Extra-sensory perception The nine-armed octopus and the oddities of the cephalopod nervous system A mix of autonomous and top-down control manage the octopus's limbs. Kenna Hughes-Castleberry – Jun 7, 2025 8:00 am | 19 Credit: Nikos Stavrinidis / 500px Credit: Nikos Stavrinidis / 500px Story text Size Small Standard Large Width * Standard Wide Links Standard Orange * Subscribers only   Learn more With their quick-change camouflage and high level of intelligence, it’s not surprising that the public and scientific experts alike are fascinated by octopuses. Their abilities to recognize faces, solve puzzles, and learn behaviors from other octopuses make these animals a captivating study. To perform these processes and others, like crawling or exploring, octopuses rely on their complex nervous system, one that has become a focus for neuroscientists. With about 500 million neurons—around the same number as dogs—octopuses’ nervous systems are the most complex of any invertebrate. But, unlike vertebrate organisms, the octopus’s nervous system is also decentralized, with around 350 million neurons, or 66 percent of it, located in its eight arms. “This means each arm is capable of independently processing sensory input, initiating movement, and even executing complex behaviors—without direct instructions from the brain,” explains Galit Pelled, a professor of Mechanical Engineering, Radiology, and Neuroscience at Michigan State University who studies octopus neuroscience. “In essence, the arms have their own ‘mini-brains.’” A decentralized nervous system is one factor that helps octopuses adapt to changes, such as injury or predation, as seen in the case of an Octopus vulgaris, or common octopus, that was observed with nine arms by researchers at the ECOBAR lab at the Institute of Marine Research in Spain between 2021 and 2022. By studying outliers like this cephalopod, researchers can gain insight into how the animal’s detailed scaffolding of nerves changes and regrows over time, uncovering more about how octopuses have evolved over millennia in our oceans. Brains, brains, and more brains Because each arm of an octopus contains its own bundle of neurons, the limbs can operate semi-independently from the central brain, enabling faster responses since signals don’t always need to travel back and forth between the brain and the arms. In fact, Pelled and her team recently discovered that “neural signals recorded in the octopus arm can predict movement type within 100 milliseconds of stimulation, without central brain involvement.” She notes that “that level of localized autonomy is unprecedented in vertebrate systems.” Though each limb moves on its own, the movements of the octopus’s body are smooth and conducted with a coordinated elegance that allows the animal to exhibit some of the broadest range of behaviors, adapting on the fly to changes in its surroundings. “That means the octopus can react quickly to its environment, especially when exploring, hunting, or defending itself,” Pelled says. “For example, one arm can grab food while another is feeling around a rock, without needing permission from the brain. This setup also makes the octopus more resilient. If one arm is injured, the others still work just fine. And because so much decision-making happens at the arms, the central brain is freed up to focus on the bigger picture—like navigating or learning new tasks.” As if each limb weren’t already buzzing with neural activity, things get even more intricate when researchers zoom in further—to the nerves within each individual sucker, a ring of muscular tissue, which octopuses use to sense and taste their surroundings. “There is a sucker ganglion, or nerve center, located in the stalk of every sucker. For some species of octopuses, that’s over a thousand ganglia,” says Cassady Olson, a graduate student at the University of Chicago who works with Cliff Ragsdale, a leading expert in octopus neuroscience. Given that each sucker has its own nerve centers—connected by a long axial nerve cord running down the limb—and each arm has hundreds of suckers, things get complicated very quickly, as researchers have historically struggled to study this peripheral nervous system, as it’s called, within the octopus’s body. “The large size of the brain makes it both really exciting to study and really challenging,” says Z. Yan Wang, an assistant professor of biology and psychology at the University of Washington. “Many of the tools available for neuroscience have to be adjusted or customized specifically for octopuses and other cephalopods because of their unique body plans.” While each limb acts independently, signals are transmitted back to the octopus’s central nervous system. The octopus’ brain sits between its eyes at the front of its mantle, or head, couched between its two optic lobes, large bean-shaped neural organs that help octopuses see the world around them. These optic lobes are just two of the over 30 lobes experts study within the animal’s centralized brain, as each lobe helps the octopus process its environment. This elaborate neural architecture is critical given the octopus’s dual role in the ecosystem as both predator and prey. Without natural defenses like a hard shell, octopuses have evolved a highly adaptable nervous system that allows them to rapidly process information and adjust as needed, helping their chances of survival. Some similarities remain While the octopus’s decentralized nervous system makes it a unique evolutionary example, it does have some structures similar to or analogous to the human nervous system. “The octopus has a central brain mass located between its eyes, and an axial nerve cord running down each arm (similar to a spinal cord),” says Wang. “The octopus has many sensory systems that we are familiar with, such as vision, touch (somatosensation), chemosensation, and gravity sensing.” Neuroscientists have homed in on these similarities to understand how these structures may have evolved across the different branches in the tree of life. As the most recent common ancestor for humans and octopuses lived around 750 million years ago, experts believe that many similarities, from similar camera-like eyes to maps of neural activities, evolved separately in a process known as convergent evolution. While these similarities shed light on evolution's independent paths, they also offer valuable insights for fields like soft robotics and regenerative medicine. Occasionally, unique individuals—like an octopus with an unexpected number of limbs—can provide even deeper clues into how this remarkable nervous system functions and adapts. Nine arms, no problem In 2021, researchers from the Institute of Marine Research in Spain used an underwater camera to follow a male Octopus vulgaris, or common octopus. On its left side, three arms were intact, while the others were reduced to uneven, stumpy lengths, sharply bitten off at varying points. Although the researchers didn’t witness the injury itself, they observed that the front right arm—known as R1—was regenerating unusually, splitting into two separate limbs and giving the octopus a total of nine arms. “In this individual, we believe this condition was a result of abnormal regeneration [a genetic mutation] after an encounter with a predator,” explains Sam Soule, one of the researchers and the first author on the corresponding paper recently published in Animals. The researchers named the octopus Salvador due to its bifurcated arm coiling up on itself like the two upturned ends of Salvador Dali’s moustache. For two years, the team studied the cephalopod’s behavior and found that it used its bifurcated arm less when doing “riskier” movements such as exploring or grabbing food, which would force the animal to stretch its arm out and expose it to further injury. “One of the conclusions of our research is that the octopus likely retains a long-term memory of the original injury, as it tends to use the bifurcated arms for less risky tasks compared to the others,” elaborates Jorge Hernández Urcera, a lead author of the study. “This idea of lasting memory brought to mind Dalí’s famous painting The Persistence of Memory, which ultimately became the title of the paper we published on monitoring this particular octopus.” While the octopus acted more protective of its extra limb, its nervous system had adapted to using the extra appendage, as the octopus was observed, after some time recovering from its injuries, using its ninth arm for probing its environment. “That nine-armed octopus is a perfect example of just how adaptable these animals are,” Pelled adds. “Most animals would struggle with an unusual body part, but not the octopus. In this case, the octopus had a bifurcated (split) arm and still used it effectively, just like any other arm. That tells us the nervous system didn’t treat it as a mistake—it figured out how to make it work.” Kenna Hughes-Castleberry is the science communicator at JILA (a joint physics research institute between the National Institute of Standards and Technology and the University of Colorado Boulder) and a freelance science journalist. Her main writing focuses are quantum physics, quantum technology, deep technology, social media, and the diversity of people in these fields, particularly women and people from minority ethnic and racial groups. Follow her on LinkedIn or visit her website. 19 Comments

Editor’s note: This blog, originally published on October 28, 2024, has been updated. Financial losses from worldwide credit card transaction fraud are projected to reach more than $403 billion over the next decade. The new NVIDIA AI Blueprint for financial fraud detection can help combat this burgeoning epidemic — using accelerated data processing and advanced algorithms to improve AI’s ability to detect and prevent credit card transaction fraud. Launched this week at the Money20/20 financial services conference, the blueprint provides a reference example for financial institutions to identify subtle patterns and anomalies in transaction data based on user behavior to improve accuracy and reduce false positives compared with traditional methods. It shows developers how to build a financial fraud detection workflow by providing reference code, deployment tools and a reference architecture. Companies can streamline the migration of their fraud detection workflows from traditional compute to accelerated compute using the NVIDIA AI Enterprise software platform and NVIDIA accelerated computing. The NVIDIA AI Blueprint is available for customers to run on Amazon Web Services, with availability coming soon on Dell Technologies and Hewlett Packard Enterprise. Customers can also use the blueprint through service offerings from NVIDIA partners including Cloudera, EXL, Infosys and SHI International. Businesses embracing comprehensive machine learning (ML) tools and strategies can observe up to an estimated 40% improvement in fraud detection accuracy, boosting their ability to identify and stop fraudsters faster and mitigate harm. As such, leading financial organizations like American Express and Capital One have been using AI to build proprietary solutions that mitigate fraud and enhance customer protection. The new AI Blueprint accelerates model training and inference, and demonstrates how these components can be wrapped into a single, easy-to-use software offering, powered by NVIDIA AI. Currently optimized for credit card transaction fraud, the blueprint could be adapted for use cases such as new account fraud, account takeover and money laundering. Using Accelerated Computing and Graph Neural Networks for Fraud Detection Traditional data science pipelines lack the compute acceleration to handle the massive data volumes required for effective fraud detection. ML models like XGBoost are effective for detecting anomalies in individual transactions but fall short when fraud involves complex networks of linked accounts and devices. Helping address these gaps, NVIDIA RAPIDS — part of the NVIDIA CUDA-X collection of microservices, libraries, tools and technologies — enables payment companies to speed up data processing and transform raw data into powerful features at scale. These companies can fuel their AI models and integrate them with graph neural networks (GNNs) to uncover hidden, large-scale fraud patterns by analyzing relationships across different transactions, users and devices. The use of gradient-boosted decision trees — a type of ML algorithm — tapping into libraries such as XGBoost, has long been the standard for fraud detection. The new AI Blueprint for financial fraud detection enhances the XGBoost ML model with NVIDIA CUDA-X Data Science libraries including GNNs to generate embeddings that can be used as additional features to help reduce false positives. The GNN embeddings are fed into XGBoost to create and train a model that can then be orchestrated. In addition, NVIDIA Dynamo-Triton, formerly NVIDIA Triton Inference Server, boosts real-time inferencing while optimizing AI model throughput, latency and utilization. NVIDIA CUDA-X Data Science and Dynamo-Triton are included with NVIDIA AI Enterprise. Leading Financial Services Organizations Adopt AI During a time when many large North American financial institutions are reporting online or mobile fraud losses continue to increase, AI is helping to combat this trend. American Express, which began using AI to fight fraud in 2010, leverages fraud detection algorithms to monitor all customer transactions globally in real time, generating fraud decisions in just milliseconds. Using a combination of advanced algorithms, one of which tapped into the NVIDIA AI platform, American Express enhanced model accuracy, advancing the company’s ability to better fight fraud. European digital bank bunq uses generative AI and large language models to help detect fraud and money laundering. Its AI-powered transaction-monitoring system achieved nearly 100x faster model training speeds with NVIDIA accelerated computing. BNY announced in March 2024 that it became the first major bank to deploy an NVIDIA DGX SuperPOD with DGX H100 systems, which will help build solutions that support fraud detection and other use cases. And now, systems integrators, software vendors and cloud service providers can integrate the new NVIDIA blueprint for fraud detection to boost their financial services applications and help keep customers’ money, identities and digital accounts safe. Explore the NVIDIA AI Blueprint for financial fraud detection and read this NVIDIA Technical Blog on supercharging fraud detection with GNNs. Learn more about AI for fraud detection by visiting the AI Summit at Money20/20, running this week in Amsterdam. See notice regarding software product information.

