Signal criticality: High
What happened: Help Net Security reported that threatModeler introduces Nexus to automate threat modeling with AI governance ThreatModeler has announced the general availability of ThreatModeler Nexus, an agentic threat modeling platform that brings governed, architecture-aware security to the way modern software is actually built. As AI writes a growing share of production code, the question is no longer whether to threat model, but where and when. ThreatModeler Nexus answers that with a platform built to threat model everything, starting wherever a team already is.
Key takeaways:
Original source: https://www.helpnetsecurity.com/2026/06/26/threatmodeler-introduces-nexus-to-automate-threat-modeling-with-ai-governance/
Signal criticality: High
What happened: SecurityWeek reported that research presented at the USENIX conference found that, in controlled tests, inserting five specially crafted texts per target question caused a RAG system to produce the attacker’s chosen answer in about 90% of cases, even when its knowledge base contained millions of legitimate texts. Human-in-the-loop traps could use a compromised agent to mislead the person expected to approve its actions. Artificial Intelligence When Information Becomes the Attack Surface Understanding AI Agent Traps From hidden content injections to cognitive state poisoning, attackers are turning trusted data sources into traps for autonomous AI.
Key takeaways:
Original source: https://www.securityweek.com/when-information-becomes-the-attack-surface-understanding-ai-agent-traps/
Signal criticality: High
What happened: The Hacker News published "New Gaslight macOS Malware Uses Prompt Injection to Disrupt AI-Assisted Analysis". A previously undocumented Rust-based macOS implant and information stealer has been found to embed a prompt injection payload designed to trick a malware analyst's artificial intelligence (AI) tools and trick it into aborting or refusing an analysis of the artifact. The malware has been codenamed Gaslight owing to this deceptive behavior. It's been assessed with high confidence that the tool is The article focuses on a concrete model, prompt, data, or integration risk with operational security implications.
Key takeaways:
Original source: https://thehackernews.com/2026/06/new-gaslight-macos-malware-uses-prompt.html
The strongest signal today is that AI security is being decided in the surrounding control layer — permissions, connectors, deterministic workflow design, response speed, and the infrastructure that still underpins trust. That is a more durable framing than generic agent hype, and it is the one worth carrying forward.