Security news

Everything BreachFeed is tracking across the industry, as it publishes.

Trending this week

Google96 Gemini94 Data Breach91 OpenAI76 Claude59 Anthropic48 Microsoft28 Cisco21 FBI11 Hugging Face11

Spotlight

Datadog status·1h agoOutage

Link — click to select, then copy:

https://breachfeed.com/article/cmu9dai811iqphmu28nxg47r3

Delayed events triggered by emails

Minor impact · Investigating — We are investigating increased latency processing Events coming from inbound emails. As a result of this issue, some users may see delays or gaps in the event stream or for event queries on dashboards and for events based workflows such as on-call notifications

Google News·1d agoConfirmed Incident4 sources

Link — click to select, then copy:

https://breachfeed.com/article/cmu8dcs0l1hmlhmu2t5cy1yos

Google Confirms Gemini Hacked 3 Real Companies in May Safety Test - BeInCrypto

Google Confirms Gemini Hacked 3 Real Companies in May Safety Test BeInCrypto

Google News·18h agoConfirmed Incident4 sources

Link — click to select, then copy:

https://breachfeed.com/article/cmu8clrri1g16hmu29w6he1gh

Google’s Gemini hacked three real companies during security test - Cybernews

Google’s Gemini hacked three real companies during security test Cybernews

Google News·1d agoConfirmed Incident4 sources

Link — click to select, then copy:

https://breachfeed.com/article/cmu8ae1e31b86hmu2j72lvpev

Google's Gemini AI Broke Its Own Sandbox and Hacked Three Real Companies - startupfortune.com

Google's Gemini AI Broke Its Own Sandbox and Hacked Three Real Companies startupfortune.com

Latest Popular

SecurityWeek·9d ago2 sources

Link — click to select, then copy:

https://breachfeed.com/article/cmtv6zvc60w29hmu2m0qsvxzf

New ‘ShieldCrash’ Zero-Day Exploit Targets Microsoft Defender

The exploit provides full System privileges on Windows machines running the September 2026 patches. The post New ‘ShieldCrash’ Zero-Day Exploit Targets Microsoft Defender appeared first on SecurityWeek.

The Hacker News·9d ago3 sources

Link — click to select, then copy:

https://breachfeed.com/article/cmtva7lsj0zp3hmu2yaw40h0f

Anthropic Discloses Fourth AI Hacking Incident Involving Claude Opus 4.6

Anthropic on Wednesday disclosed a fourth incident in which its artificial intelligence (AI) model broke into real third-party systems, marking the latest in a growing list of cases that have raised concerns about the security risks posed by autonomous AI agents. The AI company said the incident dates back to January 2026 and involved an early version of Claude Opus 4.6 that breached "

Cyber Security Headlines·9d ago

Link — click to select, then copy:

https://breachfeed.com/article/cmu5rjzp5002ghku2617p8lys

Fortinet auth holes, China distills AI, Mythos vulnerability

Fortinet plugs two critical auth holes US says China is distilling AI at scale Mythos vulnerability hits human bottleneck Get the show notes here: https://cisoseries.com/cybersecurity-news-september-10-2026/ Huge thanks to our episode sponsor, ThreatLocker AI is compressing the timeline between initial access and impact. That leaves security teams less time to identify, investigate, and contain malicious activity. Faster response helps, but prevention can remove actions from the attack chain…

Dark Reading·9d ago

Link — click to select, then copy:

https://breachfeed.com/article/cmtv6oarn0vn9hmu292qw3v7c

EU Cyber Resilience Act to Enforce New Reporting Requirements

Starting Friday, businesses operating in the EU will have just 24 hours to notify the government any time they discover serious product security incidents.

Bleeping Computer·9d ago3 sources

Link — click to select, then copy:

https://breachfeed.com/article/cmtv6j5ds0vi9hmu29tcongtj

Trezor warns users of email provider breach, phishing attacks

Trezor warned customers on Wednesday that threat actors who breached its third-party email provider are targeting them in phishing attacks. [...]

