News
Every CyberSIXT summary, newest first, with a link to its story and to the original report.
Most articles on Wed 21 Oct (96); 9–13 on weekend days.
Show as table
| Day | Articles published |
|---|---|
| Sun 11 Oct | 9 |
| Mon 12 Oct | 56 |
| Tue 13 Oct | 85 |
| Wed 14 Oct | 86 |
| Thu 15 Oct | 78 |
| Fri 16 Oct | 61 |
| Sat 17 Oct | 10 |
| Sun 18 Oct | 10 |
| Mon 19 Oct | 60 |
| Tue 20 Oct | 86 |
| Wed 21 Oct | 96 |
| Thu 22 Oct | 84 |
| Fri 23 Oct | 68 |
| Sat 24 Oct | 13 |
| Today, Sun 25 Oct | 12 so far |
Mon 20 Apr· 46 articles
-
Vercel breach after OAuth token leak in Context AI tool
Vercel disclosed that it was breached through a third‑party AI tool, Context[.]ai, after a compromised OAuth token belonging to a Vercel employee who signed up for Context’s AI Office Suite using their Vercel Google Workspace account granted broad…
Dark Reading -
ANTS ID breach may expose millions of French users' data
France’s ANTS ID System website was hit by a cyberattack that authorities say may have exposed personal data from users applying for IDs, passports, residence permits, and driver’s licences. The incident was detected on 15 April 2026, with the Interior…
Security Affairs -
Critical RCE flaws hit Lantronix, Silex serial IP converters
Researchers have identified 20 new vulnerabilities in popular models of serial-to-IP converters, devices that sit at the heart of modern industrial networks, with thousands of known vulnerabilities found in the same devices’ software stacks. The study,…
-
Scattered Spider2 articles this day · SecurityWeek first
Scottish Hacker Admits $8M Crypto Theft After Phishing Scheme · UK hacker admits $8M crypto theft via SMS phishing scheme
2 outlets-
Scottish Hacker Admits $8M Crypto Theft After Phishing Scheme
Tyler Buchanan, a 24-year-old from Scotland linked to the Scattered Spider group, pleaded guilty in the United States to hacking dozens of companies and stealing millions in cryptocurrency. The case, described in a DoJ press release, states that he admitted…
Security Affairs -
UK hacker admits $8M crypto theft via SMS phishing scheme
A British national, Tyler Robert Buchanan, 24, of Dundee, Scotland, has pleaded guilty in a US court to conspiracy to hack into dozens of companies and to stealing cryptocurrency, according to the Department of Justice. He admitted to conducting SMS phishing…
SecurityWeek
-
-
Critical CVE-2026-5760 Bug in SGLang Enables Code Execution
A critical vulnerability in SGLang, tracked as CVE-2026-5760 with a CVSS of 9.8, could allow remote code execution on affected systems. According to CERT/CC, the flaw resides in the reranking endpoint /v1/rerank, where an attacker can load a specially…
-
Rapid7’s Glasswing turns quick vuln finds into faster defence
According to Rapid7, Project Glasswing is about turning faster vulnerability discovery into faster action for defenders, emphasising that AI will be most useful when used with intention rather than simply speeding up backlogs. The piece notes that the Common…
Rapid7 -
ZionSiphon malware hits water plants via USB and Modbus
A newly identified malware strain designed to interact with operational technology (OT) systems has been analysed by security researchers, revealing capabilities aimed at water treatment and desalination infrastructure. The malware, named ZionSiphon and…
Infosecurity -
Making opportunistic cyberattacks harder by design
Microsoft’s security approach to defending against opportunistic cyberattacks centres on credential elimination and platform engineering to make attacks harder by design. The guidance emphasises removing secrets from systems, using managed identities and…
Microsoft -
BRIDGE:BREAK flaws hit Lantronix, Silex serial converters
According to SecurityWeek, Forescout researchers found 20 new vulnerabilities across Lantronix and Silex serial-to-IP converters, collectively tracked as BRIDGE:BREAK, which can be exploited for OS command injection, remote code execution, firmware…
SecurityWeek -
Phishing attacks use stealthy tricks to spread Formbook malware
Two phishing campaigns targeting organisations running Microsoft Windows use different stealthy infection techniques to deliver Formbook, a long‑running infostealer, with campaigns observed in Greece, Spain, Slovenia, Bosnia, Croatia and various South…
Infosecurity -
WhatsApp metadata leak exposes user habits, researcher says
WhatsApp leaks user metadata to attackers by revealing device fingerprints and timing information that can help an observer infer a user’s online habits, such as whether they are online or active, even if the content remains encrypted. The piece centres on…
Dark Reading -
Big Tech urged to act as older adults lose billions to scams
