Urgent.News

What's breaking now, across thousands of outlets.

Friday, 16 October 2026 Reading Room

Tech

PIA’s latest update claims to triple connection speeds on Windows — here's what's new

Private Internet Access’s v3.8.0 update brings some big changes for Windows users and some useful updates for other platforms

techradar.com ·

PIA’s latest update claims to triple connection speeds on Windows — here's what's new

Private Internet Access (PIA) recently rolled out version 3.8.0 of its VPN service, touting a potential increase in Windows connection speeds by over 200 percent. The update integrates WireGuard as the default protocol across all platforms, replacing the previous OpenVPN option. According to PIA, this transition can expedite WireGuard speeds by more than three times on high-bandwidth systems.

WireGuard is recognized for being a modern, less CPU-intensive protocol compared to OpenVPN, making it a versatile VPN suitable for various devices. This shift to WireGuard as the default protocol is particularly advantageous for everyday users who utilize VPNs for video streaming or secure file transfers, as connection speed can often determine the VPN chosen.

However, OpenVPN remains an optional protocol, which is a positive for businesses and large networks that depend on its robust infrastructure. Moreover, the update brings an OpenVPN DCO kernel driver to Windows users, potentially amplifying connection bandwidth by up to 100 percent, depending on the specific system and connection specifications.

Improving speed is not the only enhancement; PIA has also incorporated the AmneziaWG protocol for Linux and macOS, enabling the obfuscation of VPN usage. This feature is beneficial for circumventing VPN restrictions in regions or environments that curtail VPN usage. To utilize AmneziaWG, users must navigate to the Protocol page, choose WireGuard, and enable Obfuscation.

Finally, the update addresses a macOS issue that caused the application to freeze when setting up a split tunneling system extension in the system settings. PIA, known for its sporadic updates and expansions in the past, appears to be ramping up its update frequency, adding 10 new server locations across the Americas, Middle East, Africa, and Australia.

These ongoing security, privacy, and feature updates position PIA as a contender for the title of one of the best and most affordable VPNs available.

Written by urgent.news from TechRadar's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Also reported by 1 other outlet

Read the original at techradar.com →

More in Tech

Hardening an AWS EKS cluster: IRSA, NetworkPolicies, restricted Pod Security and Secrets Manager via CSI

A freshly created EKS cluster works, but it's permissive. Every pod can reach every other pod, workloads can run as root, and the easy way to give an application AWS access is to attach permissions to…

  • Implement IAM Roles for Service Accounts (IRSA) for least-privilege IAM role assumption
  • Use NetworkPolicies to enforce default-deny traffic within orders namespace
  • Utilize Secrets Store CSI driver to mount AWS Secrets Manager secrets as files

·dev.to

PO Matching Looked Like It Halved Unsafe Invoice Approvals Until I Checked Why

A LangGraph agent for Japanese invoices, killed with SIGKILL, pre-registered, and gated in CI. The headline gain from its last check turned out to be mostly routing, not detection.

  • PO matching reduced unsafe invoice approvals from 17.0% to 2.0%
  • Verification caught most errors, PO matching stopped one out of six wrong PO agreements
  • LangGraph agent had three versions, final one added duplicate checks and PO matching

·dev.to

WordPress Gets Scanned Every Hour. Core Does Nothing About It.

WordPress runs a huge share of the web, and that popularity cuts both ways: it also makes WordPress the single most-scanned CMS on the internet.

  • WordPress scanned hourly for vulnerabilities
  • xZeroProtect plugin blocks malicious requests
  • Distinguishes bots from human traffic

·dev.to

Run GLM-5.3-Flash (321B) locally with Project Maya and EXL3 on one RTX 5090

GLM-5.3-Flash is a 321B mixture-of-experts model, and its 3.05 bpw EXL3 quant is 125 GB, too big to fit in my workstation's 128 GB of RAM next to the OS.

  • GLM-5.3-Flash 321B model runs locally on RTX 5090 GPU
  • Project Maya fork used for installation and setup
  • 3.05 bits per word quantization allows model to fit in RAM

·dev.to

Framejury, a culling assistant that works offline and explains every photo it rejects

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle What I Built Framejury looks through a folder of event photos, groups the frames that were shot seconds apart…

  • Framejury is an open-source offline photo culling assistant
  • Analyzes event photos to group similar shots and select best frames
  • Provides transparent explanations for each rejection decision

·dev.to

More from Friday 16 October →