Security researchers have published a working exploit for a pre-auth remote code execution flaw in AnyDesk on Linux that gives attackers root access before anyone approves the connection. If you still run AnyDesk on Linux, you should check for version 8.0.3 or later and treat any older install as high risk until patched or removed. …
Read more “AnyDesk exploit, TP-Link lawsuits and hidden AI agents”
AWS engineers have shown that its own AgentCore AI agents can be tricked into handing over credentials just by asking, helped by tokens in metadata, weak VM isolation, and broad permissions. If you are testing AI agents in your estate, keep them away from production secrets and audit what cloud roles and metadata they can …
Read more “AWS AI agents, OneDrive cuts and Oracle lock-in risks”
Apple is set to unveil its long rumoured HomePad smart hub at the “Welcome home” event next Tuesday. If you support Apple users at home or in small offices, expect staff to ask about a HomePod with a display and start planning how it fits into your network and device standards. The latest 9to5Mac Overtime …
Read more “Apple HomePad event, AI acqui-hire and notebook price warning”
Supply chain attacks kept the spotlight this week, from GitHub workflows to AI agents and iOS exploits. There is a clear pattern of attackers going after the tools we trust most. Malicious GitHub Actions stealing credentials Researchers reported that compromised maintainer accounts were used to push a credential stealing GitHub Actions workflow into over 340 …
Read more “GitHub workflows, iOS exploits and supply chain threats”
The NCSC has urged organisations to promptly patch vulnerabilities in Citrix NetScaler ADC and Citrix NetScaler Gateway, warning that attackers could exploit unpatched systems. If you have any Citrix kit in your stack, you should be double checking versions, vendor comms, and your MSP tickets so school services are not left exposed. London Fire Brigade …
Read more “UK schools tech watch: security and assessment shifts”
AI safety firm Anthropic has launched a programme to help defend critical infrastructure and open source projects from AI driven attacks, and warned that for the next two years attackers will have the advantage as models get more capable. For a small IT team, this is a cue to treat AI as an amplifier for …
Read more “AI risk warnings, state hackers, and fragile clouds”
Apple security is in the spotlight again as researchers detail a new P7 DarkSword spyware variant hitting unpatched iPhones. If you look after iOS devices, this is a clear nudge to keep update compliance tight and to review which users are slow to move off older builds. Apple has pushed a fresh AT&T carrier bundle …
Read more “Apple security warning for iPhone and new tools for Mac teams”
Attackers in Japan have abused mobile app APIs and hit known flaws in tools like Metabase, leading to a sharp rise in web data leaks reported to JPCERT/CC. This is a reminder to treat mobile backends and third party web apps as exposed services that need the same patching and access control as anything public …
Read more “Cyber attacks reuse tools, expose data, and turn on each other”
AI in industry is moving from simple predictive tools to what MIT Technology Review calls a new phase of foundation models, physical AI, and agentic AI that can act on real equipment. That matters for small IT teams because any move toward autonomous control in plants or warehouses will pull you into new risk, safety, …
Read more “AI in factories, robots on video, and lab risks to watch”
Degree apprenticeships heat up and pressure IT Government data quoted in FE News shows there are now 11 applicants for every degree apprenticeship place in England, up from 3 just three years ago. If your trust supports apprenticeship routes, expect more demand for data on places, outcomes, and links to local employers, and plan how …