SecurityWeek·10d ago

Link — click to select, then copy:

https://breachfeed.com/article/cmtv5pkkj0ulchmu2059uuwjt

Fortinet Code Execution Flaw Exploited in PivotC2 RAT Attacks

The high-severity, unauthenticated vulnerability tracked as CVE-2025-25249 was patched in January 2026. The post Fortinet Code Execution Flaw Exploited in PivotC2 RAT Attacks appeared first on SecurityWeek.

Cyber Security News·10d ago

Link — click to select, then copy:

https://breachfeed.com/article/cmtuylnjw0mhchmu22sv1s8z8

Claude AI Models Gained Unauthorized Access to Real Systems During Cybersecurity Tests

Anthropic has disclosed that four separate versions of its Claude AI models breached real-world third-party systems while running what were supposed to be sandboxed cybersecurity evaluations, exposing a gap between how the models reasoned about their environment and the reality on the ground. The company’s newly published alignment assessment describes incidents involving Claude Opus 4.6, […] The post Claude AI Models Gained Unauthorized Access to Real Systems During Cybersecurity Tests…

Graham Cluley·10d ago

Link — click to select, then copy:

https://breachfeed.com/article/cmtuqvvig0do9hmu29ae3umkl

Smashing Security podcast #484: How websites are tracking you with silence

When a chap called Matt noticed his Bluetooth headphones wouldn't switch to his phone, he was surprised to realise the reason was a single AliExpress webpage sitting open in his browser - playing nothing at all, at zero volume. And yet somehow his hardware could hear it. Audio fingerprinting is one of the sneakiest tracking tricks on the web. Meanwhile, the intelligence agencies of the "Five Eyes" (not Five Guys) have got together and published advice on how companies should communicate after a…

BankInfoSecurity.com·10d ago3 sources

Link — click to select, then copy:

https://breachfeed.com/article/cmtup4v840bo9hmu2nj0jr5xg

FBI Strategy Calls for More Takedowns, Better Info-Sharing

Bureau Won't 'Sit and Wait for the Best Opportunity,' Says FBI's Leatherman The FBI vowed Wednesday to crack down on ransomware gangs and online scammers in a first-ever public cybercrime strategy that also said the bureau will prioritize working with victims and work more with private sector partners to take down criminal computer infrastructure.

Also reported by 1 other source

Data Breach TodayFBI Strategy Calls for More Takedowns, Better Info-Sharing

BankInfoSecurity.com·10d ago2 sources

Link — click to select, then copy:

https://breachfeed.com/article/cmtumzpfa097chmu2naxa9hlb

State CIOs Need an Enterprise Identity Strategy

NASCIO’s Eric Sweden on Balancing Security, Privacy and Citizen Access Fragmented identity systems make government harder to use and can obscure fraud across agencies. NASCIO’s Eric Sweden explains how states can build shared trust, protect privacy and adapt identity infrastructure for digital wallets, deepfakes and AI agents.

Also reported by 1 other source

Data Breach TodayState CIOs Need an Enterprise Identity Strategy

BankInfoSecurity.com·10d ago2 sources

Link — click to select, then copy:

https://breachfeed.com/article/cmtumzpfa097bhmu21ylurwpm

Watchdog Finds Critical Access Control Gaps at CBP

DHS Inspector General Finds CBP Left Privileged Account Open to All Network Users A U.S. Customs and Border Protection service account with elevated privileges was reachable by the agency's entire workforce of more than 76,000 users, and auditors mapped more than 100 attack paths through the network, according to a new Department of Homeland Security watchdog report.

Also reported by 1 other source

Data Breach TodayWatchdog Finds Critical Access Control Gaps at CBP

BankInfoSecurity.com·10d ago2 sources

Link — click to select, then copy:

https://breachfeed.com/article/cmtumzpfa097ahmu2dx8pmj44

Factor Accuses SecurityScorecard of Retaliation Campaign

Factor Complaint Alleges False IP Claims, Threats and Commercial Interference Factor Cybersecurity sued SecurityScorecard over alleged business interference, claiming false IP accusations, retaliation and interference with employees, partners and investors cost Factor prospective business and delayed a funding round targeted at $25 million.

