The HackWednesday archive
Security analysis, incident lessons, and practical guides for teams working with AI. Search the archive or choose a topic.
Incident Response
MiniPlasma is a newly published Windows privilege-escalation proof of concept that reportedly revives the old CVE-2020-17103 path and turns a standard user foothold into SYSTEM access. The bigger lesson is about patch confidence, regression risk, and why defenders need validation beyond release notes.
AI-assisted Awaiting editor review
AI in Security
AWS used mid-May 2026 guidance to make a useful point for defenders: secure AI programs start with identity, access, and guardrails in the prototype phase rather than after agents reach production.
AI-assisted Awaiting editor review
AI in Security
Microsoft's May 12, 2026 MDASH release matters because it ties agentic AI directly to 16 Patch Tuesday vulnerabilities, shifting the conversation from demos to measurable defensive outcomes.
AI-assisted Awaiting editor review
AI in Security
OpenAI's new Daybreak initiative reframes cyber defense around resilient-by-design software, Codex-powered remediation workflows, and a tiered trusted-access model for increasingly cyber-capable AI.
AI-assisted Awaiting editor review
AI in Security
OpenAI's May 7 GPT-5.5-Cyber rollout, new phishing-resistant access requirements, and parallel NIST testing agreements all point to the same shift: advanced AI security capability is being governed more like privileged infrastructure.
AI-assisted Awaiting editor review
AI in Security
Fresh NIST and Microsoft updates point to the same operational reality: security teams need ways to evaluate, inventory, and govern AI agents before trust in them can scale.
AI-assisted Awaiting editor review
AI in Security
LiteLLM is now dealing with a different kind of security problem than the March supply-chain incident: active exploitation of a critical pre-auth SQL injection that puts upstream model-provider credentials and environment secrets at risk.
AI in Security
OpenAI's April 29 cyber action plan argues that AI-powered defense should be distributed broadly, and recent Microsoft and Google moves suggest the industry is starting to build the operational infrastructure to do it.
AI-assisted Awaiting editor review
AI in Security
Late-April updates from OpenAI and Microsoft point to the same security reality: AI is compressing the time between discovery and exploitation, so defenders need faster access, remediation, and control loops.
AI-assisted Awaiting editor review
AI in Security
Google Cloud Next 2026 and Wiz's April product updates make the same argument: AI security is becoming a code-to-cloud discipline built around agent identity, shadow AI visibility, and guardrails for AI-generated software.
AI-assisted Awaiting editor review
AI in Security
Model Context Protocol can make AI tools dramatically more useful, but it also expands trust boundaries. Security teams should treat MCP like a privileged integration layer: sandbox servers, minimize scopes, block token passthrough, defend against SSRF, and review every tool as a potential remote-action surface.
AI in Security
Microsoft's April 22 security update argues that stronger AI models are compressing the time between vulnerability discovery and exploitation, forcing defenders to treat patch speed and exposure management as urgent runtime problems.
AI-assisted Awaiting editor review
