AI Red Team Engineering notes, research references, and field notes on AI systems, responsible AI, governance, and compliance.
Global search

Search AI Red Team

Search across the complete knowledge base.

Enter at least two characters to begin.
↑↓ select Enter open Browse the full research archive
Oracle documentation · October 10, 2026

Oracle Agent Memory 26.8: verify expiry, physical purge, and runtime privileges

guide Agent SecurityAI EngineeringAI Compliance
Why it matters

This October 10 documentation review covers retention in Oracle Agent Memory 26.8. The guide distinguishes expired records being hidden from searches from those records being physically deleted. Schema defaults and per-record time-to-live settings control expiry; scheduled database jobs remove expired records and leftover retrieval chunks. Setup can finish with a warning when the schema owner lacks permission to create those jobs, leaving search filtering active without completing physical cleanup. The guide also separates schema setup from runtime access. Its procedures cover the active memory store; backup and export deletion require a separate retention policy.

My takeaway: In a disposable database, create synthetic memories with short retention, inspect both direct rows and search results after expiry, and verify successful purge jobs. Test missing scheduler privileges separately; an empty search result is insufficient evidence of deletion.
Keep exploring

More curated notes connected through Agent Security and AI Engineering.

OpenAI News · framework

Pacing model development in an era of cyber-critical capabilities

OpenAI says preliminary evidence that Astra may meet its Critical cybersecurity threshold led it to pause frontier reinforcement-learning work for two weeks and keep its largest planned run on hold. New safeguards include stronger workload and network isolation, continuous boundary testing, token-level monitoring that escalates suspicious tool activity, and broader alignment checks for deception, reward hacking, and unauthorized access.

OWASP GenAI Security Project · guide

OWASP Top 10 for Agentic Applications for 2026

OWASP's community guide organizes agentic-system risk into ten categories, including goal hijacking, tool misuse, identity and privilege abuse, memory poisoning, insecure inter-agent communication, cascading failures, and rogue-agent behavior. It provides a shared taxonomy and mitigation starting point rather than a certification checklist or evidence that a deployed system is secure.

OpenAI News · framework

A blueprint for democratic governance of frontier AI

OpenAI proposes a three-part U.S. frontier-AI governance model: harmonize emerging state safety laws into a federal baseline, strengthen CAISI as an evaluation and standards institution, and coordinate a broader resilience program. Proposed controls include severe-risk evaluations, transparency reports, independent audits, safety-incident reporting, model-weight security, whistleblower protection, and periodic technical assessments.