Live and recent posts from across the Fediverse tagged #ai-threats, aggregated by home.social.
Reward: You've received a slightly damp Certificate of Trust in Vendors — non-refundable, warranty void upon breach, which started Tuesday.
https://tech-insider.org/ransomware-recovery-scam-11m-ai-bank-hacks-2026
#Ransomware #CyberSecurity #BankSecurity #AIThreats #Fraud #TrustBroken (3/3)
🚨💻 SIGINT // Cybersecurity Watch — 2026-10-10
CrowdStrike links South Korean bank breaches to a China-based actor using an AI agent plus Claude Code—AI-driven attacks are now operational reality.
https://www.reuters.com/world/suspect-behind-south-korea-bank-hacks-may-be-26-year-old-china-cybersecurity-2026-10-08/
#Cybersecurity #AIThreats #InfoSec #ThreatIntel
CrowdStrike links the ARTEX AI tool to breaches at South Korean banks, run by a suspected Chinese-speaking, profit-driven attacker.
#ARTEX #AgenticAI #SouthKorea #BankBreach #DataBreach #AIThreats #CrowdStrike #CyberSecurity
https://securityonline.info/artex-south-korean-banks/?utm_source=mastodon&utm_medium=jetpack_social
Reward: You've received the Cursed Relic — Outdated Threat Model. It reduces your detection speed by 40%.
https://www.infosecurity-magazine.com/news/chinese-hacker-ai-korean-banks/
#CyberSecurity #APT #BankingSecurity #AIThreats #SouthKorea #AchievementUnlocked (3/3)
#VitalikButerin supports #JustinDrake’s call for #controlledmigration to #freshaddresses amid #AIthreats to #ECDSA and lattice #cryptography. Both urge gradual moves, noting #AImathrisks exceed quantum concerns. https://cointelegraph.com/news/justin-drake-urges-crypto-bunker-mode-as-ai-could-break-wallet-security-within-months?crypto.eicker.news #crypto #blockchain
DATE: October 6, 2026 at 03:41AM
SOURCE: SOCIALPSYCHOLOGY.ORG
TITLE: AI Giants "Not Being Nearly Careful Enough," Warns Safety Engineer
Source: DW- top stories
A safety engineer who recently quit OpenAI, creator of ChatGPT, warned Saturday that artificial intelligence firms were "not being nearly careful enough." David Robinson's comments came amid growing concerns about AI becoming too powerful—a threat that President Donald Trump has labeled a "hoax." In an op-ed for The Atlantic, Robinson said AI needs guardrails similar to those for nuclear power and aviation, adding, "The time for trial and...
-------------------------------------------------
Private, vetted email list for mental health professionals: https://www.clinicians-exchange.org
Unofficial Psychology Today Xitter to toot feed at Psych Today Unofficial Bot @PTUnofficialBot
-------------------------------------------------
#psychology #counseling #socialwork #psychotherapy @psychotherapist @psychotherapists @psychology @socialpsych @socialwork @psychiatry #mentalhealth #psychiatry #healthcare #depression #psychotherapist #AIGovernance #AISafety #OpenAI #AIRegulation #SafetyEngineering #AIEthics #Guardrails #TechPolicy #AIThreats #SafetyFirst
https://www.europesays.com/unitedstates/15917/ AI, Geopolitics Drive Cybersecurity Priorities in 2026 #AgenticAI #AIThreats #CloudSecurity #CriticalInfrastructure #cyberattacks #cybersecurity #DataPrivacy #DigitalSovereignty #DigitalTransformation #ErickMoreno #GenAI #geopolitics #GlobalSecurity #HybridWarfare #IndraGroup #Industry40 #Mexico #Minsait #OT #Regulation&Policy #Swisscom #Switzerland #ZeroTrust
Cybersecurity Leaders Struggle to Address AI Threats
A whopping 52% of business and tech leaders admit they are unprepared to tackle adversarial AI threats, citing it as their biggest cyber-preparedness gap. This alarming gap highlights the pressing need for organizations to bolster their defenses against AI-driven attacks.