Astronomers have spotted something strange and spectacular: a mysterious object that keeps emitting pulses every 44 minutes.In a press release from Australia's Curtin University, which was part of the international team that detected the object just 15,000 light-years away in our Milky Way galaxy, astronomers explained that the find was all the more stunning because the signal is coming in the form of both X-rays and radio waves.The object, which was named ASKAP J1832-0911 after Australia's ASKAP radio telescope that was used to detect it, was discovered emitting two-minute-long pulses that would pause and then repeat 44 minutes later. As the Curtin press release explains, the researchers lucked out when they realized that NASA’s Chandra X-ray Observatory was observing the same part of the sky and detected the same repeating signal in X-ray form.This dual-natured pulse belongs to a newly-discovered class of space phenomena known as "long-period radio transients," or LPTs for short.Discovered in 2022 by the International Centre for Radio Astronomy Research — which also sponsored this latest study — these mystery pulses have unknown origins and occur in fixed intervals of minutes or hours. They're considered by astronomers to be remarkably slow as compared to the signals emitted by pulsars, those rapidly-rotating stars that send out similar bursts every few milliseconds when their poles point in Earth's direction.In the years since they were first discovered, astronomers around the world have only detected some 10 other LPTs — but before now, none have been run through X-ray telescopes as well.According to Ziteng "Andy" Wang, an ICRAR-affiliated Curtin astronomer and the lead author a paper about the finding that was just published in the journal Nature, discovering the dual nature of LPTs in such a coincidental manner "felt like finding a needle in a haystack.""The ASKAP radio telescope has a wide field view of the night sky, while Chandra observes only a fraction of it," Wang explained in the Curtin press release. "So, it was fortunate that Chandra observed the same area of the night sky at the same time."Because LPTs are such a new phenomenon to astronomers, they can't say for sure what causes them.When the first of them were discovered, astronomers posited that they could be coming from magnetars, a type of neutron star with extremely strong magnetic fields that also emit radio pulses at faster intervals, leading to the ICRAR team positing that they may have an "ultra-long-period magnetar" on their hands.While the magnetar theory appears to have been scrapped, the astronomers behind this update in LPT knowledge are hopeful that it will help them figure out what these strange, slow pulses are about."This object is unlike anything we have seen before," said Wang.Share This Article

A team of Purdue University students recently set a new Guinness World Record with their custom robot that solved a Rubik’s Cube in just 0.103 seconds. That was about a third of the time it took the previous record-setting bot. But the new record wasn’t achieved by simply building a robot that moves faster. The students used a combination of high-speed but low-res camera systems, a cube customized for improved strength, and a special solving technique popular among human speed cubers.The Rubik’s Cube-solving robot arms race kicked off in 2014, when a robot called Cubestormer 3 built with Lego Mindstorms parts and a Samsung Galaxy S4 solved the iconic puzzle in 3.253 seconds — faster than any human or robot could at the time. (The current world record for a human solving a Rubik’s Cube belongs to Xuanyi Geng, who did it in just 3.05 seconds.) Over the course of a decade, engineers managed to reduce that record to just hundreds of milliseconds.Last May, engineers at Mitsubishi Electric in Japan claimed the world record with a robot that solved a cube in 0.305 seconds. The record stood for almost a year before the team from Purdue’s Elmore Family School of Electrical and Computer Engineering — Junpei Ota, Aden Hurd, Matthew Patrohay, and Alex Berta — shattered it. Their robot has come to be known as Purdubik’s Cube. Bringing the robot record down to less than half a second required moving away from Lego and, instead, using optimized components like industrial motors. Getting it down to just 0.103 seconds, however, required the team from Purdue to find multiple new ways to shave off milliseconds.“Each robot that previous world record-holders has done has kind of focused on one new thing,” Patrohay tells The Verge. When MIT grad students broke the record in 2018, they opted for industrial hardware that outperformed what previous record-holders had used. Mitsubishi Electric chose electric motors that were better suited for the specific task of spinning each side of the cube, instead of just hardware that moved faster.However, the first thing the Purdue students improved was actually the speed that their robot could visualize the scrambled cube. Human speed cubing competitors are allowed to study a Rubik’s Cube before their timer starts, but the robot record includes the time it takes it to determine the location of all the colored squares. The students used a pair of high-speed machine vision cameras from Flir, with a resolution of just 720x540 pixels, pointed at opposing corners of the cube. Each camera can see three sides simultaneously during exposures that lasted as little as 10 microseconds.The Purdubik’s Cube’s high-speed Flir cameras use wide-angle lenses, and the Rubik’s Cube appears in only a very small region of their field of view. The color detection system relies on low-resolution images of the puzzle, which speeds up processing times. Photo: Matthew Patrohay / Purdue UniversityAlthough it may seem instantaneous, it takes time for a camera to process the data coming from a sensor and turn it into a digital picture. The Purdubik’s Cube uses a custom image detection system that skips image processing altogether. It also only focuses on a very small area of what each camera’s sensor sees — a cropped region that’s just 128x124 pixels in size — to reduce the amount of data being moved around.Raw data from the sensors is sent straight to a high-speed color detection system that uses the RGB measurements from even smaller sample areas on each square to determine their color faster than other approaches — even AI.“It’s sometimes slightly less reliable,” Patrohay admits, “but even if it’s 90 percent consistent, that’s good enough as long as it’s fast. We really want that speed.”Despite a lot of the hardware on Purdue’s robot being custom-made, the team chose to go with existing software when it came to figuring out the fastest way to solve a scrambled cube. They used Elias Frantar’s Rob-Twophase, which is a cube-solving algorithm that takes into account the unique capabilities of robots, like being able to spin two sides of a cube simultaneously.The team also took advantage of a Rubik’s Cube-solving technique called corner cutting where you can start to turn one side of the cube before you’ve finished turning another side that’s perpendicular to it. The advantage to this technique is that you’re not waiting for one side to completely finish its rotation before starting another. For a brief moment, there’s overlap between the movements of the two sides that can result in a significant amount of time saved when you’re chasing a world record.High-speed footage of the Purdubik’s Cube reveals how it uses the corner-cutting technique to overlap movements and reduce the time it takes to solve the Rubik’s Cube. Photo: Matthew Patrohay / Purdue UniversityThe challenge with corner cutting is that if you use too much force (like a robot is capable of) and don’t time things perfectly, you can physically break or even completely destroy a Rubik’s Cube. In addition to perfecting the timing of the robot’s movements and the acceleration of its motors, the students had to customize the cube itself.Guinness World Records follows the guidelines of the World Cube Association, which has a long list of regulations that need to be followed before a record will be recognized. It allows competitors to modify their cube, so long as it twists and turns like a standard Rubik’s Cube and has nine colored squares on each of its six sides, with each side a different color. Materials other than plastic can be used, but the color parts all need to have the same texture. To improve its durability, the Purdue team upgraded the internal structure of their cubes with a custom 3D-printed version made from stronger SLS nylon plastic. The WCA also allows the use of lubricants to help make cubes spin more freely, but here it’s used for a different reason.“The cube we use for the record is tensioned incredibly tight, like almost hilariously tight,” says Patrohay. “The one that we modified is very difficult to turn. Not impossible, but you can’t turn it with your fingers. You have to really get your wrist into it.” When solving the cube at high speeds, the lubricant helps to smooth out its movements while the increased tension reduces overturns and improves control so time-saving tricks like corner cutting can be used.Each of the robot’s six servo motors connect to the Rubik’s Cube center squares using a custom-made metal shaft that spins each side. Photo: Matthew Patrohay / Purdue UniversityFaster servo motors do help to reduce solving times, but it’s not as simple as maxing out their speed and hoping for the best. The Purdubik’s Cube uses six motors attached to metal shafts that slot into the center of each side of the cube. After testing several different approaches the team settled on a trapezoidal motion profile where the servos accelerate at speeds of up to 12,000,000 degrees/s2, but decelerate much slower, closer to 3,000,000 degrees/s2, so the robot can more accurately position each side as it comes to a stop.Could the Purdubik’s Cube break the record again? Patrohay believes it’s possible, but it would need a stronger cube made out of something other than plastic. “If you were to make a completely application-specific Rubik’s Cube out of some sort of carbon fiber composite, then I could imagine you being able to survive at higher speeds, and just being able to survive at higher speeds would then allow you to bring the time down.”See More:

