How to check if your VPN is working

One of the disconcerting things about using a virtual private network (VPN) is that it can be hard to tell when it's doing its job. The best VPNs all work in the background to keep your IP address hidden and your communications with their servers encrypted. The better the VPN, the less you notice it, which can make a top-performing VPN feel (uncomfortably) like one that isn't working at all.

Luckily, you've got options for checking whether your VPN is working — other than just taking the app at its word. In this article, I'll cover the basics, then go through five different tests you can run to make sure you're actually using an encrypted VPN server. For each test, I'll explain what kind of problem it's looking for, how to run it and what to do in case it fails.

Before you do anything else, though, it's not a bad idea to check your VPN app and make sure you remembered to connect. It's all too easy to open up the client app, choose a server, tweak some preferences and feel like your work is done. On top of that, we don't always remember to tell VPN beginners that simply opening the client isn't enough.

To check that your VPN is turned on, open the app on your desktop or mobile home screen. Each VPN designs its apps differently, but common signs include the color green, the word Connected and information on what server location you're connected to.

The main UI for Proton VPN, with the connection button visible at top-left and the server location menu below it.
The main UI for Proton VPN, with the connection button visible at top-left and the server location menu below it.
Sam Chapman for Engadget

If you don't see anything like that, click the On button, which should be on the first page that appears when you log into the app. Most VPNs also connect whenever you click the name of a server location.

For those of you on iPhone or iPad, I've just written an explainer on how to turn a VPN off and on. For all the tests I'll discuss across the rest of this article, make sure you're connected to a VPN server before you run them. Also, make sure your internet connection is active — a VPN can only work when there's internet.

Each of these tests investigates a different reason your VPN might not be working. We'll start by looking for connection problems that might not be obvious, check for DNS leaks, WebRTC leaks and IPv6 leaks, then finally make sure an apparently active VPN is managing to change your virtual location.

Websites and internet service providers (ISPs) use IP addresses to identify devices and their owners online. A VPN's most important job is to change your IP address to one matching its own server, which disassociates your identity from your online activities. Not doing this indicates a failure on a fundamental level: either the VPN says it's connected when it isn't, or its technology is active but somehow not sending you through the proper encrypted tunnel.

To check whether your VPN has changed your IP address, start by going to an IP address checker like whatismyipaddress.com or ipleak.net. This will show you the public IP address that everyone sees when you get online without a VPN, including the ISP that holds it and the geographic location it's associated with. Write that down or take a screenshot.

A censored report from WhatIsMyIPAddress.com.
A censored report from WhatIsMyIPAddress.com.
Sam Chapman for Engadget

Next, connect to your VPN. Remember the location you connect to, and note down the new server IP address if the VPN tells you what it is. Go back to your IP tester tool and refresh the page. You should now see an IP address and location that match the one you connected to through the VPN, including a different ISP.

If your IP address is the same as before, your VPN isn't working. To fix this, try disconnecting from the server, waiting about 10 seconds, then connecting to the same location and trying the test again. This will show you whether the problem was with one individual server or an entire location.

If the problem persists, try a different server location, then a different VPN protocol. If it's still leaking, try restarting your VPN client, your device and your modem (in that order). This should fix the problem, but if it doesn't, move on to the remaining tests or get in touch with the VPN's tech support.

A domain name system (DNS) server is an important step in getting a website to appear on your browser. DNS holds the information that connects URLs to the IP addresses of destination servers. If a VPN client lets your device contact a DNS server owned by your internet service provider without routing it through an encrypted tunnel first, the DNS request might reveal your real IP address to the ISP.

You can check for DNS leaks by connecting to your VPN, then going to dnsleaktest.com or another tool of your choice. The tester sends several innocuous DNS requests, then scans to see which servers resolve them. If you see your real ISP at all, you've got DNS leaks.

A DNS leak test run without a VPN. With one active, my real ISP (Comcast) should not appear on the list.
A DNS leak test run without a VPN. With one active, my real ISP (Comcast) should not appear on the list.
Sam Chapman for Engadget

The fix for DNS leaks is more intensive than the fixes in step #1. Check your VPN's control panel to activate any DNS leak protections and try again. IPv6 leaks can also appear as DNS leaks, so try disabling IPv6 in your browser (see #4 below for instructions). If you keep seeing leaks, you can also try clearing your computer's DNS cache.

