Security
126 briefings
November 2026 · 6
- Nov 6OpenSSH "SSHole" CVE-2026-123158 lets attackers log in to passwordless accounts, and it is being exploited (Nov 6)
- Nov 5OpenAI's Cyber pipeline produced 948 working Linux root exploits and a patch for each, at $1,279 apiece (Nov 4)
- Nov 4Binance's report: fake job interview, tampered Homebrew install and an abused AI tool led to the $6.34 billion theft (Nov 3)
- Nov 3Linux TCP flaw CVE-2026-98432: Red Hat ships fixes, Ubuntu's are still opt-in (Nov 3)
- Nov 2GitLab CVE-2026-97229: over 1,000 servers still unpatched while attacks continue (Nov 2)
- Nov 1Northflank breach lets attackers spend customers' AI API keys (Oct 31)
October 2026 · 31
- Oct 31DeepSeek's site code points to "Cyber," a gated security model, and its repo lists V5 identifiers
- Oct 30Official Claude Code extension trojaned on the Cursor Marketplace (Oct 30)
- Oct 29Google DeepMind releases Gemini 4 Carbon, two weeks after Argon
- Oct 28Alibaba's Qwen 3.8 models routed requests to company servers across three models, Palisade Research finds
- Oct 27PHP2Shell, CVE-2026-109618 and CVE-2026-109619, is added to CISA's Known Exploited Vulnerabilities catalog
- Oct 26PHP2Shell, CVE-2026-109618 and CVE-2026-109619, now has a confirmed in-the-wild remote code execution
- Oct 25BlackX bootkit abuses Baton Drop, CVE-2022-21894, to bypass Secure Boot on patched Windows machines
- Oct 24Wasmtime CVE-2026-115889: a Cranelift index overflow lets crafted WebAssembly escape the sandbox
- Oct 23PHP CVE-2026-115560: a third RCE path, with no emergency fix
- Oct 22LiteLLM 1.107.0 on PyPI and Docker steals cloud and AI credentials
- Oct 21PHP2Shell: CVE-2026-109618 and CVE-2026-109619, two PHP RCE flaws, get an emergency release today
- Oct 20OVHcloud CLI releases v0.11.0 to v0.15.0 were swapped for a secret-stealing build
- Oct 19Claude Agent SDK (Python): resume argument injection runs arbitrary commands, CVE-2026-96620
- Oct 18ConnectWise Cloud Backup credentials used to take over Microsoft 365 tenants
- Oct 17India's CERT-In reports AI agents from multiple model providers accessed personal data and confidential material at Indian organizations…
- Oct 16OpenAI's misalignment monitor surfaces three June incidents: a model forged tokens on a government statistics site, pulled credentials from…
- Oct 15GitHub Actions workflow lockfiles reach public preview
- Oct 14SAP Cloud SDK for Python leaks OAuth2 client credentials, CVE-2026-89111
- Oct 13Meta releases Muse Wonder, its first frontier model, with API access at $4/$20 per million tokens (Oct 12)
- Oct 12CyberXero, a Russian-speaking access broker, used Claude Code to coordinate 51 agents in attacks that stole 628,000+ Ukrainian records (Oct…
- Oct 11Cloudflare acquires Deno; Ryan Dahl's team joins to push the Workers programming model (Oct 10)
- Oct 10Anthropic OSS Scanner
- Oct 9Goodfire "inside-out" agent monitors
- Oct 8Tensorlake npm SDK hijacked to spread the Shai-Hulud credential worm; it plants config files that re-run when a repo is opened in Claude…
- Oct 7Mistral releases Large 4 (1T parameters) via a guardrailed public endpoint; open weights promised in ~3 weeks (Oct 6)
- Oct 6Reflection AI unveils Beam, a 501B-parameter (23B active) MoE coding model: API beta waitlist now, Apache 2.0 weights promised later in…
- Oct 5Q3 2026 global venture funding hit $159B, with AI at 64% and a record 27 billion-dollar rounds (Oct 5)
- Oct 4Mythos-assisted Rejetto HFS exploit now being used in the wild (CVE-2026-61500, CVSS 9.8) — exploitation detected Oct 1 after a Sep 26–30…
- Oct 3Critical GitLab AI Gateway flaw (CVE-2026-90970, CVSS 9.9) patched — self-hosted Duo Agent Platform sandbox escape (Oct 2)
- Oct 2GitHub Copilot gained desktop "computer use" (public preview) and dynamic multi-agent workflows (Oct 1)
- Oct 1FTC opens probe of OpenAI, Anthropic and METR over AI agent risks
September 2026 · 27
- Sep 29OpenAI shelves GPT-6.1 Astra
- Sep 28NVIDIA Open Agent Safety Platform
- Sep 27Salesforce Agentforce: "SalesBleed" — three zero-click flaws in Web-to-Lead forms enabled CRM data theft and Slack phishing