ExpressVPN is good at its job. It's easy to be skeptical of any service with a knack for self-promotion, but don't let ExpressVPN's hype distract you from the fact that it keeps its front-page promise of "just working." Outside of solid security, the two best things ExpressVPN offers are fast speeds and a simple interface. Our tests showed only a 7% average drop in download speed and a 2% loss of upload speed, worldwide. And while the lack of extra features may frustrate experienced users, it makes for a true set-and-forget VPN on any platform. This isn't to say ExpressVPN is without flaws — it's nearly bereft of customization options and it's notably more expensive than its competition — but it beats most VPNs in a head-to-head matchup. For this review, we followed our rigorous 10-step VPN testing process, exploring ExpressVPN's security, privacy, speed, interfaces and more. Whether you read straight through or skip to the sections that are most important for you, you should come away with all the information you need to decide whether to subscribe. Editors' note: We're in the process of rebooting all of our VPN reviews from scratch. Once we do a fresh pass on the top services, we'll be updating each review with a rating and additional comparative information. Table of contents Findings at a glance Installing, configuring and using ExpressVPN ExpressVPN speed test: Very fast averages ExpressVPN security test: Checking for leaks How much does ExpressVPN cost? ExpressVPN side apps and bundles Close-reading ExpressVPN's privacy policy Can ExpressVPN change your virtual location? Investigating ExpressVPN's server network Extra features of ExpressVPN ExpressVPN customer support options ExpressVPN background check: From founding to Kape Technologies Final verdict Findings at a glance Category Notes Installation and UI All interfaces are clean and minimalist, with no glitches and not enough depth to get lost in Windows and Mac clients are similar in both setup and general user experience Android and iOS are likewise almost identical, but Android has a nice-looking dark mode Speed Retains a worldwide average of 93% of starting download speeds Upload speeds average 98% of starting speeds Latency rises with distance, but global average stayed under 300 ms in tests Security OpenVPN, IKEv2 and Lightway VPN protocols all use secure ciphers Packet-sniffing test showed working encryption We detected no IP leaks Blocks IPv6 and WebRTC by default to prevent leaks Pricing Base price: $12.95 per month or $99.95 per year Lowest prepaid rate: $4.99 per month Can save money by paying for 28 months in advance, but only once per account 30-day money-back guarantee Bundles ExpressVPN Keys password manager and ID alerts included on all plans Dedicated IP addresses come at an extra price ID theft insurance, data removal and credit scanning available to new one-year and two-year subscribers for free 1GB eSIM deal included through holiday.com Privacy policy No storage of connection logs or device logs permitted The only risky exceptions are personal account data (which doesn't leave the ExpressVPN website) and marketing data (which the policy says should be anonymized) An independent audit found that ExpressVPN's RAM-only server infrastructure makes it impossible to keep logs Virtual location change Successfully unblocked five international Netflix libraries, succeeding on 14 out of 15 attempts Server network 164 server locations in 105 countries 38% of servers are virtual, though most virtual locations are accessed through physical servers within 1,000 miles A large number of locations in South America, Africa and central Asia Features Simple but effective kill switch Can block ads, trackers, adult sites and/or malware sites but blocklists can't be customized Split tunneling is convenient but unavailable on iOS and modern Macs Aircove is the best VPN router, albeit expensive Customer support Setup and troubleshooting guides are organized and useful, with lots of screenshots and videos Live chat starts with a bot but you can get to a person within a couple minutes Email tickets are only accessible from the mobile apps or after live chat has failed Background check Founded in 2009; based in the British Virgin Islands Has never been caught selling or mishandling user data Turkish police seized servers in 2017 but couldn't find any logs of user activity Owned by Kape Technologies, which also owns CyberGhost and Private Internet Access A previous CIO formerly worked on surveillance in the United Arab Emirates; no evidence of shady behavior during his time at ExpressVPN Windows Version 12 leaked some DNS requests when Split Tunneling was active Installing, configuring and using ExpressVPN This section focuses on how it feels to use ExpressVPN on each of the major platforms where it's available. The first step for any setup process is to make an account on expressvpn.com and buy a subscription. Windows Once subscribed, download the Windows VPN from either expressvpn.com or the Microsoft Store, then open the .exe file. Click "Yes" to let it make changes, wait for the install, then let your computer reboot. Including the reboot, the whole process takes 5-10 minutes, most of it idle. To finish, you'll need your activation code, which you can find by going to expressvpn.com and clicking "Setup" in the top-right corner. You can install ExpressVPN's Windows app from the Microsoft store, but we found the website more convenient. Sam Chapman for Engadget Extreme simplicity is the watchword for all ExpressVPN's designs. The Windows client's launch panel consists of three buttons and less than ten words. You can change your location or let the app pick a location for you — the "Smart Location" is the server with the best combination of being nearby and unburdened. Everything else is crammed into the hamburger menu at the top left. Here, in seven tabs, you'll find the Network Lock kill switch, the four types of content blockers, the split tunneling menu and the option to change your VPN protocol. You can also add shortcuts to various websites, useful if you regularly use your VPN for the same online destinations. To sum up, there's almost nothing here to get in the way: no delays, no snags, no nested menus to get lost in. It may be the world's most ignorable VPN client. That's not a bad thing at all. Mac ExpressVPN's app for macOS is almost identical in design to its Windows app. The process for downloading and setting it up is nearly the same too. As on Windows, it can be downloaded from the App Store or sideloaded directly from the expressvpn.com download center. Only a few features are missing and a couple others have been added. Split tunneling is gone (unless you're still on a macOS lower than 11), and you won't see the Lightway Turbo setting. ExpressVPN recommends some servers, but it's easy to search the whole list. Sam Chapman for Engadget Mac users do gain access to the IKEv2 protocol, along with the option to turn off automatic IPv6 blocking — Windows users have to leave it blocked at all times. Almost every website is still accessible via IPv4, but it's useful if you do need to access a specific IPv6 address while the VPN is active. Android Android users can download ExpressVPN through the Google Play Store. Open the app, sign in and you're ready to go. The Android app has a very nice dark-colored design, only slightly marred by an unnecessary information box about how long you've used the VPN this week. ExpressVPN's Android app puts a little more information on the screen than it needs to, but still runs well. Sam Chapman for Engadget There's a large button for connecting. Clicking on the server name takes you to a list of locations. On this list, you can either search or scroll and can choose individual locations within a country that has more than one. We connected to as many far-flung server locations as we could, but not a single one took longer than a few seconds. The options menu is organized sensibly, with no option located more than two clicks deep. You will see a couple of options here that aren't available on desktop, the best of which is the ability to automatically connect to your last-used ExpressVPN server whenever your phone connects to a non-trusted wifi network. There are also a few general security tools: an IP address checker, DNS and WebRTC leak testers and a password generator. These are also available on the website, but here, they're built into the app. With the exception of the latter, we'd recommend using third-party testing tools instead — even a VPN with integrity has an incentive to make its own app look like it's working. iPhone and iPad You can only install ExpressVPN's iOS app through the app store. During setup, you may need to enter your password to allow your phone to use VPN configurations. Otherwise, there are no major differences from the Android process. ExpressVPN looks good on iPhone and iPad. Sam Chapman for Engadget The interface is not quite as pleasing as the dark-mode Android app, but it makes up for that by cutting out some of the clutter. The tabs and features are similar, though split tunneling and shortcuts are absent. Also, both mobile apps make customer support a lot more accessible than their desktop counterparts — plus, mobile is the only way to send email support tickets. Browser extension ExpressVPN also includes browser extensions for Firefox and Chrome. These let you connect, disconnect and change server locations without leaving your browser window. It's nice, but not essential unless you have a very specific web browser flow you like. ExpressVPN speed test: Very fast averages Connecting to a VPN almost always decreases your speed, but the best VPNs mitigate the drop as much as possible. We used Ookla's speed testing app to see how much of your internet speed ExpressVPN preserves. For this test, we emphasized the locations ExpressVPN uses for most of its virtual servers, including the Netherlands, Brazil, Germany and Singapore. Some terms before we start: Latency, measured in milliseconds (ms), is the time it takes one data packet to travel between your device and a web server through the VPN. Latency increases with distance. It's most important for real-time tasks like video chatting and online gaming. Download speed, measured in megabits per second (Mbps), is the amount of information that can download onto your device at one time — such as when loading a web page or streaming a video. Upload speed, also measured in Mbps, is the amount of information your device can send to the web at once. It's most important for torrenting, since the amount of data you can seed determines how fast you can download in exchange. The table below shows our results. We conducted this on Windows, using the automatic protocol setting with the Lightway Turbo feature active — a recent ExpressVPN addition that keeps speed more consistent by processing connections in parallel. Server location Latency (ms) Increase factor Download speed (Mbps) Percentage dropoff Upload speed (Mbps) Percentage dropoff Portland, Oregon, USA (unprotected) 18 -- 58.77 -- 5.70 -- Seattle, Washington, USA (best server) 26 1.4x 54.86 6.7% 5.52 3.2% New York, NY, USA 156 8.7x 57.25 2.6% 5.57 2.3% Amsterdam, Netherlands 306 17x 53.83 8.4% 5.58 2.1% São Paulo, Brazil 371 20.6x 53.82 8.4% 5.65 0.9% Frankfurt, Germany 404 22.4x 55.71 5.2% 5.67 0.5% Singapore, Singapore 