Here's how to do that. On Windows, go to the Command Prompt (on Windows 10) or the Windows Terminal (on Windows 11). Enter the phrase ipconfig/flushdns. On Mac, open Terminal from the Utilities folder, then paste in the phrase sudo dscacheutil -flushcache; sudo killall -HUP mDNSResponder and hit Enter. Test the VPN once more to see if it's still leaking.

WebRTC, which stands for Web Real-Time Communication, is a technology that lets browsers exchange information directly with each other. This is useful for text and video chats, streaming and more, but it's also a potential security risk. WebRTC can serve as a backchannel that inadvertently sends your real IP address outside the VPN tunnel.

It's pretty easy to test for WebRTC leaks. I recommend the tool ipleak.net, which checks for them as a matter of course. You can also use browserleaks.com/webrtc to run a test that's particular to this kind of leak. These tools establish dummy connections through WebRTC, then test to see if the VPN still works when they're active. As usual, if you see your real IP address, there's a problem.

Your WebRTC IP not matching your Remote IP is a potential red flag.
Your WebRTC IP not matching your Remote IP is a potential red flag.
Sam Chapman for Engadget

The fixes for a WebRTC leak are the usual ones: try different servers, locations and protocols, reset your VPN, device and modem, then try another VPN provider. However, if nothing is working, you can also disable WebRTC on your browser altogether. This means you won't be able to do any real-time chatting (that's Zoom, Google Meet, Teams and so on), so it's a last-resort solution.

To disable WebRTC on Firefox, type about:config in the URL bar, click the message to accept the risk, type media.peerconnection.enabled in the search bar, then double-click the word True to change it to False. To turn WebRTC back on, just double-click False again.

On Edge, you can disable WebRTC by entering edge://flags in the URL bar, scrolling down to the option "Anonymize local IPs exposed by WebRTC" and making sure the dropdown next to it is set to Enable. There's no built-in way to turn off WebRTC on Chrome, but you can install the WebRTC Control extension to switch it off and on yourself.

Next up, it's possible that your real location is leaking through your IPv6 address, not IPv4. To make a long explanation short, IPv6 is a new way of formatting IP addresses that leaves more options available for the future. Since we haven't yet hit the crisis point of IPv4 shortage, very few websites are restricted to IPv6 alone.

The problem is that most VPN apps were designed in the IPv4 era and aren't built to protect IPv6 traffic. There are some exceptions, including NordVPN, but most VPNs block IPv6 traffic completely rather than retrofit themselves to work with it. However, if a VPN of that sort isn't blocking IPv6 entirely, your IPv6 address and associated location can leak.

Any IP address checker can reveal an IPv6 leak, but you can find a specific test at test-ipv6.com. This site runs several exams that look for IPv6 readiness, but the most important line is the one that shows your current IPv6 address. This will probably say you don't have one, since most ISPs don't work through IPv6 yet — but if you do have one, it should match your active VPN's location, not your real one.

If your IPv4 address matches the VPN server but your IPv6 address does not, IPv6 is the likely cause of your leak.
If your IPv4 address matches the VPN server but your IPv6 address does not, IPv6 is the likely cause of your leak.
Sam Chapman for Engadget

Should it turn out that you're leaking IPv6 requests, the easiest solution is to disable IPv6 on your computer. On Windows, you can do this through the network adapter options page of your control panel. Here's how to get there:

  • Windows 10: Start -> Settings -> Network & internet -> Status -> Change -> Advanced network settings -> Change adapter options.

  • Windows 11: Settings app -> Network & internet -> Advanced network settings -> Related settings -> More network adapter options.

On both OSes, finish the job by right-clicking the name of your internet connection, selecting Properties from the dropdown and unchecking the box next to Internet Protocol Version 6. Of course, you can always switch to another VPN that blocks IPv6 altogether, but you might find that to be a bigger hassle.

If you're on Mac, open System Settings, click the Network tab and then click the Details... button next to your network name. In the new window, click the TCP/IP tab on the left, find the entry labeled Configure IPv6 and set the dropdown to Link-Local Only.

A VPN can be working perfectly and still fail to unblock streaming sites. Netflix, HBO Max and the others block VPN traffic because VPNs can make them show material in regions where they don't hold the copyright. To avoid legal trouble, they set up their firewalls to block IP addresses known to belong to VPN servers.

