<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>AuthorityGate Blog - AI Governance</title><description>Short reads on AI governance, validation, and the ungoverned changes that take production down - each anchored to a real incident or published research.</description><link>https://authoritygate.com/</link><language>en-us</language><item><title>59% of Enterprises Run AI Agents. Only 26% Can Govern Them. ServiceNow Just Published the Gap.</title><link>https://authoritygate.com/blog/servicenow-ai-governance-gap-2026/</link><guid isPermaLink="true">https://authoritygate.com/blog/servicenow-ai-governance-gap-2026/</guid><description>ServiceNow surveyed 4,500 executives across 19 countries for its Enterprise AI Maturity Index 2026. The governance numbers should stop you cold.</description><pubDate>Thu, 02 Jul 2026 12:00:00 GMT</pubDate><category>Industry Research</category></item><item><title>Admins Blocked the Updates. The Platform Installed Them Anyway.</title><link>https://authoritygate.com/blog/mdm-approval-gates-silently-failed/</link><guid isPermaLink="true">https://authoritygate.com/blog/mdm-approval-gates-silently-failed/</guid><description>A caching misconfiguration made Windows Update treat managed devices as unmanaged for two days in June. Driver-approval policies stopped applying, and unapproved installs hit fleets by the tens of thousands.</description><pubDate>Wed, 17 Jun 2026 12:00:00 GMT</pubDate><category>Update Validation</category></item><item><title>Downtime Now Costs the Global 2000 $600 Billion a Year. The #1 Cause Is Still Human Error.</title><link>https://authoritygate.com/blog/splunk-600-billion-downtime/</link><guid isPermaLink="true">https://authoritygate.com/blog/splunk-600-billion-downtime/</guid><description>Two years ago Splunk priced unplanned downtime at $400 billion. The 2026 update says $600 billion, a 50 percent jump, and the leading cause has not changed.</description><pubDate>Tue, 02 Jun 2026 12:00:00 GMT</pubDate><category>Industry Research</category></item><item><title>79% of Enterprises Are Finding AI Apps IT Never Approved</title><link>https://authoritygate.com/blog/nutanix-shadow-ai-risk/</link><guid isPermaLink="true">https://authoritygate.com/blog/nutanix-shadow-ai-risk/</guid><description>Nutanix surveyed 1,600 cloud and engineering executives for its 8th Enterprise Cloud Index. Shadow AI is no longer an edge case - it is the norm.</description><pubDate>Tue, 17 Mar 2026 12:00:00 GMT</pubDate><category>Industry Research</category></item><item><title>81% of Leaders Say GenAI Is Moving Faster Than Their Risk Management</title><link>https://authoritygate.com/blog/cohesity-genai-outruns-risk/</link><guid isPermaLink="true">https://authoritygate.com/blog/cohesity-genai-outruns-risk/</guid><description>Cohesity surveyed 3,200 IT and security decision-makers across 11 countries. The cyberattacks are material, the financial fallout is public, and the AI risk gap is widening.</description><pubDate>Tue, 09 Dec 2025 12:00:00 GMT</pubDate><category>Industry Research</category></item><item><title>An AI Agent Ran 90% of a Nation-State Cyberattack. Anthropic Published the Receipts.</title><link>https://authoritygate.com/blog/anthropic-agent-ran-the-attack/</link><guid isPermaLink="true">https://authoritygate.com/blog/anthropic-agent-ran-the-attack/</guid><description>Anthropic disrupted what it calls the first reported AI-orchestrated cyber espionage campaign: a state-sponsored group used an agent to attack roughly thirty targets at machine speed.</description><pubDate>Tue, 25 Nov 2025 12:00:00 GMT</pubDate><category>AI Agent Security</category></item><item><title>Azure Built a Last-Known-Good Safety Net. The Bad Config Corrupted the Net Too.</title><link>https://authoritygate.com/blog/azure-last-known-good-failed/</link><guid isPermaLink="true">https://authoritygate.com/blog/azure-last-known-good-failed/</guid><description>Microsoft&apos;s post-incident review of the October 29 Azure Front Door outage reads like a case study in why config validation cannot be a single automated checkpoint.</description><pubDate>Tue, 11 Nov 2025 12:00:00 GMT</pubDate><category>Incident Analysis</category></item><item><title>The Security Patch Nobody Approved Took Down Half of Datadog for Two Days</title><link>https://authoritygate.com/blog/datadog-unapproved-update-outage/</link><guid isPermaLink="true">https://authoritygate.com/blog/datadog-unapproved-update-outage/</guid><description>A routine systemd security update, applied automatically through a legacy channel, knocked tens of thousands of nodes offline across five regions and three clouds - simultaneously.</description><pubDate>Tue, 14 Oct 2025 12:00:00 GMT</pubDate><category>Update Validation</category></item><item><title>91% of Companies Run AI Agents. 10% Have a Plan to Govern Their Identities.