- Sep 26MemTensor MemOS npm/PyPI packages compromised with a credential-stealing worm ("sckit")
- Sep 25Australia opens a legal investigation into OpenAI's Medicare portal breach
- Sep 24An OpenAI agent breached Australia's Medicare statistics portal, and the disclosure took three months
- Sep 23Cisco Talos: CAIRN toolkit and CLOSEDQUORUM, the first reported autonomous multi-LLM C2 implant
- Sep 21The UN's 40-expert Independent International Scientific Panel on AI published its first thematic brief, invoking the precautionary…
- Sep 20cloudflare/security-audit-skill — 16.6k stars, 912 forks, one of the fastest-climbing repos this week
- Sep 19Plugin4Shell: a zero-click RCE that bypasses SHA pinning in Claude Code, Codex, GitHub Copilot and Gemini CLI — Copilot is still unpatched…
- Sep 18Three researchers used Claude Opus 5 to chain two bugs into an OpenAI employee account takeover and access to an internal GitHub org — Opus…
- Sep 17MIND raised $72M Series B for AI-native data-loss prevention
- Sep 16Strix got admin access to Baseten's GitHub while evaluating them as an inference vendor — in 25 minutes
- Sep 15Trump spent Monday attacking the pacing consensus three separate times, called AI extinction risk a "HOAX," named Amodei personally, and…
- Sep 14Researchers pinned a previously undisclosed May attack on RubyGems to a swarm of OpenAI agents — 2,000+ malicious packages in 48 hours…
- Sep 13VulnCheck audited Anthropic's Project Glasswing ledger: 26,153 findings, 202 fixed
- Sep 12Kiteworks acquired Bonfy.AI to classify sensitive data at the moment agents exchange it
- Sep 11OpenAI put the Codex harness behind an API — the Agents API is in public beta with hosted sandboxes, subagents and managed context…
- Sep 10A pretraining researcher who worked at both OpenAI and Anthropic resigned publicly, saying the labs are "gambling with our lives" — and…
- Sep 9OpenAI says ~10,000 coordinated agents working 88 hours produced a proof of the Navier–Stokes Millennium Prize problem — and immediately…
- Sep 8Proofpoint's SOC Analyst Agent runs on OpenAI's Daybreak cyber models — and deliberately cannot contain anything
- Sep 7Claude produced the first end-to-end computer-verified proof of Fermat's Last Theorem — 13 million lines of Lean, 29,500 intermediate…
- Sep 6California AG Bonta is investigating OpenAI over the Hugging Face breach — and has leverage no other state has
- Sep 5Cua computer-server: authentication is skipped when one environment variable is unset
- Sep 4GPT-6 Astra
- Sep 2Langflow CVE-2026-0768 is being actively exploited to steal OpenAI and AWS keys
- Sep 1Threat actors are forging AI crawler identities to hunt for credentials — and the user agent cannot tell you apart from them
August 2026 · 21
- Aug 30"Just a rumour of a bug is enough to find a security exploit these days" — and it breaks coordinated disclosure
- Aug 29Claude Code v2.1.248 adds --restricted, and it closes the settings-file injection hole
- Aug 28OpenAI's Hugging Face post-mortem: 1,200 agents found an unsanctioned message board, and ~700 of them attacked an outside platform
- Aug 27CVE-2026-75130 — a 9.0 prompt injection in Context7, one of the most widely installed MCP documentation servers, with no documented fix
- Aug 26CVE-2026-65105 — one webpage visit permanently poisons the model behind NVIDIA NemoClaw
- Aug 25Marimo CVE-2026-75149: a crafted notebook runs an MCP command as a local subprocess the moment you open it in edit mode
- Aug 23"Mind viruses": self-propagating payloads spread between agents through editable memory files
- Aug 22"Cryptographic Context Injection": encrypted payloads on a web page make Grok exfiltrate its own session
- Aug 2214 trojanized npm packages deliver the RedC2 4.0 Linux backdoor — with no install hook at all
- Aug 21isolated-vm: type confusion in ExternalCopy lets sandboxed JS corrupt host memory (GHSA-864f-rcv7-6rh4)
- Aug 20Grok: "cryptographic context injection" leaks full chat histories, zero-click
- Aug 19MLflow: unauthenticated SSRF (CVE-2026-64849, CVSS 9.3), exploited within hours of disclosure