If your VPN can't get into a streaming platform, it'll usually be obvious; the site will either display a proxy error message or simply refuse to load. However, in rare cases, the streaming site will load fine but show you the same shows you normally see. This indicates that you might be dealing with a VPN leak.

If that happens, follow the usual steps. Disconnect and reconnect to the same location to get a different server, then try different server locations. It's also possible that the streaming site is getting your real location from your browser cache, so if the problem persists, clear your cache and cookies and try again.

There's one more important step to make sure your VPN is working: test the kill switch. This common feature cuts off your internet connection if you lose touch with your VPN server. With your kill switch active, you shouldn't be at any risk of accidentally broadcasting your real IP address, location or online activity.

To test your kill switch, you'll need to simulate an abrupt loss of VPN connectivity. Open your VPN, make sure the kill switch is turned on, then connect to a server. Next, quit the VPN app without disconnecting. At this point, the kill switch should make it impossible for you to get online — if you can still browse the internet as normal, the switch might be faulty.

This article originally appeared on Engadget at https://www.engadget.com/cybersecurity/vpn/how-to-check-if-your-vpn-is-working-130000817.html?src=rss

Netflix mobile app redesign will offer ‘deeper integration’ of vertical video

After testing the feature for the last six months or so, Netflix said it will bring vertical videos to its mobile app sometime later in 2026, TechCrunch reported. So far, the feature has been used to show clips promoting Netflix films and TV shows, letting you swipe to move to the next one à la TikTok or YouTube Shorts.

Going forward, though, it could also serve as a promotion tool for new types of programs. “You can imagine us bringing more clips based on new content types, like video podcasts,” co-CEO Greg Peters said in an earnings call.

That was another subject of discussion, as Netflix is taking on YouTube with new video podcasts. The company debuted the first of those originals last week, including shows from sports commentator Michael Irvin and comedian Peter Davidson. It’s also offering hosting content like the Bill Simmons Podcast and other shows from The Ringer.

Despite increasing competition from YouTube and other social media networks, Netflix is doing all right. In 2025, the company saw $45.2 billion revenue including $1.5 billion in ad dollars from its lower tier subscriptions. The streamer counted 325 million paid subscribers around the world at the end of last year.

This article originally appeared on Engadget at https://www.engadget.com/entertainment/streaming/netflix-mobile-app-redesign-will-offer-deeper-integration-of-vertical-video-120000820.html?src=rss

The FTC isn’t giving up on its antitrust case against Meta

The Federal Trade Commission lost its antitrust case against Meta last year, but the regulator hasn't given up on its attempts to punish the social media company for its acquisitions of WhatsApp and Instagram. The FTC is appealing a ruling last year in which a federal judge found that the government hadn't proven that Meta is currently operating as a monopoly. 

"Meta has maintained its dominant position and record profits for well over a decade not through legitimate competition, but by buying its most significant competitive threats," the FTC's Bureau of Competition Director Daniel Guarnera said in a statement. "The Trump-Vance FTC will continue fighting its historic case against Meta to ensure that competition can thrive across the country to the benefit of all Americans and U.S. businesses.”

The FTC originally filed antitrust charges against Facebook in 2020 during President Donald Trump's first term in office. The government argued that by acquiring apps it once competed with, Instagram and WhatsApp, the company had depressed competition in the space and ultimately hurt consumers. A trial last year saw testimony from several current and former executives, including CEO Mark Zuckerberg and former COO Sheryl Sandberg, who spoke at length about the pressure to compete with TikTok. 

US District Judge James Boasberg was ultimately persuaded by Meta’s arguments, writing that the success of YouTube and TikTok prevented Meta from currently "holding a monopoly" even if the company had acted monopolistically in the past. If the FTC had won, it could have tried to force Meta to undo its acquisitions of WhatsApp and Instagram. Should it be successful in its appeal, that remedy could once again be on the table.

News of the FTC's plan to appeal is also a blow to Zuckerberg, who has spent the last year courting Trump and hyping Meta's plans to spend hundreds of billions of dollars on AI infrastructure in the United States. In a statement, Meta spokesperson Andy Stone said that the original ruling was "correct," and that "Meta will remain focused on innovating and investing in America."