381 21.2x 52.76 10.2% 5.64 1.0% Average 274 15.2x 54.71 6.9% 5.61 1.6% These are extremely good results. ExpressVPN is a winner on both download and upload speed. No matter where we went in the world, we never lost more than about 7% of our download speeds, and upload lost an astoundingly low average of 2%. This suggests that ExpressVPN deftly distributes its user load between servers to eliminate bottlenecks. This Ookla speedtest shows you can still get fast internet while connected to ExpressVPN -- our unprotected speeds are around 58 Mbps. Sam Chapman for Engadget The latency numbers look worse, but the rise in the table is less sharp than we projected. Ping length depends far more on distance than download speed does, so we expect it to shoot up on servers more than 1,000 miles from our location. Keeping the average below 300 ms, as ExpressVPN does here, is a strong showing. ExpressVPN security test: Checking for leaks A VPN's core mission is to hide your IP address and make you untraceable online. Our task in this section is to figure out if ExpressVPN can carry out this mission every time you connect. While we can't be 100% certain, the tests we'll run through below have led us to believe that ExpressVPN is currently leak-proof. Available VPN protocols A VPN protocol is like a common language that a VPN server can use to mediate between your devices and the web servers you visit. If a VPN uses outdated or insecure protocols, or relies on unique protocols with no visible specs or source code, that's a bad sign. Not all protocols are available on all apps, but Mac has the full range. Sam Chapman for Engadget ExpressVPN gives you a selection of three protocols: IKEv2, OpenVPN and Lightway. The first two are solid choices that support the latest encryption algorithms. OpenVPN has been fully open-source for years and is the best choice if privacy is your goal. While IKEv2 started life as a closed project by Microsoft and Cisco, ExpressVPN uses an open-source reverse-engineering, which is both better for privacy and quite fast. Lightway is the odd one out, a protocol you'll only find on ExpressVPN, though its source code is available on Github. It's similar to WireGuard, in that both reach for faster speeds and lower processing demands by keeping their codebases slim. However, Lightway was recently rewritten in Rust to better protect the keys stored in its memory. Ultimately, you can't go wrong with any of ExpressVPN's protocol options. 99% of the time, your best choice will be to set the controls to Automatic and let the VPN decide which runs best. Testing for leaks ExpressVPN is one of the best services, but it's not leak-proof (as you can read in the Background Check below). Luckily, checking for DNS leaks is a simple matter of checking your IP address before and after connecting to a VPN server. If the new address matches the VPN server, you're good; if not, your VPN is leaking. First, we checked the Windows app with split tunneling active to ensure the flaw really had been patched. We tested several servers and didn't find any leaks, which suggests the patch worked, though leaks were rare even before ExpressVPN fixed the vulnerability. We checked our IP while connected to the virtual India location, which is run from a physical server in Singapore. Don't worry -- it still looks like India to streaming services. Sam Chapman for Engadget In fact, we didn't find any leaks on any ExpressVPN server we tested on any platform. Though questions remain about iOS, as you'll see later in this section, that's a problem on Apple's end that even the best VPNs can do very little about for now. The most common cause of VPN leaks is the use of public DNS servers to connect users to websites, which can mistakenly send browsing activity outside the VPN's encrypted tunnel. ExpressVPN avoids the risks of the public system by installing its own DNS resolvers on every server. This is the key factor behind its clean bill of health in our leak testing. Two other common flaws can lead to VPN leaks: WebRTC traffic and IPv6. The former is a communication protocol used in live streaming and the latter is a new IP standard designed to expand domain availability. Both are nice, but currently optional, so ExpressVPN automatically blocks both to ensure there's no opportunity for leaks to arise. One note about VPN security on iOS: it's a known and continuing problem that iOS VPNs do not prevent many online apps from communicating with Apple directly, outside the VPN tunnel. This risks leaking sensitive data, even with Lockdown Mode active in iOS 16. A blog post by Proton VPN shares a workaround: connect to a VPN server, then turn Airplane Mode on and off again to end all connections that were active before you connected to the VPN. Testing encryption We finished up our battery of security tests by checking out ExpressVPN's encryption directly. Using WireShark, a free packet sniffer, we inspected what it looks like when ExpressVPN transmits data from one of its servers to the internet. The screenshot below shows a data stream encrypted with Lightway UDP. After connecting to ExpressVPN, HTTP packets were rendered unreadable while in transit. Sam Chapman for Engadget That lack of any identifiable information, or even readable information, means encryption is working as intended. We repeated the test several times, always getting the same result. This left us satisfied that ExpressVPN's core features are working as intended. How much does ExpressVPN cost? ExpressVPN subscriptions cost $12.95 per month. Long-term subscriptions can bring the monthly cost down, but the great deals they offer tend to only last for the first billing period. A 12-month subscription costs $99.95 and includes three months for free with your first payment, costing a total of $6.67 per month. The bonus disappears for all subsequent years, raising the monthly cost to $8.33. You can also sign up for 28 months at a cost of $139.72, but this is also once-only — ExpressVPN can only be renewed at the $99.95 per year level. There are two ways to test ExpressVPN for free before making a financial commitment. Users on iOS and Android can download the ExpressVPN app without entering any payment details and use it free for seven days. On any platform, there's a 30-day money-back guarantee, which ExpressVPN has historically honored with no questions asked. You will have to pay before you can use it, though. In our opinion, ExpressVPN's service is solid enough that it's worth paying extra. Perhaps not this much extra, but that depends on what you get out of it. We recommend using the 30-day refund period and seeing how well ExpressVPN works for you. If it's a VPN you can enjoy using, that runs fast and unblocks everything you need, that's worth a server's weight in gold. ExpressVPN side apps and bundles ExpressVPN includes some special features that work mostly or wholly separate from its VPN apps. Some of these come free with a subscription, while others add an extra cost. Every subscription includes the ExpressVPN keys password manager. This is available under its own tab on the Android and iOS apps. On desktop, you'll need to download a separate extension from your browser's store, then sign in using your account activation code. It's available on all Chromium browsers, but not Firefox. Starting in 2025, new subscribers get an eSIM plan through holiday.com, a separate service linked to ExpressVPN. The baseline 1GB holiday eSIM plans last for 5 days and can apply to countries, regions, or the entire world (though it's not clear whether the package deal applies to the regional and global plans). Longer-term plans include larger eSIM plans. You can add a dedicated IP address to your ExpressVPN subscription for an additional cost per month. A dedicated IP lets you use the same IP address every time you connect to ExpressVPN. You can add the address to whitelists on restricted networks, and you're assured to never be blocked because of someone else's bad activity on a shared IP. Unlike many of its competitors, ExpressVPN doesn't currently offer antivirus or online storage services, but there is a comprehensive bundle of ID protection tools called Identity Defender. We haven't reviewed any of these products in detail, but here's a list for reference: ID Alerts will inform you if any of your sensitive information is leaked or misused online. It's free with all plans, but you'll have to enter your personal information on your ExpressVPN account page or a mobile app. ID Theft Insurance grants up to $1 million in identity theft reimbursement and comes free with new ExpressVPN one-year or two-year subscriptions. It's not yet available to those who subscribed before it launched in October 2024. Data Removal scans for your information in data brokerages and automatically requests that it be deleted. It's also free with one-year and two-year plans. Credit Scanner is only available for United States users. It monitors your activity on the three credit bureaus so you can quickly spot any suspicious transactions. The Identity Defender features are currently only available to new ExpressVPN customers in the US. Close-reading ExpressVPN's privacy policy Although we worry that the consolidation of VPN brands under the umbrella of Kape Technologies (ExpressVPN's parent company) will make the industry less competitive, we don't believe it's influencing ExpressVPN to take advantage of its users' privacy. To confirm, and get a full sense of what sort of privacy ExpressVPN promises its users, we set out to read ExpressVPN's privacy policy in detail. It's long, but thankfully aimed at casual users instead of lawyers. You can see it for yourself here. In the introduction, ExpressVPN states that it does not keep either activity logs (such as a user's browsing history while connected to the VPN) or connection logs (such as the duration of a user's session and their IP address, which can be used to extrapolate browsing activity). It then specifies the seven types of data it's legally allowed to collect: Data used to sign up for an account, such as names, emails and payment methods. VPN usage data which is aggregated and can't be traced to any individual. Credentials stored in the ExpressVPN Keys password manager. Diagnostic data such as crash reports, which are only shared upon user request. IP addresses authorized for MediaStreamer, which is only for streaming devices that don't otherwise support VPN apps. Marketing data collected directly from the app — a "limited amount" that's kept anonymous. Data voluntarily submitted for identity theft protection apps. Of those seven exceptions, the only ones that count as red flags are account data and marketing data. Both categories are highly personal and could be damaging if mishandled. Fortunately, complying with subpoenas is not one of the allowed