ZeroHour
DashboardNewsCVEsExploitsIncidentsAIBriefAnalyticsEntitiesWatchlistIOCsAsk the feedSourcesFeeds & APIKiosk mode

Search: “GateKeeper”

13 items in the last 7d

Searchkeywordsmart Sortlatesthotimportancerelevance
DomainAllCyberAIWindow24h3d7d30dAllIn the wildImportant onlyShow duplicates
CategoryAllAI safety & security 4Vulnerability 2AI tools & infra 1AI industry 1Malware 1Phishing & fraud 1Tools 1Exploit / PoC 1AI policy 1
SeverityAllcritical 1high 2medium 3low 0info 7

PuzzleMask: Abusing Plain Prose as a Covert AI Attack Vector

Check Point details PuzzleMask, a plain-prose technique that bypasses LLM gatekeeper policy checks, letting hidden payloads reach target models unreviewed.

Check Point Research describes PuzzleMask, a prompt-crafting technique that hides policy-violating payloads inside plain-English prose wrappers, bypassing quick LLM-based policy checks without emojis, Base64, or invisible formatting. The researchers tested 23 automated prompts against gatekeepers including GPT-4o-mini, GPT-OSS-Safeguard 20b, Claude 3 Haiku, and Llama Guard 3, and all were classified as safe despite policies that flagged the plain versions. When submitted to GPT-5 in thinking-high mode with a Python interpreter, the target model extracted and acted on the payload in over 90% of trials. The technique is not itself a jailbreak but can carry a jailbreak prompt as payload; mitigations include input paraphrasing, hardened gatekeeper policies, and output monitoring.

Check Point Researchupdated · 6d agofirst · 6d agoAI safety & security 2 sources

New AI Attack Hides Malicious Instructions in Normal-Looking Text to Evade Safety Filters

Check Point researchers show crafted prose hides policy-violating instructions that bypass all tested LLM gatekeepers, including GPT-4o mini and Llama Guard 3.

A new prompt-crafting technique embeds malicious payloads inside grammatical, natural-looking text without Base64, invisible Unicode, or obvious encodings, defeating lightweight pre-screening gatekeepers. In testing, all four evaluated gatekeeper models—gpt-4o-mini-2024-07-18, gpt-oss-safeguard:20b, claude-3-haiku-20240307, and llama-guard3:8b—classified the crafted wrappers as safe at a 100% bypass rate across 23 obfuscated prompts. GPT-5 Thinking in high-reasoning mode recovered and acted on the hidden instruction in 17 of 18 tests (~94.4%), often spending over a minute and multiple Python executions. Researchers recommend paraphrasing untrusted input, hardening gatekeeper policies, and applying defense-in-depth controls for agentic deployments.

GBHackers · 6d agoAI safety & security 2 sources

Hackers Can Hide Malicious AI Commands Inside Normal English to Bypass Security Filters

Check Point's PuzzleMask technique hides malicious prompts in ordinary English that fast gatekeeper models miss but high-reasoning downstream models execute.

Check Point researchers disclosed PuzzleMask, a technique concealing policy-breaking instructions in natural-language prose without encodings or invisible characters. Fast screening models classified all 23 crafted wrappers as safe, while a high-reasoning model recovered and acted on the hidden instruction in 17 of 18 tests (94.4%). The gap stems from capability imbalance between gatekeeper and target models, with defenses including paraphrasing untrusted input, stricter self-referential wording rules, and output/tool-call monitoring.

Cyber Security News · 6d agoAI safety & security2

Apple Updates Everything, (Mon, Sep 14th)

Apple patched a record 261 vulnerabilities across iOS 27, macOS Golden Gate 27, watchOS 27, and other platforms, with none flagged as exploited.

Apple's annual OS update shipped iOS/iPadOS 27, macOS Golden Gate 27, tvOS 27, watchOS 27, and visionOS 27 plus bug-fix releases for the 26 and 15 branches, fixing a record 261 vulnerabilities. Notable issues include multiple kernel flaws allowing root privilege escalation (CVE-2026-43689, CVE-2026-43691, CVE-2026-43698, CVE-2026-43786), remote code execution in CUPS (CVE-2026-43692), kernel memory corruption via malicious NFS servers (CVE-2026-43686, CVE-2026-43687), and WebKit memory corruption from crafted web content (CVE-2026-43715). No vulnerabilities are labeled as actively exploited, and Apple does not assign per-CVE severities. Users report iOS 26.7 being downloaded when iOS 27 is intended, and tools like Little Snitch and BlockBlock need updates before upgrading to macOS 27.