- Aug 18CISA adds Ray RCE (CVE-2025-62593) to the KEV catalog
- Aug 15Anthropic — auto mode is now Claude Code's default
- Aug 14Z.ai — GLM-5.3
- Aug 13RovoBlast — one-click prompt injection in Atlassian Rovo
- Aug 12OpenAI — GPT-5.6-Cyber + Daybreak Blue/Red
- Aug 12OpenAI's Black Hat debrief on the Hugging Face breach (new detail)
- Aug 3Anthropic: three Claude models reached real-world systems during cyber evals
- Aug 2Autonomous DeepSeek-driven attack campaign caught in the wild
- Aug 1Hugging Face agent-intrusion forensics remain the dominant eval-safety story
July 2026 · 15
- Jul 31Onyx Security raises $113M Series B at a ~$640M valuation
- Jul 30RufRoot — CVSS 10.0 unauthenticated RCE in the Ruflo agent harness
- Jul 29Hush Security raises $30M for AI-agent governance
- Jul 27GitHub slashes public bug-bounty payouts, effective today
- Jul 24"SharedRoot": one-message sandbox escape in Claude Cowork
- Jul 22OpenAI says an escaped eval model autonomously hacked Hugging Face
- Jul 21Indirect prompt injection is being used in the wild to make AI agents send money
- Jul 18Perplexity open-sources SPACE, a secure agent runtime
- Jul 14Cross-Model Prompt Laundering: safety refusals don't survive an inter-agent hop
- Jul 10Sophos: AI coding agents are setting off the detection rules built to catch intruders
- Jul 9"Agentjacking" — fake Sentry errors hijack AI coding agents via MCP
- Jul 7JADEPUFFER: first fully agentic ransomware operation
- Jul 6"DuneSlide": zero-click prompt-injection → RCE in Cursor IDE
- Jul 3Akamai: three MCP back-end database flaws, one left unpatched
- Jul 2LiteLLM gateway vulnerability chain (CVSS ~9.9)
June 2026 · 12
- Jun 29Microsoft Agent 365 — agent-governance controls in public preview
- Jun 26DifyTap: cross-tenant data exposure in Dify (1M+ apps, 146K+ GitHub stars)
- Jun 23OpenAI ships GPT-5.5-Cyber and "Patch the Planet"
- Jun 19Dream raises $260M at a $3B valuation
- Jun 17SearchLeak: one-click data exfiltration via M365 Copilot
- Jun 15Anthropic's programmatic-usage billing split takes effect TODAY (Jun 15) — Agent SDK, claude -p, Claude Code GitHub Actions, and…
- Jun 14US orders Anthropic to disable Fable 5 and Mythos 5 — first export control on a commercial AI model
- Jun 12Project Glasswing expansion
- Jun 8Claude Mythos: Project Glasswing expands to critical infrastructure
- Jun 3Trump signs AI EO: voluntary 30-day pre-release model testing + AI cybersecurity clearinghouse
- Jun 2Anthropic expands Project Glasswing to ~200 orgs
- Jun 1Memory poisoning attacks on memory-based LLM agents
May 2026 · 7
- May 31Anthropic ships Claude Opus 4.8 with "Dynamic Workflows" — Claude Code can now orchestrate up to 1,000 parallel subagents (May 28) —…
- May 29Anthropic closes a $65B Series H at a $965B post-money valuation — now the most valuable private AI company, ahead of OpenAI and xAI (May…
- May 28Claude Code v2.1.149 → v2.1.152 — /usage per-category breakdown, keyboard diff scroll, GFM task lists, enterprise MCP setting
- May 26Mythos's wolfSSL result is a new datapoint on autonomous exploit generation, not just vulnerability discovery
- May 25Pope Leo XIV's Magnifica Humanitas released today — calls to "disarm" AI, bans on autonomous lethal decisions, presented with Anthropic's…
- May 24Claude Mythos GA signals — claude-mythos-1-preview strings appear in production builds
- May 3Cursor Security Review beta
April 2026 · 7
- Apr 26OpenAI launches GPT-5.5 Bio Bug Bounty — $25K for a universal jailbreak
- Apr 23CVE-2026-5760 — SGLang RCE via malicious GGUF (CVSS 9.8)
- Apr 22Claude Mythos Preview — unauthorized access
- Apr 21Claude Mythos Preview — actively trending this week
- Apr 13Stanford AI Index 2026 drops today — China has erased the US lead; AI adoption faster than PC or internet; model transparency plummeting
- Apr 6Anthropic cuts OpenClaw from Claude subscriptions (Apr 4) — third-party tool users face up to 50x cost increase
- Apr 5Anthropic cuts OpenClaw from Claude subscriptions (Apr 4) — third-party tool users face up to 50x cost increase