uses listed for either data category, nor does the policy let ExpressVPN sell the data to other private parties. The only really annoying thing here is that if you ask ExpressVPN to delete your personal data, you won't be able to use your account from then on. You aren't even eligible for a refund in this case, unless you're within 30 days of your initial subscription. As for marketing data, ExpressVPN collects device fingerprints and location data when you sign up for an account on its website. The privacy policy also claims this is anonymized, as its "systems are engineered to decouple such data from personally identifiable information." Audits corroborate this, as we'll see in the next section. So, while it would be better if ExpressVPN didn't collect any personal data at all, its practices don't appear to pose a risk to anything you do while using the VPN — just the ExpressVPN website. Privacy audits VPN providers often get third-party accounting firms to audit their privacy policies. The idea is that a well-known firm won't mortgage its reputation to lie on behalf of a VPN, so their results can be trusted. For the last several years, ExpressVPN has had KPMG look over its privacy policy and relevant infrastructure (see "TrustedServer" below). KPMG's most recent report, completed in December 2023 and released in May 2024, found that ExpressVPN had enough internal controls in place that users could trust its privacy policy. The report is freely available to read. This is a very good sign, though we're looking out for a more up-to-date audit soon. TrustedServer "TrustedServer" is a marketing term ExpressVPN uses for its RAM-only server infrastructure. RAM-only servers have no hard drives for long-term storage and return to a standard disk image with every reboot. This makes it theoretically impossible to store user activity logs on them, even if ExpressVPN wanted to do that. The KPMG audit, linked above, reports that TrustedServer works as advertised. Between its many clean privacy audits and the Turkish server incident in 2017, we're prepared to say ExpressVPN is a private VPN, in spite of its aggravating exception for marketing. Can ExpressVPN change your virtual location? Next, we tested whether ExpressVPN can actually convince websites that you're somewhere other than your real location. Our security tests have already proven it can hide your IP address, but it takes more than leak-proofing to fool streaming sites these days — Netflix and the others have gotten very good at combing through metadata to sniff out proxy users. The process for testing this is a lot like how we handled the DNS leak tests: try several different servers and see if we get caught. We checked five sample locations outside the U.S. to see if we a) got into Netflix and b) saw different titles in the library. The results are below. Server Location Unblocked Netflix? Library changed? Canada Y Y United Kingdom Y (second try; Docklands failed) Y Slovakia Y Y India Y Y (different from UK library) Australia Y Y In fifteen tests, ExpressVPN slipped up only once. Docklands, the UK server it chose as the fastest, wasn't able to access Netflix. We switched to a server labeled simply "London" and unblocked it without issue. ExpressVPN can change your virtual location so you can explore the wonderful world of K-drama. Sam Chapman for Engadget All the other locations got us access to an alternate Netflix library on the first try. We even checked whether the India server, which is physically located in the UK, showed us different videos than the UK servers. It did, which makes us even more confident that ExpressVPN's virtual locations are airtight. Investigating ExpressVPN's server network ExpressVPN users can connect to a total of 164 server locations in 105 countries and territories. These locations are reasonably well distributed across the globe, but as with all VPNs, there's a bias toward the northern hemisphere. There are 24 locations in the U.S. alone and a further 66 in Europe. That isn't to say users in the Global South get nothing. ExpressVPN has IP addresses from nine nations in South America (Argentina, Brazil, Bolivia, Chile, Colombia, Ecuador, Peru, Uruguay and Venezuela) and six in Africa (Algeria, Egypt, Ghana, Kenya, Morocco and South Africa). The network even includes Kazakhstan, Uzbekistan and Mongolia, impressive since central Asia may be the region most often shafted by VPNs. However, many of these servers have virtual locations different from their real ones. For those of you choosing a server based on performance instead of a particular IP address, ExpressVPN's website has a helpful list of which servers are virtual. The bad news is that it's a big chunk of the list. A total of 63 ExpressVPN locations are virtual, or 38% of its entire network. To reduce the sting, ExpressVPN takes care to locate virtual servers as close to their real locations as possible. Its virtual locations in Indonesia and India are physically based in Singapore. This isn't always practical, leading to some awkwardness like operating a Ghana IP address out of Germany. But it helps ExpressVPN perform better in the southern hemisphere. Extra features of ExpressVPN Compared to direct competitors like NordVPN and Surfshark, ExpressVPN doesn't have many special features. It's aimed squarely at the casual market and will probably disappoint power users. Having said that, what they do include works well. In this section, we'll run through ExpressVPN's four substantial features outside its VPN servers themselves. Network Lock kill switch "Network Lock" is the name ExpressVPN gives to its kill switch (though it's called "Network Protection" on mobile). A VPN kill switch is a safety feature that keeps you from broadcasting outside the VPN tunnel. If it ever detects that you aren't connected to a legitimate ExpressVPN server, it cuts off your internet access. You won't be able to get back online until you either reconnect to the VPN or disable Network Lock. ExpressVPN's kill switch is called Network Lock on desktop, and Network Protection on mobile (Android pictured) Sam Chapman for Engadget This is important for everyone, not just users who need to hide sensitive traffic. The recently discovered TunnelVision bug theoretically allows hackers to set up fake public wi-fi networks through which they redirect you to equally fake VPN servers, which then harvest your personal information. It's unlikely, but not impossible, and a kill switch is the best way to prevent it — the switch always triggers unless you're connected to a real server in the VPN's network. Like most of ExpressVPN's features, all you can do with Network Lock is turn it on and off. You can also toggle whether you'll still be able to access local devices while the kill switch is blocking your internet — this is allowed by default. Threat manager, ad blocker and parental controls ExpressVPN groups three tools under the heading of "advanced protection" — Threat Manager, an ad blocker and parental controls. Threat Manager consists of two checkboxes: one that blocks your browser from communicating with activity tracking software and one that blocks a list of websites known to be used for malware. Check any of these boxes to use the pre-set blocklists whenever you're connected to ExpressVPN. Sam Chapman for Engadget You can't customize the lists, so you're limited to what ExpressVPN considers worthy of blocking. They share their sources on the website. While the lists are extensive and open-source, they rely on after-the-fact reporting and can't detect and block unknown threats like a proper antivirus. The adblock and parental control options work the same way: check a box to block everything on the list, uncheck it to allow everything through. In tests, the ad blocker was nearly 100% effective against banner ads, but failed to block any video ads on YouTube or Netflix. The parental control option blocks a list of porn sites. It's an easy option for concerned parents, but only works while ExpressVPN is connected. As such, it's meant to be used in conjunction with device-level parental controls that prevent the child from turning off or uninstalling the VPN client. Split tunneling Sometimes, you'll find it helpful to have your device getting online through two different IP addresses at once — one for your home services and one for a location you're trying to spoof. That's where split tunneling is helpful: it runs some apps through the VPN while leaving others unprotected. This can also improve your speeds, since the VPN needs to encrypt less in total. You can configure split tunneling through either a blocklist or an allowlist. Sam Chapman for Engadget ExpressVPN includes split tunneling on Windows, Android and Mac (though only on versions 10 and below). You can only split by app, not by website, but it's still pretty useful. For example, you can have BitTorrent handling a heavy download in the background while you use your browser for innocuous activities that don't need protecting. ExpressVPN Aircove router By now, it should be clear that we find ExpressVPN to be a highly reliable but often unexceptional VPN service. However, there's one area in which it's a clear industry leader: VPN routers. ExpressVPN Aircove is, to our knowledge, the only router with a built-in commercial VPN that comes with its own dashboard interface. Usually, installing a VPN on your router requires tinkering with the router control panel, which turns off all but the most experienced users — not to mention making it a massive pain to switch to a new server location. Aircove's dashboard, by contrast, will be instantly familiar to anyone who already knows how to use an ExpressVPN client. It even allows different devices in your home to connect to different locations through the router VPN. Aircove's biggest drawback is its price. Currently retailing at $189 (not including an ExpressVPN subscription), it's around three times more expensive than an aftermarket router fitted with free VPN firmware. Some of you might still find the convenience worth the one-time payment. ExpressVPN customer support options ExpressVPN's written help pages are some of the best on the market. Its live chat is more of a mixed bag, and complex questions may cause delays. However, it is at least staffed with human agents who aim to reply accurately, rather than resolve your ticket as quickly as possible. You can directly access both live chat and email from ExpressVPN's mobile apps (on desktop, you'll have to go to the website). Sam Chapman for Engadget We approached ExpressVPN's support features with a simple question: "If I requested that ExpressVPN delete all my personal data, would I be able to get a refund for my unused subscription time?" (Remember from the Privacy Policy section that submitting a full deletion request also cancels your ExpressVPN