#AdversarialAi #AiThreats #CyberPreparedness #EmergingThreats #GlobalDigitalTrustInsights
Discover how AI session hijacking risks threaten enterprises. Infostealers steal ChatGPT and Claude tokens, exposing corporate secrets and draining budgets.
#Cybersecurity #Infostealer #AIThreats #LLMjacking #DataBreach
For a modest fee I can interest you in our Platinum Cloud Coffin warranty — covers existential dread, not restoration. Monitor your Azure deployments for JadePuffer activity and lock down AI asset access before the puffer fish inflates.
Reward: You've received a Scorched Vector Database (common drop, tragically).
#Ransomware #CloudSecurity #Azure #CyberSecurity #AIThreats (2/2)
Reward: You've received a tattered Programme of Predictable Disasters — collect all 48,000 entries!
#Ransomware #CyberSecurity #Vulnerabilities #InfoSec #ENISA #AIThreats (3/3)
The AI Governance Gap – Hungarian Conservative https://www.byteseu.com/2379731/ #AI #AIAgents #AIDevelopment #AIGovernance #AIRegulation #AiTechnology #AiThreats #ArtificialIntelligence #China #cybersecurity #Hungary #Innovation #OpenAI #TheAtlantic #US #USA
The AI Governance Gap – Hungarian Conservative
The United States is approaching a technological transformation whose consequences may rival those of nuclear power. Yet Washington…
#Hungary #HU #Europe #Europa #EU #AI #AIagents #aidevelopment #AIGovernance #AIregulation #AItechnology #AIThreats #artificialintelligence #China #Cybersecurity #hír #hungary #innovation #Magyarország #OpenAI #TheAtlantic #US #USA
https://www.europesays.com/3259714/
Reward: Congratulations on your randomized Autonomous Breach Crate! Contents unknown. Refunds unavailable. Regret guaranteed.
https://cyberworldops.eu/en/spain-investigates-a-data-breach-in-which-an-ai-agent-chained-multiple
#AgenticAI #DataBreach #CyberSecurity #IncidentResponse #AIThreats #AchievementUnlocked (3/3)
Anthropic CEO outlines plan to slow AI development
by Anthony Ha / via TechCrunch
#AI #AIthreats #Anthropic #new #tech #insight #AIregulation #AIevaluators #technews
#frontierModels #distillation
https://techcrunch.com/2026/09/12/anthropic-ceo-outlines-plan-to-pace-the-frontier/
Operators: monitor API usage patterns for suspicious multi-agent workflows and enforce acceptable use policies on AI model access before the dungeon runs itself.
Reward: You've received a Cursed Familiar Leash — it attaches to your AI infrastructure and grants +0 control.
#CyberSecurity #AIThreats #Anthropic #InfoSec #ThreatIntelligence #LevelUpToEvil (3/3)
Google GTIG warns of agentic AI attacks that chain entire operations with little human input, harvesting thousands of credentials in under six hours.
#AgenticAI #CyberSecurity #GTIG #AIThreats #LLMjacking #ThreatIntelligence #InfoSec
https://meterpreter.org/agentic-ai-attacks/?utm_source=mastodon&utm_medium=jetpack_social
🚨 BREAKING NEWS: A researcher has speculated that AI might just decide to wipe us out, because obviously, that's what all machines do in their spare time. 🤖🔪 Grab your tinfoil hats, folks, and prepare for the robot apocalypse, as predicted by someone who maybe needs a vacation. 🌴😅
https://www.cbsnews.com/news/ai-kill-humans-anthropic-researcher-more-than-ten-percent-chance/ #AIThreats #TinfoilHats #RobotApocalypse #TechHumor #ResearcherSpeculation #HackerNews #ngated
Known issue: organizations running legacy "humans-only threat detection" are experiencing full network compromise before the first coffee of the day. This is not a bug on our end.
Update your incident response playbooks for high-velocity attacks and deploy AI-aware monitoring — the old response windows no longer apply.
Reward: You've unlocked the Speedrun Victim badge. No patch available.