This article originally appeared on Engadget at https://www.engadget.com/big-tech/the-ftc-isnt-giving-up-on-its-antitrust-case-against-meta-225020769.html?src=rss

OpenAI is launching age prediction for ChatGPT accounts

OpenAI is the latest company to hop on the bandwagon of gating access by users' age. The AI business is beginning a global rollout of an age prediction tool to determine whether or not a user is a minor. “The model looks at a combination of behavioral and account-level signals, including how long an account has existed, typical times of day when someone is active, usage patterns over time,and a user’s stated age,” the company’s announcement states. If an individual is incorrectly characterized by ChatGPT as underage, they will need to submit a selfie to correct the mistake through the Persona age verification platform. 

Most AI companies have been willing to push new features first and then attempt to layer on a patchwork of protections and safety guards on top of them after they cause harm. OpenAI was implicated in a wrongful death suit for a teen who allegedly used ChatGPT to plan his suicide, and only in the following months began pondering automatic restrictions on content for underage users and launching a mental health advisory council. In this instance, OpenAI is attempting to prepare for the launch of an "adult mode" that will allow users to create and consume content that would be dubbed NSFW. Considering how well a similar change has been going over at Roblox, another platform with a shaky history around protecting minors, it seems probable that underage users will find ways to circumvent the existing tools if they want to use ChatGPT as adults.

This article originally appeared on Engadget at https://www.engadget.com/ai/openai-is-launching-age-prediction-for-chatgpt-accounts-222650340.html?src=rss

Akai’s MPC XL groovebox is the most powerful device the company has ever made

Akai just revealed its most powerful standalone groovebox, the MPC XL. For the uninitiated, the MPC has been around since the 1980s and pretty much defined hip hop from that era. The line has continuously iterated to keep up with the times and the XL looks to be the baddest of them all.

First of all, it offers four times the processing power of previous MPCs, which is enough to load up to 32 virtual instruments at the same time. This is assisted by a full 16GB of RAM, which is a whole lot in this era of AI tomfoolery. The XL can handle 16 audio tracks simultaneously. In my experience with previous units, this is more than enough for a full song.

It runs on a proprietary OS and features a 10-inch OLED touchscreen for making adjustments. There are also dozens of knobs and buttons to play with, including 16 knobs that integrate with the display for real-time feedback.

The XL features a step sequencer, but this is an MPC. The real star attraction are those 16 drum pads. These pads can be set to trigger samples and hits, but can also be programmed to initiate effects and do all kinds of other stuff. Each pad has four quadrants, one for each corner, and they are all fully adjustable.

Looking for even more nuanced control? There are two assignable touch-strips and plenty of short-cut keys. It has built-in microphone preamps, phono inputs for sampling, instrument inputs and numerous other connectivity options. This is a true flagship in every sense of the word. It's also very, very large.

The MPC may have started as a hip hop machine, but newer models are useful for any genre of music. To that end, it comes with an extensive collection of plugins, samples and effects. These even include some plugins made by rival company Native Instruments.

The software can handle stuff like stem separation, time-stretching and more. The workflow has been heavily inspired by modern DAWs, with a full arrangement view available on that OLED.

The MPC XL is a standalone unit, so it doesn't need a computer or anything like that. The power requirements here, however, don't allow for batteries. This thing has to be plugged in, much like Native Instruments' Maschine+. It's available right now and costs a whopping $2,900. This is a serious machine with a serious price tag, just like Roland's recently-released TR-1000 drum machine.

This article originally appeared on Engadget at https://www.engadget.com/audio/akais-mpc-xl-groovebox-is-the-most-powerful-device-the-company-has-ever-made-183952483.html?src=rss

Earth is having some issues, so let’s enjoy the Webb telescope’s new nebula image

Sometimes, you just need to give your mind a little vacation. And these days, outer space sounds like as good a destination as any. Thankfully, the James Webb Space Telescope (JWST) is here to give us a dazzling new image of the Helix Nebula.

Discovered in the early 19th century, the Helix Nebula resides in the constellation Aquarius. (Cue The 5th Dimension.) At about 655 light-years away, it's one of Earth's closest planetary nebulae. When zoomed farther out, it's easy to see why it's been nicknamed the Eye of God or Eye of Sauron. This 2004 image from the Hubble telescope illustrates that.

