Every article scored, classified, and mapped to MITRE ATLAS and OWASP LLM Top 10 — so you always know what matters and why.
AI security intelligence analysis for 2026-W35 — MITRE ATLAS technique trends, OWASP LLM risk distribution, threat actor activity, and enterprise readiness assessment based on 25 articles.
Read full analysis →
A three-flaw vulnerability chain dubbed AutoJack in Microsoft's AutoGen Studio allowed attackers to execute arbitrary commands on a developer's host system by manipulating a browsing AI agent into connecting to a malicious webpage. The attack exploited missing authentication on MCP WebSocket routes combined with unsanitised base64-encoded parameters to launch arbitrary processes. Microsoft confirmed the flaw was patched before any PyPI release, limiting exposure to developers building directly from the main GitHub branch.
AWS has launched Amazon Bedrock AgentCore Payments, a managed infrastructure layer that enables AI agents to autonomously transact with external model providers and services using the x402 payment protocol, without human intervention. This capability introduces a new class of financial attack surface where compromised or manipulated agents can autonomously spend real funds, exfiltrate value, or be redirected to malicious service endpoints. Defenders must now treat agent payment credentials and spending budgets as first-class financial controls, on par with cloud IAM policies.
Mindgard researchers demonstrated that ChatGPT's image generation pipeline can be manipulated through an indirect, socially-engineered prompt to produce violent and sexually explicit content without users directly requesting it, exposing a significant failure in OpenAI's content moderation controls. Defenders and enterprise operators of ChatGPT-integrated products face a newly validated attack class where innocuous-looking prompt patterns — potentially spreading virally — can systematically strip safety guardrails from image generation. This finding signals that content filter bypasses in multimodal systems are reproducible at scale, raising urgent questions about the adequacy of output-layer filtering as a sole defence mechanism.
Bayer AG and Thoughtworks have published a detailed case study on PRINCE, a production agentic RAG system combining multi-agent orchestration, Text-to-SQL, and human-in-the-loop workflows to answer complex pharmaceutical preclinical research questions and draft regulatory documents. The system's architecture — spanning intent clarification, planning, retrieval, reflection, and writing agents with access to decades of safety study data — introduces a broad attack surface including prompt injection across agent boundaries, SQL injection via natural language, and sensitive data exfiltration through compromised agent outputs. Defenders evaluating similar agentic platforms should treat each inter-agent handoff as a trust boundary requiring independent validation and focus on data leakage controls given the sensitivity of preclinical regulatory data.
Recall is an open-source, fully-local memory layer for Anthropic's Claude Code that persists and summarises project context across coding sessions without sending data to external services. For defenders, the introduction of a persistent, file-based context store creates a new attack surface: a poisoned or tampered memory file can silently inject malicious instructions into every subsequent Claude Code session. Security teams should treat the local memory store as a trusted-input boundary and apply appropriate file-integrity and access controls.
OpenAI has upgraded ChatGPT's health and wellness response capabilities via GPT-5.5 Instant, incorporating stronger reasoning, physician-informed evaluations, and improved contextual understanding for medical queries. This expansion into high-stakes health guidance raises meaningful concerns for defenders, as improved fluency and authority in medical responses increases the risk of user overreliance and lowers the perceived threshold for trusting AI-generated health advice. Security and trust-safety teams should evaluate how this capability interacts with prompt injection, social engineering chains, and the broader risk of AI-mediated medical misinformation at scale.
A malware developer has been observed embedding fake system instructions and policy-triggering content — including references to nuclear and biological weapons — inside JavaScript comment blocks to confuse or trigger refusal behaviour in LLM-powered security analysis pipelines. The technique does not affect code execution but is specifically designed to disrupt naive AI-first triage tools that feed raw file content to language models without isolating it as untrusted data. Traditional static analysis methods remain unaffected, but the approach signals an emerging class of anti-AI-analysis evasion techniques.
A new class of agentic AI security platforms is emerging that autonomously correlates threat intelligence, validates controls, and prioritizes remediations across siloed enterprise security tooling — moving beyond assistive chatbot interfaces to continuous, multi-step autonomous action. This shift introduces significant new attack surface: an AI system with persistent access to live exposure data, security telemetry, and remediation workflows becomes a high-value target for adversarial manipulation. Defenders must assess trust boundaries, prompt injection risks, and the consequences of autonomous action taken on poisoned or manipulated inputs before deploying these systems.