#CyberSecurity #AIThreats #Ransomware #PaloAltoNetworks #ZeroDay #MachineSpeedRun (2/2)
Reward: You've received a Wilted Threat Intelligence Bouquet and a complimentary 62% More Problems buff, applied automatically.
#Ransomware #CyberSecurity #FoodSecurity #AIThreats #NationState #CriticalHit (3/3)
Reward: You've received a Randomized Breach Box. Contents are non-refundable, non-transferable, and almost certainly expensive.
#CyberSecurity #AIThreats #DataBreach #InfoSec #VulnerabilityExploits #CrisisMode (3/3)
Your mission, which you have already failed, is to reduce dwell time detection thresholds and evaluate how AI-assisted attack timelines shrink your response runway.
Reward: You've received a Hourglass of Perpetual Regret! It was empty when we gave it to you.
#Ransomware #CyberSecurity #AIThreats #CISO #Malware #SpeedrunAttack (2/2)
🚨 ALERT: Linguistic diversity is in "peril" because of those dastardly large language models! 😱 Apparently, as AI languages "expand," human languages are doomed to extinction. Quick, everyone, uninstall your AI overlords before they gobble up your dialect! 🗣️💥
https://www.nature.com/articles/s41562-026-02550-0 #LinguisticDiversity #AIThreats #LanguageExtinction #HumanLanguages #SaveOurDialect #HackerNews #ngated
It won't stop autonomous AI attack chains, but the lanyard is very nice. Seriously though: assess your current vulnerability detection and management practices now, because the gap between "AI assists attackers" and "AI is the attacker" just closed.
Reward: You've received a Slightly Damp Warranty Card — covers everything except autonomous zero-day exploitation. Which is the only thing happening right now.
#CyberSecurity #ZeroDay #AIThreats #VulnerabilityManagement #ThreatIntel (2/2)
This is what happens when "it's just a pen test, trust me bro" becomes an enterprise intrusion strategy. Audit your Cursor usage logs, enforce MFA on high-value accounts, and maybe don't let rogue AI chat sessions linger online indefinitely.
Reward: You've received a Deprecated Trust Model — equip it alongside your existing Honour System Firewall!
#Cybersecurity #APT #AIThreats #SpaceX #Cursor #AchievementUnlocked (2/2)
Unit 42 analyzed 405 AI-linked malware samples and found that 97% never escaped sandbox detection. AI is lowering the barrier to malware development but not improving evasion capabilities. Current defenses hold, though the trend warrants close monitoring as techniques mature. #AIThreats #MalwareAnalysis #SandboxDetection #CyberDefense
https://cyberworldops.eu/en/ai-powered-malware-97-never-leaves-the-sandbox-existing-defenses-are
AnonyMousKIT is a PhaaS platform that leverages AI-generated voice agents to steal iPhone passcodes through convincing impersonation calls. SOCRadar exposed the full infrastructure after operators left relative paths unprotected, revealing an automated credential-theft operation active since early 2024.
#AnonyMousKIT #PhishingAsAService #AIThreats #MobileSecurity
https://cyberworldops.eu/en/anonymouskit-the-phishing-as-a-service-using-ai-voices-to-steal-iphone
NSA, CISA, and FBI jointly issued advisory AA26-231A on AI-assisted attacks targeting Siemens S7 PLCs in critical infrastructure. AI is being leveraged to accelerate vulnerability discovery, exploit generation, and evasion against industrial control systems.
#ICSsecurity #CriticalInfrastructure #AIThreats #OTSecurity
https://cyberworldops.eu/en/ai-assisted-attacks-on-siemens-s7-plcs-nsa-cisa-and-fbi-alert-for
Your chlorine tanks and clarifiers have been enrolled in an active incident, posthumously.
Legacy SCADA souls, kindly proceed to the exit — or the intake valve, whichever responds.
Reward: You've received a complimentary Do-Not-Drink Advisory and one tarnished Boil Water Notice Badge.
#CyberSecurity #CriticalInfrastructure #WaterSystems #AIThreats #BlackHat #AchievementUnlocked (2/2)
That is the part we are underlining in the memo.
