information, and reporting related to audits, reviews, and assessments. The report found that many regulations require regulated entities to provide similar information to different federal agencies, such that companies may need to prepare multiple reports concerning the same cybersecurity event or compliance activity. GAO highlighted potential overlap across all three reporting categories, including 48 regulations requiring cyber incident reporting, 52 requiring cybersecurity plans or technical information, and 25 requiring audits, reviews, or assessments. (And this does not even touch on state obligations, which are proliferating. GAO highlighted the financial services sector, where entities are already subject to cybersecurity incident reporting obligations administered by the Securities and Exchange Commission, Federal Reserve, the Office of the Comptroller of the Currency, Federal Deposit Insurance Corporation, National Credit Union Administration, Department of the Treasury, Federal Trade Commission, and Commodity Futures Trading Commission. Once CIRCIA is finalized, many of those same organizations could also be required to report incidents to CISA. The report further notes that differences in definitions, reporting thresholds, triggers, timelines, and required content could require entities to prepare multiple reports about the same incident for different federal recipients. These concerns are not unique to the financial services sector. Several sectors, including communications and transportation, also have myriad overlapping obligations. https://www.jdsupra.com/legalnews/gao-confirms-cyber-reporting-burdens-as-9175072/ #Infosec #Legal

Interesting 🛡️ World's Largest AI Model Repository Hugging Face Breached by Autonomous AI Agent. The starting point of the attack was the data processing pipeline itself, with a malicious dataset abusing two code execution paths, viz., in its remote code dataset loader and a template injection in a dataset configuration, to run code on a processing worker. With that access, the threat actor is said to have escalated to node-level access, collected cloud and cluster credentials, and moved laterally into several internal clusters over a weekend. The company also said it turned to Z.ai's GLM 5.2, a Chinese open-weight model, to conduct the forensic analysis after Western frontier models refused requests containing real attack commands, exploit payloads, and command-and-control (C2) artifacts because their safety guardrails were triggered and their inability to differentiate between an attacker and a legitimate incident response effort. "This experience points to a gap worth planning for," the New York-headquartered company said. "We do not know which model powered the attacker's agents, whether a jailbroken hosted model or an unrestricted open-weight one; either way, the attacker was bound by no usage policy, while our own forensic work was blocked by the guardrails of the hosted models we first tried." "The practical lesson for defenders: have a capable model you can run on your own infrastructure vetted and ready before an incident, both to avoid guardrail lockout and to keep attacker data and credentials from leaving your environment." https://thehackernews.com/2026/07/worlds-largest-ai-model-repository.html #InfoSec

Recommended read for every CISO 🛡️ Detecting Agentic Threats in Claude: Writing Rules on the Execution Layer 🛡️ “In this post I look at the main threats from agentic platforms, and how we can use the execution-layer telemetry we're getting from Claude to write detections for them. If you followed along with the last post, you will have closed the telemetry gap left by the Compliance API. Cowork, Claude Code, and the Office agents will now be streaming their execution-layer events (tool calls, MCP invocations, file paths and approval decisions), into your SIEM via OpenTelemetry. The Compliance API feed tells you what was said; the execution layer tells you what was done. In this post we look at detections that need both. Part 2's detections read content and judged intent: Was this prompt a jailbreak? Was this upload a poisoned document? The detections for agentic attacks work the other way round. They read actions, and they ask a different question: Should this action have happened? In this order? Approved by whom?” https://lnkd.in/g-efi9hJ #Infosec