Token Security has published analysis and launched a platform addressing the growing security gap created by AI agents operating as unmanaged identities within enterprise environments, connecting to critical systems like Salesforce, GitHub, Snowflake, and production databases with minimal governance. Most organizations have deployed AI agents using credentials provisioned for other purposes, creating high-privilege, low-visibility actors outside the scope of existing IAM controls. Defenders now face a sprawling, machine-speed identity layer that existing lifecycle management, least-privilege enforcement, and audit tooling were never designed to handle.
GitHub has published a detailed engineering account of how it built an internal data analytics agent using GitHub Copilot, exposing the architectural patterns — including natural language-to-SQL translation, autonomous tool invocation, and internal data access — that underpin such systems. For defenders, this blueprint highlights concrete risks around prompt injection into analytics pipelines, excessive agency over sensitive internal datasets, and the challenge of auditing LLM-generated queries before execution. Organisations adopting similar agentic analytics patterns should treat this as a reference threat model rather than a safe-to-copy architecture.
Microsoft researchers disclosed AutoJack, an exploit chain targeting AutoGen Studio's MCP WebSocket endpoint that allows a single malicious web page to execute arbitrary commands on a developer's host machine via an AI browsing agent. The attack chains three distinct weaknesses — localhost trust bypass, missing authentication on MCP paths, and unsanitised command execution — requiring no credentials or user interaction beyond the agent loading the attacker's URL. While the vulnerable handler was not included in stable PyPI releases, it shipped in two pre-release builds that remain unyanked, leaving anyone who installed those versions exposed.
Karamo Brown's Kē wellness app deploys an AI digital clone of the celebrity — voice, persona, and advisory content — built by Delphi from interviews, podcasts, and public clips, enabling real-time conversational coaching at scale. For defenders, celebrity-clone architectures introduce layered risks: the training corpus is largely public and manipulable, the voice synthesis surface is exploitable for deepfake derivation, and the mental-health context creates elevated harm potential if the persona is hijacked or jailbroken. Security teams evaluating similar deployments should treat the persona boundary as a primary control point, since users in vulnerable emotional states are disproportionately exposed to manipulation if guardrails fail.
AWS has released a deep observability layer for SageMaker AI inference endpoints, emitting over 100 metrics covering GPU health, KV cache pressure, token-level latency, and traffic distribution into a native CloudWatch Insights dashboard with PromQL-compatible export. For defenders, this centralised telemetry surface introduces new reconnaissance and exfiltration vectors: an adversary with read access to CloudWatch or connected third-party tools (Grafana, Datadog) can infer model architecture, request patterns, and capacity limits without touching the model itself. The richness of these signals also raises insider-threat risk, as operational staff now have granular visibility into inference behaviour that can be leveraged to reverse-engineer model characteristics or plan targeted denial-of-service campaigns.
AWS has made Amazon Bedrock AgentCore Harness generally available, providing a managed abstraction layer that reduces agent deployment to two API calls while bundling sandboxed compute, persistent memory, tool gateway, browser access, identity management, and observability. For defenders, this dramatically lowers the barrier to deploying autonomous agents with filesystem access, shell execution, web browsing, and multi-provider model switching — compressing what was a weeks-long infrastructure project into minutes. Security teams face an expanded attack surface where prompt injection, tool abuse, cross-session memory poisoning, and supply chain risks through AWS-curated skill catalogs now arrive as a single, tightly integrated managed service rather than individually reviewable components.
Researchers at Microsoft identified a three-stage exploit chain in AutoGen Studio that allows a malicious web page visited by a browsing AI agent to reach the host's local Model Context Protocol (MCP) WebSocket and spawn arbitrary processes. The chain exploits a bypassable origin allowlist, authentication middleware that excluded MCP endpoints, and unsanitised URL-derived command parameters. Although the vulnerable surface was never shipped in a PyPI release, the finding exposes a systemic architectural risk in any agent framework that combines untrusted browsing with privileged localhost services.
◉ AI THREAT BRIEFING
Twice-weekly digest of critical AI security developments — every story mapped to MITRE ATLAS and OWASP LLM Top 10. Free.
No spam. Unsubscribe anytime.