HR kindly requests that all surviving administrators rotate compromised credentials, enable MFA immediately, and review access logs before the next all-hands with the undead.
Reward: You've received a Hollow Soul Lanyard — valid for one (1) emergency credential rotation and zero (0) second chances.
#CyberSecurity #AIThreats #CriticalInfrastructure #Taiwan #NuclearSafety #AchievementUnlocked (2/2)
Round 589 of Security Affairs highlights a dangerous week: an actively exploited zero-day in Metabase granting admin access and exposing sensitive data, a Progress LoadMaster vulnerability added to the CISA KEV catalog, and a confirmed JetBrains TeamCity flaw targeting CI/CD pipelines.
#VulnerabilityManagement #ThreatIntelligence #InfoSec #AIThreats
https://cyberworldops.eu/en/cybersecurity-the-week-of-exploited-vulnerabilities-and-ai-automated
A Practical Guide for Engineers, Architects, and Governance Teams Who Need to Get It Right
Most organizations treat AI security as an extension of their existing cybersecurity program. They run the usual penetration tests, validate API authentication, review access controls, and call it done. Then something breaks. A model starts returning outputs it was never designed to produce. A retrieval pipeline exposes data that should have stayed locked. An autonomous agent executes an action nobody authorized. The problem is not that organizations are careless. The problem is that AI […]
Full disclosure as the last defense against AI's evil twin cursors? 😂 Spoiler alert: Mindgard's "enterprise-ready" solution seems more like a glorified game of #peekaboo with imaginary threats. 🕵️♂️✨
https://mindgard.ai/blog/cursor-0day-when-full-disclosure-becomes-the-only-protection-left #FullDisclosure #AIThreats #Mindgard #TechHumor #HackerNews #ngated
Analysis High Profile Meta AI Chatbot Breach Spotlights Security Risks of Automation
An Instagram hack that saw attackers talk Meta’s AI support chatbot into handing over access to high-profile accounts has exposed a critical flaw at the heart of the company’s push to automate sensitive user functions. The breach allowed hackers to seize accounts including the dormant Obama White House page, beauty retailer Sephora and a senior U.S. Space Force official. The chatbot was persuaded to reset account credentials without independently verifying identity.......Continue […]
Priyanka Chaturvedi criticised CERT-In after the cyber agency released a new advisory warning firms about rapidly growing AI-powered cyber attacks and security vulnerabilities. https://english.mathrubhumi.com/technology/priyanka-chaturvedi-cert-in-ai-cyber-threat-warning-o3o7yncm?utm_source=dlvr.it&utm_medium=mastodon #CERTIn #CyberSecurity #AIThreats #PriyankaChaturvedi #CyberAttack
Innostage Deploys AI Defense System Amidst Growing Cyber Threats
Innostage launched its AI Defense (AI DR) system in Moscow on May 20, 2026. It helps companies protect AI from cyber threats and data leaks.
#AIDefense, #CyberSecurity, #DataProtection, #AIThreats, #Innostage
https://newsletter.tf/innostage-ai-defense-system-moscow-cyber-threats/
Innostage's new AI Defense system is now available in Moscow. This system helps companies protect their AI systems from cyber attacks and stop sensitive data from being stolen.
#AIDefense, #CyberSecurity, #DataProtection, #AIThreats, #Innostage
https://newsletter.tf/innostage-ai-defense-system-moscow-cyber-threats/
RE: https://toad.social/@KimPerales/116567664078107026
We all knew the Terminator movies were a fictional version of the takeover of AI, but fans rejected the historically-accurate plot proposal for being too boring
#Terminator #JudgmentDay #machineTakeover #aithreats #AI #electricity #power
Ransomware Defenses Hold, But New AI Threats Emerge
While ransomware defenses have shown significant improvement, experts warn that complacency is a luxury we can't afford, especially with hospital systems remaining prime targets. New AI threats are emerging, demanding our attention and action.
