Mullvad VPN review: Near-total privacy with a few sacrifices

Mullvad, a virtual private network (VPN) named after the Swedish word for "mole," is often recognized as one of the best VPNs for privacy. I put it on my best VPN list for exactly that reaso [...]

CyberGhost VPN review: Despite its flaws, the value is hard to beat

CyberGhost is the middle child of the Kape Technologies VPN portfolio, but in quality, it's much closer to ExpressVPN than Private Internet Access. I mainly put it on my best VPN list because it& [...]

Windscribe review: Despite the annoyances, it has the right idea

Windscribe is a virtual private network (VPN) with intense "How do you do, fellow kids?" energy. It has servers in 69 countries and an annual plan that costs $69, an obsession with the sex n [...]

Anthropic says its most powerful AI cyber model is too dangerous to release publicly — so it built Project Glasswing

Anthropic on Tuesday announced Project Glasswing, a sweeping cybersecurity initiative that pairs an unreleased frontier AI model — Claude Mythos Preview — with a coalition of twelve major technolo [...]

Private Internet Access VPN review: Both more and less than a budget VPN

I came into this review thinking of Private Internet Access (PIA) as one of the better VPNs. It's in the Kape Technologies portfolio, along with the top-tier ExpressVPN and the generally reliable [...]

Norton VPN review: A VPN that fails to meet Norton's standards

One thing I need to make clear right from the start: this is a review of Norton VPN (formerly Norton Secure VPN, and briefly Norton Ultra VPN) as a standalone app, not of the VPN feature in the Norton [...]

Flaw in 17 Google Fast Pair audio devices could let hackers eavesdrop

Now would be a good time to update all your Bluetooth audio devices. On Thursday, Wired reported on a security flaw in 17 headphone and speaker models that could allow hackers to access your devices, [...]

200,000 MCP servers expose a command execution flaw that Anthropic calls a feature

Anthropic created the Model Context Protocol as the open standard for AI agent-to-tool communication. OpenAI adopted it in March 2025. Google DeepMind followed. Anthropic donated MCP to the Linux Foun [...]

Anthropic's Claude Code Security is available now after finding 500+ vulnerabilities: how security leaders should respond

Anthropic pointed its most advanced AI model, Claude Opus 4.6, at production open-source codebases and found a plethora of security holes: more than 500 high-severity vulnerabilities that had survived [...]