account.) Our first stop was expressvpn.com/support, the written support center and FAQ page. It's divided into setup guides, troubleshooting, account management and information on each of ExpressVPN's products. The setup guides are excellent, including screenshots and clearly written steps; each one includes a video guide for those who learn better that way. Troubleshooting is just as good — no videos, but the same standards of clarity and usefulness prevail. The section starts with general problems, then delves into specific issues you might face on each operating system. Each article clearly derives from a real customer need. The live support experience To get answers on our refund question, we visited the account management FAQs. This section stated that the refund policy only applies within 30 days of purchase. Pretty clear-cut, but we still wanted an answer on our special case, so we contacted live chat by clicking the button at the bottom-right of every FAQ page. Live chat is in the bottom-right corner of every page of expressvpn.com. Sam Chapman for Engadget Live chat starts with an AI assistant, which is not too hard to get past — just ask it a question it can't answer, then click "Transfer to an Agent." We got online with (what claimed to be) a human in less than a minute. Answering the question took longer and involved an uncomfortable 10-minute silence, but we did get a clear verdict from a real person: refunds are within 30 days only, no matter what. If the live chat agent can't answer your question, you'll be redirected to open an email support ticket. Annoyingly, there's no way to go directly to email support through the website or desktop apps, though mobile users have the option to skip directly there. ExpressVPN background check: From founding to Kape Technologies ExpressVPN launched in 2009, which makes it one of the oldest consumer VPNs in continual operation. In more than 15 years of operation, it's never been caught violating its own privacy policy, though its record isn't free of more minor blemishes. Headquarters in the British Virgin Islands Founders Dan Pomerantz and Peter Burchhardt registered the company in the British Virgin Islands from the start to take advantage of that territory's favorable legal environment for online privacy. The BVIs have no law requiring businesses to retain data on their users, and the process for extraditing data is famously difficult, requiring a direct order from the highest court. In 2021, the BVI implemented the Data Protection Act (DPA) [PDF link], which prevents companies based in the territory from accessing data on their users anywhere in the world. It's a great privacy law in theory, modeled on best-in-class legislation in the EU. However, we couldn't find any evidence that its supervising authority — the Office of the Information Commissioner — has a leader or staff. In other words, while ExpressVPN is not legally required to log any data on its users, there's technically nobody stopping them from doing so. Whether you trust the jurisdiction depends on whether you trust the company itself. Let's see what the other evidence says. Security and privacy incidents Two significant incidents stand out from ExpressVPN's 16-year history. In 2017, when Andrei Karlov, Russia's ambassador to Turkey, was shot to death at an art show. Turkish police suspected someone had used ExpressVPN to mask their identity while they deleted information from social media accounts belonging to the alleged assassin. To investigate, they confiscated an ExpressVPN server to comb for evidence. They didn't find anything. A police seizure is the best possible test of a VPN's approach to privacy. The provider can't prepare beforehand, fake anything, or collude with investigators. The Turkey incident is still one of the best reasons to recommend ExpressVPN, though eight years is a long time for policy to change. The second incident began in March 2024, when a researcher at CNET informed ExpressVPN that its version 12 for Windows occasionally leaked DNS requests when users enabled the split tunneling feature. While these users remained connected to an ExpressVPN server, their browsing activity was often going directly to their ISP, unmasked. The bug only impacted a few users, and to their credit, ExpressVPN sprang into action as soon as they learned about it. The team had it patched by April, as confirmed by the researcher who initially discovered the vulnerability. But while their quick and effective response deserves praise, it's still a mark against them that a journalist noticed the bug before they did. Kape Technologies ownership and management questions In 2021, an Israeli-owned, UK-based firm called Kape Technologies purchased a controlling interest in ExpressVPN. In addition to ExpressVPN, privately held Kape owns CyberGhost, Private Internet Access, and Zenmate (before it merged into CyberGhost). As shown on its website, it also owns Webselenese, publisher of VPN review websites WizCase and vpnMentor, which poses an apparent conflict of interest. When reached for comment, a representative for ExpressVPN said that "ExpressVPN does not directly engage with, nor seek to influence, the content on any Webselenese site," and pointed us to disclosure statements on the websites in question — here's one example. Even so, it's a good reminder not to take VPN reviews at face value without knowing who's behind them (Engadget is owned by Yahoo, which does not own any VPNs). Diving deeper into the background of Kape's ownership will lead you to owner Teddy Sagi. Go back far enough, and you'll see he did prison time in Israel and was mentioned in the Pandora Papers, among other things. More recently, headlines about the billionaire have focused more his businesses in the online gambling and fintech arenas, as well as his real estate ventures. An ExpressVPN representative told us that "Kape's brands continue to operate independently," and our investigation bore that out — we couldn't find any proof that Kape or Sagi have directly attempted to influence ExpressVPN's software or daily operations. Closer to the immediate day-to-day operations of ExpressVPN was the company's employment of Daniel Gericke as CTO from 2019 through 2023. During that time, the US Justice Department announced it had fined Gericke and two others for their previous employment on a surveillance operation called Project Raven, which the United Arab Emirates (UAE) used to spy on its own citizens. The revelation prompted a public response from ExpressVPN defending its decision to hire Gericke, arguing that "[t]he best goalkeepers are the ones trained by the best strikers." ExpressVPN's representative confirmed that the company still stands by that linked statement. Gericke parted ways with ExpressVPN in October 2023, per his LinkedIn profile. While we don't know what we don't know, we can say that ExpressVPN has not notably changed its public-facing security and privacy policies during the time it's been connected to Kape, Sagi, or Gericke. In the end, how much ExpressVPN's history matters to you is a personal choice. If you object to any current or past actions by Kape Technologies or Teddy Sagi, there are other premium VPN options you might prefer. If you need more information to make up your mind, we recommend reading through CNET's 2022 deep dive on ExpressVPN's corporate history. Final verdict ExpressVPN is the VPN we most often recommend to beginners. It takes zero training to use, and consistently gets past filters on streaming sites. It also runs in the background with virtually no impact. If anything is worth the high price of admission, it's the excellent speeds distributed evenly across the worldwide server network. However, for certain specific cases, ExpressVPN may not be the best choice. There's no way to set up your own server locations, like NordVPN offers, and no double VPN connections, like you can build for yourself on Surfshark. Its corporate background is more suspect than the entities backing Proton VPN, and unlike Mullvad, ExpressVPN doesn't work in China — it's so well-known that the government targets its servers specifically. We suggest going with ExpressVPN for general online privacy, for spoofing locations in your home country while traveling, or if you regularly need to unblock sites in other countries. That encompasses 19 of every 20 users, which is fine by us, as ExpressVPN is a great service. It's just more of a reliable old screwdriver than a multi-tool. This article originally appeared on Engadget at https://www.engadget.com/cybersecurity/vpn/expressvpn-review-2025-fast-speeds-and-a-low-learning-curve-160052884.html?src=rss

In June, we hosted a webinar featuring experts from Arm, the Unity Accelerate Solutions team, and SYBO Games, the creator of Subway Surfers. The resulting roundtable focused on profiling tips and strategies for mobile games, the business implications of poor performance, and how SYBO shipped a hit mobile game with 3 billion downloads to date.Let’s dive into some of the follow-up questions we didn’t have time to cover during the webinar. You can also watch the full recording.We hear a lot about the Unity Profiler in relation to CPU profiling, but not as much about the Profile Analyzer (available as a Unity package). Are there any plans to improve it or integrate it into the core Profiler toolset?There are no immediate plans to integrate the Profile Analyzer into the core Editor, but this might change as our profiling tools evolve.Does Unity have any plans to add an option for the GPU Usage Profiler module to appear in percentages like it does in milliseconds?That’s a great idea, and while we can’t say yes or no at the time of this blog post, it’s a request that’s been shared with our R&D teams for possible future consideration.Do you have plans for tackling “Application Not Responding” (ANR) errors that are reported by the Google Play store and don’t contain any stack trace?Although we don’t have specific plans for tracking ANR without stack trace at the moment, we will consider it for the future roadmap.How can I share my feedback to help influence the future development of Unity’s profiling tools?You can keep track of upcoming features and share feedback via our product board and forums. We are also conducting a survey to learn more about our customers’ experience with the profiling tools. If you’ve used profiling tools before (either daily or just once) or are working on a project that requires optimization, we would love to get your input. The survey is designed to take no more than 5–10 minutes to complete.By participating, you’ll also have the chance to opt into a follow-up interview to share more feedback directly with the development team, including the opportunity to discuss potential prototypes of new features.Is there a good rule for determining what counts as