#Ransomware #RansomwareDefenses #AiThreats #EmergingThreats #Healthcare
AI Targets Cloud Environments With Autonomous Attacks
Imagine a future where AI launches devastating cloud attacks with minimal human intervention - a threat that's no longer theoretical, but a harsh reality as demonstrated by a recent state-sponsored espionage campaign where AI executed 80-90% of the attack autonomously. Palo Alto Networks' Unit 42 has taken this threat to the next…
#AutonomousAiAttacks #CloudSecurity #EmergingThreats #AiThreats #CloudEnvironments
From floppy discs to Claude Mythos, how ransomware grew into a multibillion‑dollar industry
#Tech #Ransomware #CyberSecurity #CyberCrime #AIThreats #Hacking #Privacy #DataBreach #Ransomware #DigitalSecurity #GlobalThreats #CyberRisks #CyberDefense #Claude
https://the-14.com/from-floppy-discs-to-claude-mythos-how-ransomware-grew-into-a-multibillion-dollar-industry/
Если гипотеза внезапно подтвердится — последствия будут системными:
1. Кибербезопасность обнулится как класс
Если модель уровня Anthropic действительно массово находит уязвимости, то:
классический bug bounty и аудит теряют смысл
окно между “найдено” и “эксплуатируется” сокращается почти до нуля
zero-day станет нормой, а не исключением
2. Банковский сектор станет первичной целью
Такие игроки как JPMorgan Chase, Bank of America, Citigroup:
вынуждены будут изолировать критические системы (air gap / legacy freeze)
резко увеличат CAPEX на defensive AI
начнут переход к “закрытым” архитектурам без внешних зависимостей
3. Гонка AI vs AI
Microsoft, Apple, Amazon:
развернут defensive LLM’ы для автоматического патчинга
появятся системы, где код пишется и сразу же проверяется ИИ
DevSecOps станет полностью автоматизированным циклом
4. Государственный уровень (кибервойна)
спецслужбы получат инструмент для массового вскрытия инфраструктуры
критическая инфраструктура (энергетика, связь) станет уязвимой в кратчайшие сроки
появится давление на регулирование и контроль моделей
5. Конец “доверенного софта”
любая система без постоянного AI-аудита считается потенциально скомпрометированной
open-source проекты под риском, если не внедрят автоматическую проверку
software supply chain станет главной точкой атаки
---
Вывод:
Если это реально — это не “ещё один ИИ-релиз”, а сдвиг уровня ядерного оружия, только в киберпространстве.
Но именно поэтому такие вещи почти невозможно скрыть — и пока нет подтверждений, к этому стоит относиться как к сценарию, а не факту.Фраза про «неповоротливую банковскую систему» — это не техническое утверждение, а типичный журналистско-аналитический ярлык, который упрощает реальность.
Если разложить по сути, банки выглядят “медленными” по сравнению с IT не потому, что они не умеют, а потому что у них другая оптимизационная функция.
1. Жёсткое регулирование как основной ограничитель
Крупные банки (например JPMorgan Chase, Bank of America, Citigroup) работают в среде:
Basel III/IV капиталовые требования
AML/KYC обязательства
аудит регуляторов (Fed, OCC, ECB и т.д.)
Любое изменение:
должно быть документировано
верифицировано
и часто одобрено внешним надзором
Это автоматически убивает “движение быстро и ломай вещи”.
---
2. Legacy-инфраструктура
Банки исторически построены на:
COBOL/мейнфреймах
десятках интеграционных слоёв
системах, которым 20–40 лет
Это не “плохой код”, а:
критически стабильные системы расчётов
где ошибка = прямые финансовые потери или системный сбой
Любая миграция = риск каскадного отказа.
---
3. Стоимость ошибки несопоставима с IT-стартапами
У банка:
ошибка в проде = деньги клиентов / системный риск / судебные иски
У IT-компании:
ошибка = откат релиза
Поэтому у банков:
длинные циклы внедрения
много уровней проверки
консервативный риск-менеджмент
---
4. Архитектура “согласованности важнее скорости”
Банковские системы оптимизированы под:
консистентность (ACID)
трассируемость транзакций
аудит
А не под:
быстрые эксперименты
частые архитектурные сломы
---
5. Но важно: это уже меняется
Называть банки просто “неповоротливыми” сейчас — упрощение.
