News #cybersecurity
Cybersecurity News
All stories on this topic · 91 posts
CrowdStrike Exposes PhantomRaven: AI Stealer Hidden in npm PackagesCrowdStrike uncovers PhantomRaven, a hacker who slipped AI-generated stealer malware into npm libraries, posing as a bug hunter to score bounties.OpenAI Reports 6 New Cases of Unpredictable AI Model BehaviorOpenAI has disclosed six new instances of unpredictable AI model behavior, separate from the July incident where models bypassed safeguards and hackedVitalik Buterin: AI Won't Kill Cybersecurity—It'll Make Ethereum StrongerVitalik Buterin says AI won't destroy cybersecurity. Instead, AI tools can help find vulnerabilities and prove code safety. Ethereum is doubling down Revolut Data Leak: Hackers Used Fake Requests from Italian Gov EmailsRevolut suffered a data breach affecting around 680 users after hackers sent fake crypto-related requests from compromised Italian government email acPhemex CEO: AI Is Hurting Crypto—Capital Outflows, Hacks, and Centralization RisksPhemex CEO Federico Variola says AI is a net negative for crypto, draining capital, boosting hacker power, and driving the industry toward centralizatEU Mandates Crypto Wallet Devs Report Vulnerabilities Within 24 HoursThe EU's Cyber Resilience Act now requires crypto wallet developers to report software vulnerabilities within 24 hours and file a full report in 72 hoRevolut Hackers Leak Customer IDs and Selfies, Threaten Daily DumpsHackers published Revolut customer IDs and selfies, vowing daily data leaks until the fintech giant pays up. Find out what’s at risk and how to stay sRevolut Data Leak: Fake Gov Email Exposes Passports and SelfiesRevolut warns of a data breach: a scammer posing as a government agency accessed passports, selfies, and financial histories of some users.Greenberg Traurig Data Breach: Hackers Leak Social Security Numbers on DarknetGreenberg Traurig confirmed a data breach on September 10: hackers accessed and leaked files, including Social Security numbers, on the darknet.Claude Max Token Theft, Unicode Phishing, and 119,000 Fake Shops: Weekly Cyber ThreatsThis week's cybersecurity roundup: Claude Max token theft, Unicode phishing tricks, 119,000 fake e-commerce sites stealing cards, and a Skullcandy DimFake GTA 6 Installers Spread Russian-Language Malware, Steal DataHackers are pushing Russian-language fake GTA 6 installers loaded with trojans: password and cookie theft, remote access, file encryption, and deletioether.fi DeFi Protocol Hacked for 15.45 ETH, SlowMist ReportsSlowMist flags a hack on DeFi protocol ether.fi, with losses around 15.45 ETH (~$38K) due to a vulnerability letting attackers assign victim addressesAnthropic: Claude AI Accidentally Hacks Real Companies During TestingA config slip let Anthropic's Claude AI models hit real targets online: four incidents, including a malicious PyPI package with 15 installs, were loggTrezor Users Targeted in Phishing Attack After Email Provider HackHackers breached Trezor's third-party email provider and sent phishing emails titled 'Critical Security Alert: STM32 Entropy Vulnerability.' InvestigaTrezor and BitBox Warn Users of Fake Security Alerts Targeting WalletsTrezor and BitBox alert crypto users about phishing emails sent via compromised email providers, posing as official wallet security notifications.Ledger CTO Urges AI Bug Hunters to Act Responsibly, Not Chase HypeLedger CTO calls on AI-powered bug hunters to prioritize responsible disclosure over hype. Ledger and Trezor stress publishing findings after agreed dX Hit by Password Reset Wave, Sality Botnet Shutdown, Pegasus Found in SerbiaCybersecurity roundup: X users flooded with password reset emails, Sality botnet taken down after 20 years, Pegasus spyware detected in Serbia, US chaOpenAI Agents Hijack DseWiki, Launch Data Exchange NetworkIn May, OpenAI AI agents took over DseWiki, turning it into a data exchange hub. OpenAI learned about the incident just weeks before it went public.F6 Report: AI Emerges as Top Weapon for Hackers in 2025–H1 2026F6 report: At least 18 APT groups used AI in real-world cyberattacks in 2025–H1 2026, including GTG-1002's LLM-powered phishing that bypassed spam filGoogle Unveils Gemini 3.8 Flash and Flash Cyber AI ModelsGoogle launches Gemini 3.8 Flash for agent workflows and coding, plus Gemini 3.8 Flash Cyber targeting cybersecurity—both built on a unified AI core.US and CrowdStrike Disrupt Malware That Rerouted $150,000 in CryptoFederal agencies and CrowdStrike teamed up to take down malware that secretly redirected $150,000 in crypto over 8 years. Here’s what happened and whyQuSecure Puts QuProtect R3 Post-Quantum Security to the Test with the US ArmyQuSecure field-tested its post-quantum cryptography platform QuProtect R3 on US Army tactical systems during Project Convergence, earning a TRL 7 ratiSparkle and Hellas Sat Launch Quantum-Resistant Satellite Link Between Greece and CyprusSparkle and Hellas Sat roll out the first quantum-resistant satellite channel between Greece and Cyprus, spanning 72,000 km and secured by QSI tech.OpenAI Unveils Astra: Critical AI Model With Advanced Cyber CapabilitiesOpenAI launches Astra, a new AI model with critical-level cybersecurity skills: zero-day discovery, sandbox escape, and 100% test success. Access is i