The article argues that Big Tech could do more to stop scams, highlighting how older adults are often blamed for fallings for fraud rather than addressing the incentives of the platforms themselves. It cites data suggesting that in 2024, older adults in the…
Malwarebytes -
Anthropic’s AI Mythos can chain cyber exploits on its own
Mythos: An AI tool too powerful for public release discusses Anthropic’s most capable model to date, Claude Mythos Preview (aka Mythos), described as a “step change” in AI performance, particularly for cybersecurity tasks. The article notes Mythos was kept…
Malwarebytes -
ShinyHunters2 articles this day · SecurityWeek first
Attackers breached Vercel via compromised Context[.]ai OAuth app · Vercel breach traced to compromised AI tool, $2M data sale
2 outlets-
Attackers breached Vercel via compromised Context[.]ai OAuth app
On 19 April 2026, Vercel disclosed a security incident after a threat actor publicly claimed to be selling stolen corporate data on the Dark Web. According to BreachForums, the post, attributed to the group ShinyHunters, offered access keys, source code, and…
SOCRadar -
Vercel breach traced to compromised AI tool, $2M data sale
Vercel confirmed on Sunday that it has suffered an intrusion after a hacker offering to sell data allegedly stolen from the company’s systems appeared on BreachForums, with the sale pegged at $2 million. The hacker, using the moniker ShinyHunters, claimed to…
SecurityWeek
-
-
CVE-2023-33538 under attack for a year, but exploitation still unsuccessful
Hackers have targeted CVE-2023-33538, a command-injection flaw in several old TP-Link routers, for over a year, but no successful exploitation has been observed so far. The vulnerability resides in the /userRpm/WlanNetworkRpm endpoint, where the ssid1…
-
QEMU Abused to Hide Ransomware While Android Banking Trojans Surge
The Hacker News’ weekly recap highlights a pattern of trusted paths being exploited, with Vercel’s Context[.]ai breach cited as an example where an employee’s Google Workspace access enabled exposure of some internal environments and environment variables.…
The Hacker News -
Minidoka Hospital Hit by Cyber Attack, Threatens 576GB Data Leak
Minidoka Memorial Hospital in Rupert, Idaho, said a cyber incident on Easter morning, 5 April 2026, limited imaging services and led to some emergency patient transfers, though the hospital and its clinics continued treating patients. In an April 17 social…
DataBreaches -
Mirai based Nexcorium botnet exploits TBK DVR flaw CVE-2024-3721
A newly identified malware campaign has been observed exploiting a command injection flaw in digital video recorder (DVR) devices to deploy a Mirai-based botnet, according to analysis by FortiGuard Labs. The activity targets CVE-2024-3721 in TBK DVR systems,…
-
Cloudflare rolls out AI code review for massive merge workflow
According to The Cloudflare Blog, Cloudflare has built a CI-native orchestration system around OpenCode to run AI-assisted code reviews at scale, with up to seven specialised reviewers handling security, performance, code quality, documentation, release…
Cloudflare -
Cloudflare’s R&D Engineers Widely Adopt AI Coding Assistants
According to Cloudflare, in the last 30 days 93% of its R&D organisation used AI coding tools powered by infrastructure built on its own platform. Eleven months ago the team began integrating AI into the engineering stack, forming a tiger team called iMARS…
Cloudflare -
Cloudflare Unveils Agentic Cloud Tools After Agents Week 2026
Today marks the end of Cloudflare’s first Agents Week 2026, an event described as an innovation week focused on the age of agents, with the aim of scaling and enabling many parallel agents. The company outlined a full slate of new offerings across four…
Cloudflare -
Be Prime hit by cyberattack exposing 12.6 GB of client data
Be Prime, a Mexico-based cybersecurity company that provides connectivity and security services to clients including Iberdrola, ArcelorMittal, Whirlpool and Alsea, was reportedly the victim of a cyberattack that allegedly resulted in the leak of 12.6 GB of…
DataBreaches -
Gentlemen RaaS adds ESXi locker, infects 320+ victims worldwide
The DFIR Report from Check Point Research outlines how The Gentlemen, a Go‑based ransomware‑as‑a‑service, has attracted affiliates with multi‑platform lockers for Windows, Linux, NAS, BSD, and an ESXi variant, and has publicly claimed over 320 victims with…
Check PointThe Gentlemen -
Hackers Abuse QEMU for Defense Evasion
According to Sophos, hackers have been abusing the QEMU machine emulator in at least two campaigns to deploy ransomware and remote access tools. The first campaign, observed in November 2025 and tracked as STAC4713, is potentially linked to the PayoutsKing…