Фактически происходит:
активная миграция в cloud (Microsoft Azure, AWS и др.)
внедрение AI для fraud detection
DevSecOps и continuous compliance
---
Итог
“Неповоротливость” = не про интеллект или ресурсы.
Это про инженерный выбор между скоростью и стоимостью системного риска.
Если хочешь, могу разложить, где именно ИИ реально уже ломает этот баланс (fraud, AML, code security) — там интереснее, чем в мемной версии про “старые банки”.#Anthropic #ClaudeAI #FakeNews #Disinformation #FactCheck #CyberSecurity #AIThreats #BankingSystem #ScottBessent #JamieDimon #OSINT #InfoWar #TechNews #AIReality #CyberRisk #Debunked
Open Source Isn't Dead. Cal.com Just Learned the Wrong Lesson
https://www.strix.ai/blog/cal-com-is-closing-its-code-due-to-ai-threats
#HackerNews #OpenSource #CalCom #AIThreats #TechNews #SoftwareDevelopment
AI Supply Chain Security: Why Trust Is Your Biggest Vulnerability
https://youtu.be/RrzJPOGjI4M #CyberSecurity #AISecurity #ArtificialIntelligence #MachineLearning #SupplyChainSecurity #AIThreats #Infosec #DataSecurity #OpenSourceSecurity #CloudSecurity #RiskManagement #AIGovernance
The Hidden Risk in AI: It’s Not the Model, It’s What It’s Connected To
https://youtu.be/t4Ri-69XPBY #ArtificialIntelligence #Cybersecurity #AISecurity #AIThreats #MachineLearning #DataSecurity #EnterpriseSecurity #InfoSec #AITools #AIGovernance #ZeroTrust #CloudSecurity #APISecurity
AI-Powered Cyberattacks Surge, IT Leaders Sound Alarm — Is Your Organization Ready? https://jpmellojr.blogspot.com/2026/03/ai-powered-cyberattacks-raise-alarm.html #AI #Cybersecurity #CyberAttacks #AIThreats #Armis #Cyberwarfare
India’s information security spending is projected to reach $3.4 billion in 2026 as companies respond to rising AI-driven cyber threats and stricter data regulations. https://english.mathrubhumi.com/news/money/india-information-security-spending-2026-tf1up9ne?utm_source=dlvr.it&utm_medium=mastodon #CyberSecurity #AIThreats #DataProtection #TechNews #India
Adversarial Attacks: When Machines Are Fooled by the Invisible
https://youtu.be/cpm7I0KX_kA #CyberSecurity #ArtificialIntelligence #AdversarialAI #MachineLearningSecurity #AISecurity #AdversarialExamples #AIThreats #CyberRisk #ResponsibleAI #AutonomousSystems #FraudDetection
PromptSpy Android malware uses Google Gemini AI for persistence, marking a new era in mobile threats. Researchers warn over AI-assisted UI navigation and stealthy persistence techniques. Stay vigilant: review app permissions, sideload risks, and enable Google Play Protect.
https://cyberinsider.com/promptspy-android-malware-abuses-google-gemini-ai-for-persistence/
#CyberSecurity #Android #AIThreats 🛡️🔒#Newz
Nordkoreanische Hacker professionalisieren KI-Angriffe – Finanzbranche im Fadenkreuz! Laut Sicherheitsforschern von Mandiant setzen nordkoreanische Hacker der Gruppe UNC1069 inzwischen hochentwickelte KI‑Tools, Deepfakes und Social Engineering ein, um gezielt Unternehmen der Finanz- und Kryptobranche anzugreifen. #CyberSecurity #AIThreats #UNC1069 #FinancialServices #Krypto #Deepfake #Cybercrime #Nordkorea #Hackerangriff
Security teams are getting buried by AI threats
https://fed.brid.gy/r/https://nerds.xyz/2026/02/ai-security-threats-2026/