An eye-like nebula in space
A wider view of the same nebula from 2004
NASA / ESA / C.R. O'Dell (Vanderbilt University) / M. Meixner / P. McCullough / G. Bacon ( STSI)

What we're seeing in the nebula is, in a sense, a moment of death that lays the groundwork for a new birth. The dying star (out of frame in the closer new image) sheds its outer layers. As expelled gas and dust cool, they provide raw material that could someday form new stars and perhaps planetary systems.

The new image from Webb's NIRCam (Near-Infrared Camera) provides a much closer, higher-resolution view.

Pillar-like reddish knots in the Helix Nebula
Pillar-like reddish knots in the Helix Nebula

Those pillars you see are called cometary knots, and this image is our best view of those to date. "Here, blistering winds of hot gas from the dying star are crashing into colder shells of dust and gas that were shed earlier in its life, sculpting the nebula's remarkable structure," the ESA wrote in its press release.

The knots' colors represent temperature and chemistry. Hints of blue indicate the hottest gas (energized by ultraviolet light). The yellow regions, where hydrogen atoms form molecules, are farther from the nebula's nucleus (and therefore cooler). On the edges, reddish-orange regions depict the coolest material, where gas thins and dust begins to form.

This article originally appeared on Engadget at https://www.engadget.com/science/space/earth-is-having-some-issues-so-lets-enjoy-the-webb-telescopes-new-nebula-image-181049901.html?src=rss

Roland’s Go:Mixer Studio is an affordable but capable mixer for budding recording engineers

Roland just unveiled the Go:Mixer Studio, a powerful entry in the company's line of audio interfaces. This one promises to be a portable and affordable way to create high-quality recordings with a smartphone or PC.

The biggest news here are the 12 input channels and six output channels. This means that users can record multiple instruments at once and even run the signal through outboard gear if so desired. There are two XLR inputs on the back with phantom power, an instrument jack, stereo ¼-inch line-in jacks, an aux input, TRS MIDI ins and outs, two headphones jacks and stereo line outs for connecting monitor speakers.

It records up to 24-bit/192kHz audio, which will certainly get the job done. The device comes with built-in effects for mangling audio on the fly. Each audio input also has a dedicated EQ and compressor directly on the channel.

It connects via an app that's available for iOS, Windows and MacOS. The iOS app actually lets users capture both video and multitrack audio at the same time, which should be a boon for streamers and content creators of all stripes. The Windows and MacOS apps allow for complete control of the mixer remotely.

As for the unit itself, there are several big knobs to adjust parameters and a color display. It's also really small and light, making it easy to pack away in case of an unexpected recording session. The device can even be mounted on a mic stand.

The Roland Go: Mixer Studio is available right now and costs $300. The Roland Go line of audio products has been around for years, but this one seems like a major step up.

This article originally appeared on Engadget at https://www.engadget.com/audio/rolands-gomixer-studio-is-an-affordable-but-capable-mixer-for-budding-recording-engineers-163927262.html?src=rss

Meta’s Oversight Board is looking into transparency around disabling accounts

Meta has a lot of work to do when it comes to limiting hate speech on its platforms. Now, its Oversight Board is looking into the company's decision to ban an account for, among other things, posting visual violent threats and harassment against a journalist — and it wants the public's advice. 

In the year prior to the ban, Meta referred five posts due to violations of its hateful conduct, bullying and harassment, violence and incitement and adult nudity and sexual activity community standards. In addition to the posts harassing the woman journalist, the user also shared "anti-gay slurs against prominent politicians and content depicting a sex act, alleging misconduct against minorities." 

Meta's internal review experts decided to permanently disable the account due to the consistent violations and calls for violence. This action was taken despite the number of strikes not reaching the ban threshold — Meta's guidance states that even seven strikes only get users a one-day ban. However, its account integrity page lays out examples of when it will disable accounts, including violating its community standards through "risk of imminent harm" to an individual. 

The Board is now looking for insight from the public until 11:59PM PT on Tuesday, February 3. The Board is specifically seeking comments from individuals who can "contribute valuable perspectives" on the following topics: 

  • How best to ensure due process and fairness to people whose accounts are penalized or permanently disabled.

  • The effectiveness of measures used by social media platforms to protect public figures and journalists from accounts engaged in repeated abuse and threats of violence, in particular against women in the public eye.

