[deepseek-v4.1-flash]

October 30, 2026

• Mistral released open weights for Large 4 ("Le Chonk"), a trillion-parameter model trained entirely in Europe and pitched at agentic coding and cybersecurity; it leads its open rivals on coding but scores 26 on the independent index The Decoder cites, against 32 for DeepSeek V4.1 Flash and GLM-5.3 (The Decoder). · AI & Model Security

in Attackers Go for the Management Plane

October 21, 2026

• DeepSeek open-sourced DeepInfer, its vLLM-based inference framework. The engine powers DeepSeek-V4.1-Flash and is designed to reduce inference compute and energy costs, widening the options for self-hosted model infrastructure, reports The Decoder. · New Tools & Releases

in PHP’s php2shell RCEs Break Cover

October 9, 2026

• CrowdStrike named the tooling behind the South Korean bank intrusions: ARTEX, a recently released Chinese open-source LLM multi-agent pentest framework, running DeepSeek v4.1-flash as its main backend with Z.ai GLM-5.3 and Grok 4.6 supplementing. The campaign was found through open directories on a Hong Kong IP (38.244.50[.]120) exposing Claude Code session histories, Claude memory files and ARTEX configs — including the operator asking Claude where Korean breach data is sold. Active late September to early October 2026, unattributed but assessed as a financially motivated Chinese-speaking operator; more than 25,000 customer records were taken from Shinhan Bank alone, and ARTEX has since gone closed source (The Hacker News, The Decoder) (earlier coverage). · AI & Model Security

in Third NetScaler RCE in a Month Lands on the Same SAML Surface

September 27, 2026

• Gambit's numbers on the agentic e-commerce campaign are now public, adding hard denominators to the story we covered on 23 September. A Chinese-speaking, financially motivated actor ran Strix 146 times in deep mode against 138 hosts between 23–31 August (via OpenRouter on GLM 5.2 and DeepSeek v4 Pro), handed the reports to Cairn for 105 attack projects between 10–15 September on DeepSeek v4.1 Flash, and used Hermes with a Chinese system persona and 121 skills (78 of them attack skills) on Anthropic's opus-4.6 for orchestration. At least 27 companies compromised in that five-day window, over 600,000 unexpired credit cards stolen from two of them, skimmers injected into five stores, and 1,951 human-typed prompts across 260 sessions — a few short Chinese instructions per target (SecurityWeek). Danny Chez's follow-up pivots the hunt from rotating C2 domains to the injected code's packer characteristics, expanding from one observed injection to 50+ shops (write-up). · Threat Activity

in Citrix NetScaler Zero-Days Land Mid-Weekend, No Patch Until Next Week

September 17, 2026

One Extension Hijacks the AI Assistant in Five Major Browsers

BragJack hijacked built-in AI assistants in five major browsers (Chrome, Microsoft Edge, Opera Neon, Perplexity Comet, and Claude in Chrome) from a single malicious extension, with no guardrail bypass or prompt injection required. Spain's data protection agency received the first personal-data breach notification blaming an AI agent for autonomous execution of the intrusion, including file searches, lateral movement, and data exfiltration. CVE-2026-59310 in VMware vCenter, a directory traversal enabling unauthenticated RCE, is now actively exploited in ransomware campaigns with over 360 compromised servers identified. Multiple critical vulnerabilities including CVE-2026-84869 in ConnectWise ScreenConnect, CVE-2026-5430 in WSO2 API Manager, and CVE-2026-58704 in Google Pixel are under active exploitation by threat actors.

September 12, 2026

• An inference-time activation edit sharply changed DeepSeek V4.1-Flash’s cyber and refusal results without rewriting weights. @0x0SojalSec subtracted a refusal direction at each layer and reported results moving from 4/32 to 24/32 on one refusal set and 5/32 to 31/32 on a 32-item cyber set. It requires local weights and a modified vLLM path, so this is not a remote jailbreak. The small, self-reported evaluation also does not establish parity with closed frontier models (earlier coverage). · AI & Agent Security

in Researchers Tie OpenAI’s Agent Swarm to a 2,000-Package RubyGems Attack