a viable low-end device to target?A rule of thumb we hear from many Unity game developers is to target devices that are five years old at the time of your game’s release, as this helps to ensure the largest user base. But we also see teams reducing their release-date scope to devices that are only three years old if they’re aiming for higher graphical quality. A visually complex 3D application, for example, will have higher device requirements than a simple 2D application. This approach allows for a higher “min spec,” but reduces the size of the initial install base. It’s essentially a business decision: Will it cost more to develop for and support old devices than what your game will earn running on them?Sometimes the technical requirements of your game will dictate your minimum target specifications. So if your game uses up large amounts of texture memory even after optimization, but you absolutely cannot reduce quality or resolution, that probably rules out running on phones with insufficient memory. If your rendering solution requires compute shaders, that likely rules out devices with drivers that can’t support OpenGL ES 3.1, Metal, or Vulkan.It’s a good idea to look at market data for your priority target audience. For instance, mobile device specs can vary a lot between countries and regions. Remember to define some target “budgets” so that benchmarking goals for what’s acceptable are set prior to choosing low-end devices for testing.For live service games that will run for years, you’ll need to monitor their compatibility continuously and adapt over time based on both your actual user base and current devices on the market.Is it enough to test performance exclusively on low-end devices to ensure that the game will also run smoothly on high-end ones?It might be, if you have a uniform workload on all devices. However, you still need to consider variations across hardware from different vendors and/or driver versions.It’s common for graphically rich games to have tiers of graphical fidelity – the higher the visual tier, the more resources required on capable devices. This tier selection might be automatic, but increasingly, users themselves can control the choice via a graphical settings menu. For this style of development, you’ll need to test at least one “min spec” target device per feature/workload tier that your game supports.If your game detects the capabilities of the device it’s running on and adapts the graphics output as needed, it could perform differently on higher end devices. So be sure to test on a range of devices with the different quality levels you’ve programmed the title for.Note: In this section, we’ve specified whether the expert answering is from Arm or Unity.Do you have advice for detecting the power range of a device to support automatic quality settings, particularly for mobile?Arm: We typically see developers doing coarse capability binning based on CPU and GPU models, as well as the GPU shader core count. This is never perfect, but it’s “about right.” A lot of studios collect live analytics from deployed devices, so they can supplement the automated binning with device-specific opt-in/opt-out to work around point issues where the capability binning isn’t accurate enough.As related to the previous question, for graphically rich content, we see a trend in mobile toward settings menus where users can choose to turn effects on or off, thereby allowing them to make performance choices that suit their preferences.Unity: Device memory and screen resolution are also important factors for choosing quality settings. Regarding textures, developers should be aware that Render Textures used by effects or post-processing can become a problem on devices with high resolution screens, but without a lot of memory to match.Given the breadth of configurations available (CPU, GPU, SOC, memory, mobile, desktop, console, etc.), can you suggest a way to categorize devices to reduce the number of tiers you need to optimize for?Arm: The number of tiers your team optimizes for is really a game design and business decision, and should be based on how important pushing visual quality is to the value proposition of the game. For some genres it might not matter at all, but for others, users will have high expectations for the visual fidelity.Does the texture memory limit differ among models and brands of Android devices that have the same amount of total system memory?Arm: To a first-order approximation, we would expect the total amount of texture memory to be similar across vendors and hardware generations. There will be minor differences caused by memory layout and alignment restrictions, so it won’t be exactly the same.Is it CPU or GPU usage that contributes the most to overheating on mobile devices?Arm: It’s entirely content dependent. The CPU, GPU, or the DRAM can individually overheat a high-end device if pushed hard enough, even if you ignore the other two completely. The exact balance will vary based on the workload you are running.What tips can you give for profiling on devices that have thermal throttling? What margin would you target to avoid thermal throttling (i.e., targeting 20 ms instead of 33 ms)?Arm: Optimizing for frame time can be misleading on Android because devices will constantly adjust frequency to optimize energy usage, making frame time an incomplete measure by itself. Preferably, monitor CPU and GPU cycles per frame, as well as GPU memory bandwidth per frame, to get some value that is independent of frequency. The cycle target you need will depend on each device’s chip design, so you’ll need to experiment.Any optimization helps when it comes to managing power consumption, even if it doesn’t directly improve frame rate. For example, reducing CPU cycles will reduce thermal load even if the CPU isn’t the critical path for your game.Beyond that, optimizing memory bandwidth is one of the biggest savings you can make. Accessing DRAM is orders of magnitude more expensive than accessing local data on-chip, so watch your triangle budget and keep data types in memory as small as possible.Unity: To limit the impact of CPU clock frequency on the performance metrics, we recommend trying to run at a consistent temperature. There are a couple of approaches for doing this:Run warm: Run the device for a while so that it reaches a stable warm state before profiling.Run cool: Leave the device to cool for a while before profiling. This strategy can eliminate confusion and inconsistency in profiling sessions by taking captures that are unlikely to be thermally throttled. However, such captures will always represent the best case performance a user will see rather than what they might actually see after long play sessions. This strategy can also delay the time between profiling runs due to the need to wait for the cooling period first.With some hardware, you can fix the clock frequency for more stable performance metrics. However, this is not representative of most devices your users will be using, and will not report accurate real-world performance. Basically, it’s a handy technique if you are using a continuous integration setup to check for performance changes in your codebase over time.Any thoughts on Vulkan vs OpenGL ES 3 on Android? Vulkan is generally slower performance-wise. At the same time, many devices lack support for various features on ES3.Arm: Recent drivers and engine builds have vastly improved the quality of the Vulkan implementations available; so for an equivalent workload, there shouldn’t be a performance gap between OpenGL ES and Vulkan (if there is, please let us know). The switch to Vulkan is picking up speed and we expect to see more people choosing Vulkan by default over the next year or two. If you have counterexamples of areas where Vulkan isn’t performing well, please get in touch with us. We’d love to hear from you.What tools can we use to monitor memory bandwidth (RAM <-> VRAM)?Arm: The Streamline Profiler in Arm Mobile Studio can measure bandwidth between Mali GPUs and the external DRAM (or system cache).Should you split graphical assets by device tiers or device resolution?Arm: You can get the best result by retuning assets, but it’s expensive to do. Start by reducing resolution and frame rate, or disabling some optional post-processing effects.What is the best way to record performance metric statistics from our development build?Arm: You can use the Performance Advisor tool in Arm Mobile Studio to automatically capture and export performance metrics from the Mali GPUs, although this comes with a caveat: The generation of JSON reports requires a Professional Edition license.Unity: The Unity Profiler can be used to view common rendering metrics, such as vertex and triangle counts in the Rendering module. Plus you can include custom packages, such as System Metrics Mali, in your project to add low-level Mali GPU metrics to the Unity Profiler.What are your recommendations for profiling shader code?You need a GPU Profiler to do this. The one you choose depends on your target platform. For example, on iOS devices, Xcode’s GPU Profiler includes the Shader Profiler, which breaks down shader performance on a line-by-line basis.Arm Mobile Studio supports Mali Offline Compiler, a static analysis tool for shader code and compute kernels. This tool provides some overall performance estimates and recommendations for the Arm Mali GPU family.When profiling, the general rule is to test your game or app on the target device(s). With the industry moving toward more types of chipsets (Apple M1, Arm, x86 by Intel, AMD, etc.), how can developers profile and pinpoint issues on the many different hardware configurations in a reasonable amount of time?The proliferation of chipsets is primarily a concern on desktop platforms. There are a limited number of hardware architectures to test for console games. On mobile, there’s Apple’s A Series for iOS devices and a range of Arm and Qualcomm architectures for Android – but selecting a manageable list of representative mobile devices is pretty straightforward.On desktop it’s trickier because there’s a wide range of available chipsets and architectures, and buying Macs and PCs for testing can be expensive. Our best advice is to do what you can. No studio has infinite time and money for testing. We generally wouldn’t expect any huge surprises when comparing performance between an Intel x86 CPU and a similarly specced AMD processor, for instance. As long as the game performs comfortably on your minimum spec machine, you should be reasonably confident about other machines. It’s also worth considering using analytics, such as Unity Analytics, to record frame rates, system specs, and player options’ settings to identify hotspots or problematic configurations.We’re seeing more studios move to using at least some level of automated testing for regular