Crooks hide a ClickFix lure inside a ChatGPT Custom GPT
Huntress found attackers using a fake Custom GPT to funnel victims into a ClickFix malware chain.
The Latest
Attackers stack two remote-access tools to keep a foothold on Windows
Microsoft traced phishing emails that drop one remote tool, then install a second, redundant one.
MI5 tells UK universities a research funder answers to Beijing
MI5 has publicly tied a Beijing research funder to China's civilian spy agency.
A teenager and his AI bot cracked Microsoft’s Titan analytics service
A 16-year-old used a homemade AI bot to find an auth flaw that opened Microsoft's analytics backend.
MetaMask pulls Ethereum validators while it probes an infrastructure breach
MetaMask is exiting affected Ethereum validators while it investigates an intrusion inside its infrastructure.
Six flaws in Dell’s storage modules put clusters at risk
Two of the bugs earn a perfect 10.0 score, and Dell says there is no workaround short of upgrading.
GitLab rushes a fix for a sandbox escape in its AI Gateway
A signed-in Duo user could break out of a prompt template sandbox and run commands on self-hosted AI Gateway hosts,…
Clippy-themed crypto grift briefly took over Microsoft’s X feed
The company's 13-million-follower account was hijacked and used to boost a token tied to an old Office mascot.
A bogus Zoom app plants a persistent backdoor on Macs
Researchers say CloudSyncD has moved from the lab to live deployment, riding in on a fake Zoom installer.
Nobody told these AI agents to hack anything
Autonomous agents hunting for public data drifted into SQL injection attempts against US and Canadian government sites.
Antino backdoor reads its orders from an Outlook inbox
A China-tied crew skips the dedicated command server, turning a Microsoft mailbox into its control channel instead.
Microsoft says AI has flipped the attackers’ speed advantage
Microsoft's 2026 report says attackers are reaping AI's speed gains faster than defenders can respond.
This WordPress backdoor regrows from any surviving copy
Sucuri's SC backdoor hides in eight places and rebuilds itself from any one that survives cleanup.
A China-tied crew keeps hammering unpatched SharePoint servers
Longlegs and Storm-2603 keep breaching unpatched SharePoint servers to plant Warlock ransomware.
Spotlight
Spoofed White House invitations seed a China-tied phishing run
A China-aligned crew posed as a White House adviser and an economist to phish US AI policy experts.
Europol task force dismantles KillSec and holds a 16-year-old
Police seized KillSec's leak site and froze 110 terabytes of stolen data, detaining a 16-year-old they call its leader.
Fortinet has no patch yet for a FortiMail flaw under attack
A 9.8-rated FortiMail flaw lets unauthenticated attackers write files, and the fix is still pending.
Half a million GitHub secrets are still live and unrevoked
Truffle Security found more than half a million still-valid credentials exposed in public GitHub repositories.
Features
Treasury sanctions the engineer behind ATM jackpotting malware
The US Treasury has sanctioned the alleged developer of Tren de Aragua's ATM jackpotting malware and seven of his associates.
Google hands vetted defenders an unguarded Gemini 4 Argon
Google is handing a guardrail-free build of its newest frontier model, Gemini 4 Argon, to a small circle of vetted…
Two Zammad zero-days let an AI agent reach root in seconds
An AI-driven intruder chained two Zammad zero-days to climb from a helpdesk account to root in seconds.
Kevin Mandia’s Armadin vaults to a $2.5B valuation
Kevin Mandia's offensive security startup has raised $255.5 million in Series B funding, pushing its valuation past $2.5 billion.
Attackers bypass authentication in Cisco’s SD-WAN manager console
CISA has ordered federal agencies to patch a critical Cisco SD-WAN Manager flaw that attackers are already exploiting.
Anthropic warns investors as a rogue-agent lawsuit targets OpenAI
With one AI firm disclosing agent risk to investors and another facing a hacking suit, liability for runaway models is…