SANS Internet Storm Center · 2d agoVulnerabilityCVE-2022-3437CVE-2026-20683CVE-2026-28899+27 CVEs

ClickFix attacks infecting PCs and Macs are going viral

ClickFix social engineering attacks are spreading widely, delivering malware to Windows and macOS users by tricking them into running malicious terminal commands that bypass Gatekeeper.

BlueVoyant reports the Lorem Ipsum malware operation pivoted to ClickFix in late May 2026, dropping code-signing infrastructure and broadening its victim pool from Microsoft Teams searchers to anyone browsing a compromised website. Jamf and independent researchers documented macOS ClickFix variants that bypass Gatekeeper protections. Cisco Talos observed attackers abusing published Google Sheets for control traffic, while Netskope counted 5,400 sites beaconing to one campaign; Sandworm has separately hosted control infrastructure in blockchain smart contracts.

Ars Technica · Security · 6d agoPhishing & fraud in the wild1

14th September – Threat Intelligence Report

Check Point weekly digest: Microsoft's record 974-vuln Patch Tuesday ships two actively exploited Windows zero-days; IDScan.net, Mathspace, Revolut suffer breaches.

Microsoft's September 2026 Patch Tuesday addressed a record 974 vulnerabilities, including two actively exploited privilege-escalation zero-days, CVE-2026-85880 and CVE-2026-81963, plus 20 flaws allowing unauthenticated remote code execution. Disclosed breaches include IDScan.net (identity documents), Mathspace (over 1 million people via Metabase CVE-2026-72898), Revolut, and Florida DMV (ShinyHunters). GitLab fixed critical CVSS 10.0 path traversal CVE-2026-85706, and MikroTik fixed chainable RouterOS flaws CVE-2026-67276 and CVE-2026-86060. The report also covers the PuzzleMask LLM jailbreak technique, GoldFactory's Gigabud Android fraud, and the BlueMoon Chromium exploit chain (CVE-2026-85046).

Check Point Research · 3d agoExploit / PoC in the wildCVE-2026-72898CVE-2026-85880CVE-2026-81963+4 CVEs2· 1 read

16 governance tools for securing your AI fleet

CSO Online reviews 16 AI governance and security tools, including Collibra, Credo AI, F5/CalypsoAI, Fiddler AI, and Guardrails AI, for managing LLM risks.

CSO Online surveys 16 vendors in the emerging AI governance and guardrails market for keeping production LLMs in check. Featured products include Collibra's AI Command Center, Confident Security's OpenPCC, Credo AI's Govern AI Assistant, F5's acquired CalypsoAI, Fiddler AI's control plane, and Guardrails AI's Snowglobe simulator. The tools address hallucination tracking, PII leakage, prompt injection and jailbreak defense, and compliance with frameworks such as the EU AI Act, SOC2, ISO-42001, and GDPR.

CSO Online · 4h agoAI tools & infra

BambooToken: The Malware That Speaks MQTT to Stay Under the Radar

Lumen's Black Lotus Labs uncovered BambooToken, a Windows and Linux malware family using MQTT broker-based C2 and DLL sideloading across Asia since February 2023.

Lumen Black Lotus Labs identified BambooToken, a multiplatform malware family that exchanges commands through MQTT brokers so infected hosts never contact the C2 server directly, active from at least February 2023 through July 2026. The Windows variant sideloads via Tendyron's OnKey hardware-token software used in Chinese banking and government, or impersonates Kingsoft Office, without either vendor's signing certificate being compromised; a Linux build appeared by December 2025 with shell, file transfer, and system information commands. Victims include MikroTik and DrayTek routers in Singapore, Cambodia, and Vietnam reached after internet-wide SNMP scanning, and Lumen cannot attribute the family to any known actor.

Security Affairs · 16h agoMalware in the wild1

Apple Rolls Out Massive Security Update Fixing 273 Vulnerabilities Across Its Devices

Apple's coordinated rollout patches 273 unique vulnerabilities across iOS 27, macOS Golden Gate 27, watchOS and Safari, including remote code execution flaws.

