Week in review: Cisco patches exploited email gateway 0-day, Revolut breach

Last week’s security news highlighted major breaches, active exploits, and AI-related risks, including incidents at Revolut, CenterPoint Energy, HBO Max, and the first data breach blamed on an autonomous AI agent in Spain. It also covered critical vulnerabilities in Cisco Secure Email Gateway, Cisco ISE, Acronis, Parallels Desktop, and AI coding agents, alongside new guidance and tools from CISA, ENISA, NIST, Google, and Microsoft. #Revolut #CenterPointEnergy #HBOMax #CiscoSecureEmailGateway #CiscoISE #Acronis #ParallelsDesktop #ENISA #CISA #NIST #Google #Microsoft

Read More
Researchers escape OpenAI Codex sandbox to run commands on host

Researchers uncovered two sandbox escapes in OpenAI Codex, including Heapjack and Overpatch, that could let attacker-controlled code run commands outside Codex’s intended restrictions. OpenAI patched both issues quickly, and users are advised to update to Codex Desktop build 26.818.21641 or Codex CLI 0.149.0 and later. #OpenAICodex #Heapjack #Overpatch #CodexDesktop #CodexCLI

Read More
Malicious npm packages evade install-script defenses at runtime

The indexed-btree npm campaign hides malicious code in normal runtime behavior to bypass new GitHub and npm supply chain defenses, while impersonating the legitimate sorted-btree library. Checkmarx found the package and nine related npm packages, which exposed victims to data theft, C2 activity through Slack, Telegram, and an Ethereum smart contract, and possible trace removal by the attackers. #indexed-btree #sorted-btree #Checkmarx #GitHub #npm #Sepolia

Read More
The Defender 2026 by Top Cyber News Magazine

The author expresses gratitude for being included in The Defender 2026 globally represented list of CISOs and thanks Top Cyber News MAGAZINE and Ludmila Morozova-Buss, Ph.D. The post also congratulates the other recognized cybersecurity leaders and acknowledges the many professionals who are strengthening the industry. #TheDefender2026 #TopCyberNewsMAGAZINE #LudmilaMorozovaBuss #KerissaVarma #GaryHayslip #KunalSehgal #IvaTasheva #FedericoCasano #IsabelMariaGomez #RiverNygryn #VivianLyon #LydieNgoNogol #CarinaBirarda #YohannBauzil

Read More
North Korean hackers infect thousands of devices across 100 countries as part of ‘WaterPlum’ campaign

North Korean hackers tied to the WaterPlum campaign stole more than $10.5 million by posing as AI and blockchain recruiters to target job seekers and infect their devices. The operation overlapped with North Korean IT worker schemes and used malware such as BeaverTail, InvisibleFerret, OtterCookie, OtterCandy, and StoatWaffle to steal cryptocurrency…

Read More
Ransom! Siddhi Green Excellence Pvt. Ltd (SEP-2026)

Cassias MG Government in Brazil was targeted by threat actor emperador using ransomware against government credentials and network access, later involving justice panels and sectors such as finance, police, and sensitive records. The data exposure included personal identity and civil documents (e.g., RG/CIN, birth certificates, CPF), along with medical information from Brazil’s SUS, threatening operational continuity under a negotiation deadline! #Brazil

Read More
Ransom! Siddhi Green Excellence Pvt. Ltd (SEP-2026)

Fanatics, a global sports commerce/e-commerce platform in the United States, reported that threat actor N0n gained destructive control over its cloud data estate, initiating deletion of 46,902 order files (108 GB) containing customer personal data, accounts-payable invoices, customer balances, bank transaction archives, tax exemption certificates, and fraud-prevention datasets. The incident is active with a deadline of 2026-09-23 01:01 UTC, threatening availability and exposure of sports commerce information. #UnitedStates

Read More
Cybersecurity News | Daily Recap [19 Sep 2026]

Daily Recap, TigerByte Cyber emerged from stealth with $3 million in funding, while multiple nation-state and AI-driven advances underscored how fast attackers are operationalizing new techniques. North Korea’s WaterPlum campaign reportedly infected 30,000 devices and used fake job-seeker lures, and research showed AI-assisted exploitation could chain weaknesses to gain access to OpenAI staff accounts. #TigerByteCyber #WaterPlum #TransparentTribe #RustBackdoor #SolarWinds #ARMHardCodedKey #WordPressClick2Shell #Gyazo #Click2Shell #ClaudeOpus5 #Rapuncel #LastPassAuthenticator #ScatteredSpider #AhmedElbadawy #OpenAI

Read More