Entity trail
Breaking Defense
Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.
Briefing refs
0
Findings
5
Edges
1
Sources
5
Corpus findings
- 2026-06-24 / thought-leaders-researcherZico Kolter and Matt Fredrikson on Latent Space: 'AI Security Is Not Just Cybersecurity With AI'On swyx's Latent Space (June 22), OpenAI board member Zico Kolter and Gray Swan CEO Matt Fredrikson argued AI security is a distinct discipline — as enterprises deploy autonomous agents like Claude Code and OpenClaw, prompt injection and indirect attacks form a genuinely new risk class. They contend specialized red-teaming models now outperform humans at breaking AI systems, so robust defense requires dedicated guardrail models (e.g., Cygnal) plus continuous adversarial testing via platforms like Gray Swan Arena. The builder takeaway: agent security needs purpose-built models and ongoing red-teaming, not bolt-on filters.
- 2026-05-24 / arxiv-researcherTimeGuard: First Backdoor Defense Tailored for Time Series ForecastingTimeGuard addresses an underexplored threat: backdoor attacks on time series forecasting models used in finance, energy, and healthcare. Systematic evaluation of 13 existing backdoor defenses across the TSF lifecycle reveals two fundamental failure modes — data entanglement diluting channel-level signals and task-formulation shifts breaking classification-oriented defenses. TimeGuard introduces channel-wise pool training to isolate and neutralize backdoor triggers without degrading forecasting accuracy.
- 2026-05-12 / arxiv-researcherRe-Triggering Safeguards: Embedding Disruption Method Reactivates Built-In LLM SafetyThis paper proposes a jailbreak detection method that works by disrupting embeddings to re-activate the model's built-in safeguards rather than serving as a standalone defense. The insight is that jailbreaking prompts are inherently fragile — small perturbations in embedding space cause the model to recognize the harmful intent it was tricked into missing. Unlike prior defenses, this approach is complementary to existing safety layers rather than replacing them.
- 2026-05-04 / arxiv-researcherDefense Against Poisoning Attacks in Shuffle-DP: Breaking the Honest-User AssumptionThis paper addresses a critical weakness in shuffle differential privacy protocols: the assumption that all users behave honestly. In real-world deployments, adversarial users can exploit this vulnerability through poisoning attacks that compromise both privacy guarantees and utility. The proposed defense works under full-bandit feedback conditions, making it practical for production shuffle-DP systems where user honesty cannot be guaranteed.
- 2026-04-24 / agents-researcherPentagon Workers Vibe-Code 100,000+ AI Agents in Five Weeks on GenAI.mil — IL5 Authorization, Classified Expansion PlannedDefense Department personnel created over 103,000 AI agents using Google Gemini's Agent Designer on GenAI.mil in under five weeks, logging 1.1 million agent sessions. The low-code/no-code 'vibe-coded' agents have Impact Level 5 authorization for unclassified networks, with expansion to classified and top-secret systems under active discussion. Most popular agents automate staff work like After Action Reports and operation estimates.
Graph relationships
Source trail
Graph sources
entity graphfindings textkg entities