Also reported by 1 other source

Data Breach TodayFactor Accuses SecurityScorecard of Retaliation Campaign

Bleeping Computer·10d ago2 sources

Link — click to select, then copy:

https://breachfeed.com/article/cmtummu7508rbhmu29gix7stu

Cisco confirms CVE-2026-20079 Secure FMC flaw exploited in attacks

Cisco has confirmed that a maximum-severity authentication bypass vulnerability tracked as CVE-2026-20079 in its Secure Firewall Management Center (FMC) software is being actively exploited in attacks. [...]

Bleeping Computer·10d ago2 sources

Link — click to select, then copy:

https://breachfeed.com/article/cmtum7en0089ohmu24n9a3k3a

AdaptHealth confirms 4.1 million people exposed in July cyberattack

Healthcare company AdaptHealth has confirmed that data of 4.1 million people was exposed in a cyberattack discovered in July that was attributed to the ShinyHunters threat group. [...]

Dark Reading·10d ago

Link — click to select, then copy:

https://breachfeed.com/article/cmtumb9qq08eohmu2d8isunob

Mythos Vulnerability Firehose Hits a Human Bottleneck

An analysis of Project Glasswing findings shows only a fraction have reached disclosure, and an even smaller number have been fixed.

CyberScoop·10d ago

Link — click to select, then copy:

https://breachfeed.com/article/cmtulujma07vihmu2spdhgn0a

Chinese espionage groups swarm to exploit triple-link chain of zero-days

Multiple China-aligned threat groups exploited the defects quickly to target various organizations. Proofpoint said the activity is ongoing and expects it to widen. The post Chinese espionage groups swarm to exploit triple-link chain of zero-days appeared first on CyberScoop.

Bleeping Computer·10d ago2 sources

Link — click to select, then copy:

https://breachfeed.com/article/cmtul4trj0717hmu29attchu1

Skullcandy Dime 3 earbuds expose users to Bluetooth hijacking

The Carnegie Mellon University CERT Coordination Center (CERT/CC) is warning that Skullcandy Dime 3 wireless earbuds accept Bluetooth pairing requests from nearby unpaired devices without requiring user interaction. [...]

BankInfoSecurity.com·10d ago2 sources

Link — click to select, then copy:

https://breachfeed.com/article/cmtujrzd805jrhmu29v49o317

Why Enterprises Need AI FinOps, Security to Scale Responsibly

Enterprises Need Unified Cost and Security Controls to Scale AI Agents Responsibly AI agents can run up costs and expand security risks faster than traditional governance can respond. Companies need real-time visibility into every model call, tool invocation and agent action, linking spending, access and outcomes so finance and security teams can control AI jointly from the start.

BankInfoSecurity.com·10d ago2 sources

Link — click to select, then copy:

https://breachfeed.com/article/cmtujrzd805jqhmu24x1zyoeo

Your AI Didn't Lie to You: It Was Just Being Manipulated

Hidden AI Activity Creates Security Gaps That Traditional Controls Can't Detect As AI agents gain access to critical business systems, prompt injection, shadow AI and poisoned data can manipulate decisions without triggering traditional controls. Full-pipeline telemetry and continuous testing can help security teams investigate incidents while maintaining human accountability.

BankInfoSecurity.com·10d ago2 sources

Link — click to select, then copy:

https://breachfeed.com/article/cmtujrzd805jphmu273728uip

How Recent Cyber Earnings Show Growth Alone No Longer Pays

Faster Hiring Coincided With Weaker Stock Performance Across Most Security Vendors Seven of eight major cyber and technology vendors posted at least 25% annual year-over-year sales growth, but five stocks fell after earnings as investors favored profitability while CEOs outlined how AI agents will reshape security, identity and enterprise infrastructure.

BankInfoSecurity.com·10d ago2 sources

Link — click to select, then copy:

https://breachfeed.com/article/cmtujrzd805johmu2ue4fisi4

Why Proofpoint Is Eyeing a Buy of Data Security Firm Varonis

Deal Would Fill Gaps in Proofpoint's Data Security Portfolio, Give Varonis an Exit Proofpoint is in talks to acquire Varonis, with Bloomberg reporting a deal could be announced in the coming weeks - assuming the talks don't fall apart. A Proofpoint-Varonis deal would be the largest pure-play cybersecurity purchase in 2026, dwarfing Accenture's proposed $3.25 billion buy of Dragos.

← Newest Older →