  • Challenges in identifying and considering off-platform context when assessing threats against public figures and journalists.

  • Research into the efficacy of punitive measures to shape online behaviors, and the efficacy of alternative or complementary interventions.

  • Good industry practices in transparency reporting on account enforcement decisions and related appeals.

This instance marks the first time the Board has looked into Meta permanently disabling an account. It stated that this "represents a significant opportunity to provide users with greater transparency on Meta’s account enforcement policies and practices, make recommendations for improvement, and expand the types of cases the Board can review."

This article originally appeared on Engadget at https://www.engadget.com/big-tech/metas-oversight-board-is-looking-into-transparency-around-disabling-accounts-145757717.html?src=rss

Meta’s Oversight Board is looking into transparency around disabling accounts

Meta has a lot of work to do when it comes to limiting hate speech on its platforms. Now, its Oversight Board is looking into the company's decision to ban an account for, among other things, posting visual violent threats and harassment against a journalist — and it wants the public's advice. 

In the year prior to the ban, Meta referred five posts due to violations of its hateful conduct, bullying and harassment, violence and incitement and adult nudity and sexual activity community standards. In addition to the posts harassing the woman journalist, the user also shared "anti-gay slurs against prominent politicians and content depicting a sex act, alleging misconduct against minorities." 

Meta's internal review experts decided to permanently disable the account due to the consistent violations and calls for violence. This action was taken despite the number of strikes not reaching the ban threshold — Meta's guidance states that even seven strikes only get users a one-day ban. However, its account integrity page lays out examples of when it will disable accounts, including violating its community standards through "risk of imminent harm" to an individual. 

The Board is now looking for insight from the public until 11:59PM PT on Tuesday, February 3. The Board is specifically seeking comments from individuals who can "contribute valuable perspectives" on the following topics: 

  • How best to ensure due process and fairness to people whose accounts are penalized or permanently disabled.

  • The effectiveness of measures used by social media platforms to protect public figures and journalists from accounts engaged in repeated abuse and threats of violence, in particular against women in the public eye.

  • Challenges in identifying and considering off-platform context when assessing threats against public figures and journalists.

  • Research into the efficacy of punitive measures to shape online behaviors, and the efficacy of alternative or complementary interventions.

  • Good industry practices in transparency reporting on account enforcement decisions and related appeals.

This instance marks the first time the Board has looked into Meta permanently disabling an account. It stated that this "represents a significant opportunity to provide users with greater transparency on Meta’s account enforcement policies and practices, make recommendations for improvement, and expand the types of cases the Board can review."

This article originally appeared on Engadget at https://www.engadget.com/big-tech/metas-oversight-board-is-looking-into-transparency-around-disabling-accounts-145757717.html?src=rss

Adobe unveils new AI-powered video editing tools for Premiere

Adobe has announced updates for Premiere and After Effects, including new AI-powered tools that are meant to speed up your video editing tasks. In Premiere, the company’s video-editing software, it has unveiled a new AI-powered Object Mask feature that lets you easily pick and track persons or objects moving through your video clips. You simply have to hover over that object and click to generate a mask overlay in seconds. While the mask is supposed to be accurate from the start, you can adjust and resize it as needed. Adobe says the feature uses its own AI model for the feature and that the processing happens on-device. It also says that it doesn’t use your activities and data to train its models.

The company has also given its Shape Mask tool an upgrade. You can generate its redesigned Ellipse, Rectangle and Pen masks directly from the toolbar. Further, it updated their controls to make moving or adjusting the masks more precise. The masks can now also track objects on your video clips 20 times faster than their predecessors, which means you won’t have to keep such a close eye on the status bar. Another new Premiere update lets you easily bring media from from Firefly Boards, Adobe’s AI-powered digital canvas, into the program. In addition, Adobe Stock is now fully integrated within Premiere.

For After Effects, Adobe has rolled out an update that lets you import SVG files that are commonly used in Illustrator. You can now also build graphics and photorealistic objects inside images with 3D parametric meshes, consisting of cubes, spheres, cylinders, cones, toris and planes, within the visual effects software.

This article originally appeared on Engadget at https://www.engadget.com/ai/adobe-unveils-new-ai-powered-video-editing-tools-for-premiere-140000970.html?src=rss