-
AI Demos Wow, But Real Deployments Stall Over Data and Governance
According to The Hacker News, the fastest way to fall in love with an AI tool is the demo, where prompts land cleanly and outputs appear in seconds, giving a strong impression of a new era for teams. Yet the piece argues that most AI initiatives stall not…
The Hacker News -
Bluesky Suffers Sophisticated DDoS Attack Lasting About 24 Hours
Bluesky was disrupted by a distributed denial-of-service attack described as sophisticated, causing intermittent outages across its platform. The DDoS attack appears to have begun late on 15 April (Pacific Time) and continued into 16 April, lasting roughly…
SecurityWeek313 Team -
Anthropic MCP Design Vulnerability Enables RCE, Threatening AI Supply Chain
Anthropic’s Model Context Protocol (MCP) has been identified as having a design weakness that could allow remote code execution and impact the AI supply chain, according to The Hacker News. Researchers from OX Security describe the flaw as baked into…
-
Senate renews Section 70
According to Associated Press, the Senate approved a short-term renewal until 30 April of a controversial surveillance programme used by U.S. spy agencies. The measure was cleared by voice vote in the Senate, the same method used hours earlier in the House,…
SecurityWeek -
SOCRadar spots Vercel key sale, Binance data leak, FALKONc2 RAT
published on 20 April 2026, SOCRadar’s Dark Web Team identified multiple new underground posts, including an alleged Vercel access key and source code sale framed as a supply chain risk, and a separate listing claiming a 1.5 million record Binance dataset.…
SOCRadar -
Vercel breach linked to compromised AI tool Context[.]ai
Vercel has disclosed a security breach triggered by the compromise of a third‑party AI tool, Context[.]ai, used by one of its employees. The attacker took over the employee’s Google Workspace account and used it to access parts of Vercel’s internal systems,…
Security Affairs -
Frontier AI models ignite zero day hunt threatening open source
Unit 42’s hands-on with frontier AI models suggests a major shift in how quickly and at what scale vulnerabilities can be identified, moving beyond simple coding assistance to autonomous security research. The researchers warn of autonomous zero-day…
unit42.paloaltonetworks.com -
Millions of Exposed FTP Servers Still Lack Encryption, Researchers Warn
Approximately 6 million internet-accessible systems are using FTP today, and almost half of them do not use encryption, according to Censys. The number of hosts running an internet-facing FTP service has fallen by about 40% since 2024, from 10.1 million to…
SecurityWeek -
9.8 Billion Credentials Leaked from Exposed Elasticsearch Servers
Public Elasticsearch servers exposed a staggering 9.8 billion credential records across enterprise, cloud, and AI platforms, according to SOCRadar. In total, three publicly accessible Elasticsearch instances contained 9,879,060,029 records, with server 1…
SOCRadar -
NCSC rolls out NHS cyber resilience plan with new tools
The National Cyber Security Centre (NCSC) has outlined a coordinated plan to boost cyber resilience across the NHS, detailing how government and industry have deepened collaboration over the past 18 months to reduce risk and improve detection. The plan rests…
Infosecurity -
Securelist 404 shows HoneyMyte, Evasive Panda, Cloud Atlas update
The page you reached on Securelist returns a 404 error, stating that the requested page cannot be found. It nevertheless surfaces a snapshot of recent reports and links to the latest posts, including items about HoneyMyte, Cloud Atlas and Evasive Panda among…
securelist.comMUSTANG PANDA -
FakeWallet crypto stealer spreads via iOS App Store phishing apps
According to Securelist, FakeWallet’s crypto stealer has been spreading through iOS apps in the App Store, with the campaign disclosed on 20 April 2026. In March 2026, more than twenty phishing apps masquerading as popular wallets were identified, with 26…
securelist.com -
Kyrgyz Exchange Grinex Blames Foreign Hackers for £1B Theft
A Kyrgyzstan-based exchange, Grinex, said it had been subjected to a “large-scale cyber-attack” by “foreign” intelligence agencies, claiming the raid led to the theft of one billion rubles from Russian customers. The company, described as the successor to…
Infosecurity -
AI generated Chrome exploit built for just £2,300 raises alarm
According to Security Affairs, Claude Opus managed to produce a functional Chrome exploit for just $2,283, illustrating how widely available AI models can already find and weaponise vulnerabilities. The article breaks down the experiment cost, showing $2,014…
Security Affairs