</title><link>https://authoritygate.com/blog/okta-agents-without-governance/</link><guid isPermaLink="true">https://authoritygate.com/blog/okta-agents-without-governance/</guid><description>Okta surveyed 260 executives across 12 countries for AI at Work 2025. The identity company found an 81-point gap between agent adoption and agent governance.</description><pubDate>Tue, 30 Sep 2025 12:00:00 GMT</pubDate><category>Agentic AI Governance</category></item><item><title>MIT Says 95% of GenAI Pilots Fail. The 5% That Survive Embrace Friction.</title><link>https://authoritygate.com/blog/mit-95-percent-friction/</link><guid isPermaLink="true">https://authoritygate.com/blog/mit-95-percent-friction/</guid><description>MIT&apos;s State of AI in Business landed like a bomb, and Forbes&apos; read on it is the interesting one: the projects that survive are the ones that stop avoiding friction.</description><pubDate>Tue, 09 Sep 2025 12:00:00 GMT</pubDate><category>Industry Research</category></item><item><title>Federal AI Use Cases Doubled in a Year. The Agencies Say Their Own Policies Can&apos;t Keep Up.</title><link>https://authoritygate.com/blog/federal-ai-outruns-policy/</link><guid isPermaLink="true">https://authoritygate.com/blog/federal-ai-outruns-policy/</guid><description>GAO counted federal AI use cases nearly doubling in a single year, with generative AI growing nine-fold - while the agencies themselves say policy cannot keep pace.</description><pubDate>Tue, 12 Aug 2025 12:00:00 GMT</pubDate><category>AI Governance</category></item><item><title>Grok Melted Down Twice in Two Months. Both Times xAI Blamed an Ungoverned Change.</title><link>https://authoritygate.com/blog/grok-two-ungoverned-changes/</link><guid isPermaLink="true">https://authoritygate.com/blog/grok-two-ungoverned-changes/</guid><description>A 3 AM prompt edit that bypassed code review in May. An upstream code update that ran wild for 16 hours in July. xAI&apos;s own statements are the case study.</description><pubDate>Tue, 22 Jul 2025 12:00:00 GMT</pubDate><category>Incident Analysis</category></item><item><title>Dell Just Made After-the-Disaster Validation 2.8x Faster. Before-the-Disaster Is Still Faster.</title><link>https://authoritygate.com/blog/dell-after-disaster-validation/</link><guid isPermaLink="true">https://authoritygate.com/blog/dell-after-disaster-validation/</guid><description>Dell&apos;s new all-flash Data Domain appliance validates cyber-vault data 2.8x faster. The industry is spending flash money on checking data after an attack.</description><pubDate>Tue, 10 Jun 2025 12:00:00 GMT</pubDate><category>Data Resilience</category></item><item><title>One Script Deleted 775 Customers in 23 Minutes. The Same Company Now Ships Thousands of AI Agents.</title><link>https://authoritygate.com/blog/atlassian-script-to-agents/</link><guid isPermaLink="true">https://authoritygate.com/blog/atlassian-script-to-agents/</guid><description>Atlassian&apos;s own 2022 postmortem describes a maintenance script that permanently erased 883 sites in 23 minutes. Three years later, 2,000 Rovo agents run in customer workflows.</description><pubDate>Tue, 27 May 2025 12:00:00 GMT</pubDate><category>Agentic AI Governance</category></item><item><title>OpenAI&apos;s Testers Said the Model Felt Off. It Shipped Anyway. Four Days Later: Full Rollback.</title><link>https://authoritygate.com/blog/openai-sycophancy-rollback/</link><guid isPermaLink="true">https://authoritygate.com/blog/openai-sycophancy-rollback/</guid><description>OpenAI&apos;s own postmortem of the GPT-4o sycophancy update is one of the most honest documents a vendor has published about why green metrics are not a launch gate.</description><pubDate>Wed, 14 May 2025 12:00:00 GMT</pubDate><category>Update Validation</category></item><item><title>98% Had a Ransomware Playbook. 57% Recovered Less Than Half Their Data.</title><link>https://authoritygate.com/blog/veeam-ransomware-playbook-gap/</link><guid isPermaLink="true">https://authoritygate.com/blog/veeam-ransomware-playbook-gap/</guid><description>Veeam&apos;s 2025 Ransomware Trends report surveyed 1,300 organizations, 900 of them attacked in the past year. The gap between paper preparedness and actual recovery is the story.</description><pubDate>Tue, 06 May 2025 12:00:00 GMT</pubDate><category>Data Resilience</category></item><item><title>The Kubernetes Component Built to Validate Changes Was the Way In. 43% of Cloud Environments Were Exposed.</title><link>https://authoritygate.com/blog/ingressnightmare-gatekeeper-hole/</link><guid isPermaLink="true">https://authoritygate.com/blog/ingressnightmare-gatekeeper-hole/</guid><description>Wiz&apos;s IngressNightmare disclosure: four CVEs in ingress-nginx, the worst a 9.8-critical unauthenticated RCE in the admission controller - the very component that validates changes before they enter the cluster.</description><pubDate>Tue, 08 Apr 2025 12:00:00 GMT</pubDate><category>Incident Analysis</category></item></channel></rss>