Apple shipped one of its largest coordinated security updates on September 14, 2026, fixing 273 unique CVEs across iOS 27, iPadOS 27, macOS Golden Gate 27, watchOS 27, tvOS 27, visionOS 27, Safari 27 and Xcode 27. Highlights include CVE-2026-65414, a Bluetooth out-of-bounds write enabling remote code execution, and CVE-2026-84607, an AVEVideoEncoder race condition granting kernel privileges to sandboxed apps. macOS Golden Gate 27 covers the broadest set with 210 CVEs, and Apple states none of the flaws were exploited in the wild.

Cyber Security News · 1d agoVulnerabilityCVE-2026-65414CVE-2026-84607CVE-2026-64752+7 CVEs1

[AINews] AEF-1 standard emerges for Third Party Evaluators, as Xai, OpenAI, and Anthropic all cosign

xAI, OpenAI, and Anthropic cosign the AEF-1 third-party evaluation standard while Dario Amodei proposes embedded evaluators for safety verification.

The AI Evaluator Forum published AEF-1, a baseline standard for independent third-party AI evaluations covering access, conflicts of interest, funding relationships, recusal, and transparency, cosigned by xAI, OpenAI, and Anthropic. Dario Amodei wrote a rare personal blogpost proposing embedded evaluators such as METR with desks, badges, company laptops, and internal-risk-team-level access to verify safety commitments, plus democratic and global coordination frameworks. The roundup also covers the pacing debate: Bilal Chughtai left Google DeepMind arguing progress may outrun alignment, while critics including Aidan Gomez and Cohere push back against slowdowns and lab gatekeeping. Additional items include Cline Desktop's launch with open-weight model support.

Latent Space · 2d agoAI safety & security

Who gets to define the rules for AI?

Cohere CEO Aidan Gomez attacks big-lab antitrust exemption proposals as cartel behavior that lets incumbents write AI safety rules.

Cohere CEO Aidan Gomez argues that proposals from large AI labs—particularly Anthropic's roadmap requesting antitrust exemptions for safety coordination—amount to a cartel letting incumbents define rules for everyone else. He draws parallels to the 1975 SEC NRSRO credit-rating designations and the EU's 1985 Motor Vehicle Block Exemption, where safety justifications produced incumbent-protecting market structures. Gomez supports independent review of highly capable AI systems but disputes who writes the standards, who conducts review, and who participates. He also warns AI cyber offense is getting cheaper faster than defenses are improving.

Hacker News · AI · 3d agoAI policyHN 41↑ · 29 comments

The Rise of the Forward Deployed Engineer — and How To Do the Job Right

Palantir veteran Vinoo Ganesh traces the forward deployed engineer role and shares practices for building effective FDE teams.

Kepler CEO and former Palantir forward deployed engineer Vinoo Ganesh argues that labs, startups, and PE firms hire FDEs without a shared definition of the role. He recounts Palantir's Project Frontline rotation, which trained about 250 software engineers as FDEs, many now leading forward deployed teams at OpenAI, Anthropic, xAI, and Anduril. A 2013 failure of the Phoenix transaction store at a bank, where real-world data gaps caused roughly 2.3 million keyspaces and an out-of-memory crash, illustrates why FDEs must own the gap between design and production reality. At Kepler he places the FDE function inside product rather than sales.

Latent Space · 4d agoAI industry 4 sources1

WordPress Uses AI to Stop Malicious Plugin Updates Before They Reach Millions of Websites

WordPress.org now auto-blocks plugin updates flagged by AI review after a backdoor incident, adding a supply-chain gate for millions of sites.

Since June 5, 2026, every WordPress.org plugin and theme release passes a mandatory six-hour cooldown while multiple AI models and Jetpack Scan analyze code changes and produce a consolidated security score; releases above the risk threshold are blocked automatically. The change followed a July 28, 2026 incident where a backdoor was pushed into a plugin with roughly 20,000 active installs, which Wordfence flagged and the Plugins Team pulled 26 minutes later before distribution. Authors are notified of blocking findings and can republish corrected releases or appeal false positives to the Plugins Team.

Cyber Security Newsupdated · 6d agofirst · 6d agoTools 2 sources1