Cybersecurity doesn't have to be complicated. I am here to break down concepts, threats, tips, and tools in simple language. For mere mortals and developers alike.
Since I have been asked this a lot: The best way to support this site, is to buy my books.
Ring’s new TAKE encryption shortens how long the company keeps video keys while preserving cloud features. Here is what that protects, what it leaves unchanged, and which setting suits your home.
Meta has agreed to time limits, overnight blocks, and quieter school hours for teen accounts. The useful protections come with an age-assurance system that families should inspect just as carefully.
A shopping page ran an inaudible audio test that helped describe a visitor’s device. Here is what the finding proves, what it does not prove, and how to make browser tracking less reliable.
Nottinghamshire Police has begun scanning faces in public against a watchlist. Here is what the system does, what deletion cannot undo, and which records the public should demand before the next deployment.
Apple sent mercenary-spyware warnings across 110 countries, and investigators saw a record response. Here is how to verify the alert, reduce exposure, protect other people, and get expert help without destroying useful evidence.
Flock Safety has promised shorter default storage and stronger checks on police searches. The useful parts deserve credit, but your privacy still depends on local rules, warrants, sharing limits, and public oversight.
Signal's new Automatic Key Verification checks whether an encrypted conversation received the expected public key. Here is what the green check proves, where its limits sit, and when you should still compare safety numbers.
UC San Diego researchers found that a dealer-installed alarm system uses the same Bluetooth key across every vehicle, and that key can be pulled from the official app. Anyone in range can open the doors or immobilise the car. Roughly half of owners never asked for the device.
A flaw in the Adobe Acrobat Chrome extension let a malicious webpage reach into WhatsApp Web and pull out messages and contacts. No malware, no stolen password. Just an extension with more reach than anyone realised.
OpenAI launched Health in ChatGPT with a promise that connected medical data would not train models or target ads. That promise is real and narrow. Here are the eight other questions that decide whether sensitive data in an AI product is actually protected.
Every Windows installation since Vista registers a unique 64-bit identifier with Microsoft on first connection. It sits in plain text in the registry, travels with Store and telemetry traffic, and has been used to link devices to people in court.
Jscrambler found European and US financial institutions sending customer data to ad platforms through tracking pixels, including unencrypted names, tax IDs, and loan details, sometimes in direct violation of the consent the user had given.