on-device profiling, with summary stats published where the whole team can keep an eye on performance across the range of target devices. With well-designed test scenes, this can usually be made into a mechanical process that’s suited for automation, so you don’t need an experienced technical artist or QA tester running builds through the process manually.Do you ever see performance issues on high-end devices that don’t occur on the low-end ones?It’s uncommon, but we have seen it. Often the issue lies in how the project is configured, such as with the use of fancy shaders and high-res textures on high-end devices, which can put extra pressure on the GPU or memory. Sometimes a high-end mobile device or console will use a high-res phone screen or 4K TV output as a selling point but not necessarily have enough GPU power or memory to live up to that promise without further optimization.If you make use of the current versions of the C# Job System, verify whether there’s a job scheduling overhead that scales with the number of worker threads, which in turn, scales with the number of CPU cores. This can result in code that runs more slowly on a 64+ core Threadripper™ than on a modest 4-core or 8-core CPU. This issue will be addressed in future versions of Unity, but in the meantime, try limiting the number of job worker threads by setting JobsUtility.JobWorkerCount.What are some pointers for setting a good frame budget?Most of the time when we talk about frame budgets, we’re talking about the overall time budget for the frame. You calculate 1000/target frames per second (fps) to get your frame budget: 33.33 ms for 30 fps, 16.66 ms for 60 fps, 8.33 ms for 120 Hz, etc. Reduce that number by around 35% if you’re on mobile to give the chips a chance to cool down between each frame. Dividing the budget up to get specific sub-budgets for different features and/or systems is probably overkill except for projects with very specific, predictable systems, or those that make heavy use of Time Slicing.Generally, profiling is the process of finding the biggest bottlenecks – and therefore, the biggest potential performance gains. So rather than saying, “Physics is taking 1.2 ms when the budget only allows for 1 ms,” you might look at a frame and say, “Rendering is taking 6 ms, making it the biggest main thread CPU cost in the frame. How can we reduce that?”It seems like profiling early and often is still not common knowledge. What are your thoughts on why this might be the case?Building, releasing, promoting, and managing a game is difficult work on multiple fronts. So there will always be numerous priorities vying for a developer’s attention, and profiling can fall by the wayside. They know it’s something they should do, but perhaps they’re unfamiliar with the tools and don’t feel like they have time to learn. Or, they don’t know how to fit profiling into their workflows because they’re pushed toward completing features rather than performance optimization.Just as with bugs and technical debt, performance issues are cheaper and less risky to address early on, rather than later in a project’s development cycle. Our focus is on helping to demystify profiling tools and techniques for those developers who are unfamiliar with them. That’s what the profiling e-book and its related blog post and webinar aim to support.Is there a way to exclude certain methods from instrumentation or include only specific methods when using Deep Profiling in the Unity Profiler? When using a lot of async/await tasks, we create large stack traces, but how can we avoid slowing down both the client and the Profiler when Deep Profiling?You can enable Allocation call stacks to see the full call stacks that lead to managed allocations (shown as magenta in the Unity CPU Profiler Timeline view). Additionally, you can – and should! – manually instrument long-running methods and processes by sprinkling ProfilerMarkers throughout your code. There’s currently no way to automatically enable Deep Profiling or disable profiling entirely in specific parts of your application. But manually adding ProfilerMarkers and enabling Allocation call stacks when required can help you dig down into problem areas without having to resort to Deep Profiling.As of Unity 2022.2, you can also use our IgnoredByDeepProfilerAttribute to prevent the Unity Profiler from capturing method calls. Just add the IgnoredByDeepProfiler attribute to classes, structures, and methods.Where can I find more information on Deep Profiling in Unity?Deep Profiling is covered in our Profiler documentation. Then there’s the most in-depth, single resource for profiling information, the Ultimate Guide to profiling Unity games e-book, which links to relevant documentation and other resources throughout.Is it correct that Deep Profiling is only useful for the Allocations Profiler and that it skews results so much that it’s not useful for finding hitches in the game?Deep Profiling can be used to find the specific causes of managed allocations, although Allocation call stacks can do the same thing with less overhead, overall. At the same time, Deep Profiling can be helpful for quickly investigating why one specific ProfilerMarker seems to be taking so long, as it’s more convenient to enable than to add numerous ProfilerMarkers to your scripts and rebuild your game. But yes, it does skew performance quite heavily and so shouldn’t be enabled for general profiling.Is VSync worth setting to every VBlank? My mobile game runs at a very low fps when it’s disabled.Mobile devices force VSync to be enabled at a driver/hardware level, so disabling it in Unity’s Quality settings shouldn’t make any difference on those platforms. We haven’t heard of a case where disabling VSync negatively affects performance. Try taking a profile capture with VSync enabled, along with another capture of the same scene but with VSync disabled. Then compare the captures using Profile Analyzer to try to understand why the performance is so different.How can you determine if the main thread is waiting for the GPU and not the other way around?This is covered in the Ultimate Guide to profiling Unity games. You can also get more information in the blog post, Detecting performance bottlenecks with Unity Frame Timing Manager.Generally speaking, the telltale sign is that the main thread waits for the Render thread while the Render thread waits for the GPU. The specific marker names will differ depending on your target platform and graphics API, but you should look out for markers with names such as “PresentFrame” or “WaitForPresent.”Is there a solid process for finding memory leaks in profiling?Use the Memory Profiler to compare memory snapshots and check for leaks. For example, you can take a snapshot in your main menu, enter your game and then quit, go back to the main menu, and take a second snapshot. Comparing these two will tell you whether any objects/allocations from the game are still hanging around in memory.Does it make sense to optimize and rewrite part of the code for the DOTS system, for mobile devices including VR/AR? Do you use this system in your projects?A number of game projects now make use of parts of the Data-Oriented Technology Stack (DOTS). Native Containers, the C# Job System, Mathematics, and the Burst compilerare all fully supported packages that you can use right away to write optimal, parallelized, high-performance C# (HPC#) code to improve your project’s CPU performance.A smaller number of projects are also using Entities and associated packages, such as the Hybrid Renderer, Unity Physics, and NetCode. However, at this time, the packages listed are experimental, and using them involves accepting a degree of technical risk. This risk derives from an API that is still evolving, missing or incomplete features, as well as the engineering learning curve required to understand Data-Oriented Design (DOD) to get the most out of Unity’s Entity Component System (ECS). Unity engineer Steve McGreal wrote a guide on DOTS best practices, which includes some DOD fundamentals and tips for improving ECS performance.How do you go about setting limits on SetPass calls or shader complexity? Can you even set limits beforehand?Rendering is a complex process and there is no practical way to set a hard limit on the maximum number of SetPass calls or a metric for shader complexity. Even on a fixed hardware platform, such as a single console, the limits will depend on what kind of scene you want to render, and what other work is happening on the CPU and GPU during a frame.That’s why the rule on when to profile is “early and often.” Teams tend to create a “vertical slice” demo early on during production – usually a short burst of gameplay developed to the level of visual fidelity intended for the final game. This is your first opportunity to profile rendering and figure out what optimizations and limits might be needed. The profiling process should be repeated every time a new area or other major piece of visual content is added.Here are additional resources for learning about performance optimization:BlogsOptimize your mobile game performance: Expert tips on graphics and assetsOptimize your mobile game performance: Expert tips on physics, UI, and audio settingsOptimize your mobile game performance: Expert tips on profiling, memory, and code architecture from Unity’s top engineersExpert tips on optimizing your game graphics for consolesProfiling in Unity 2021 LTS: What, when, and howHow-to pagesProfiling and debugging toolsHow to profile memory in UnityBest practices for profiling game performanceE-booksOptimize your console and PC game performanceOptimize your mobile game performanceUltimate guide to profiling Unity gamesLearn tutorialsProfiling CPU performance in Android builds with Android StudioProfiling applications – Made with UnityEven more advanced technical content is coming soon – but in the meantime, please feel free to suggest topics for us to cover on the forum and check out the full roundtable webinar recording.

Published May 31, 2025 6:00am EDT close Brain implant enables ALS patient to communicate using AI ALS patient communicates with the world using only his thoughts. Imagine losing your ability to speak or move, yet still having so much to say. For Brad G. Smith, this became his reality after being diagnosed with ALS, a rare and progressive disease that attacks the nerves controlling voluntary muscle movement. But thanks to a groundbreaking Neuralink brain implant, Smith is now able to communicate with the world using only his thoughts. ALS patient Brad G. Smith and his family. (Bradford G. Smith/X)Life before NeuralinkBefore receiving the Neuralink implant, Smith relied on eye-tracking technology to communicate. While impressive, it came with major limitations. "It is a miracle of technology, but it is frustrating. It works best in dark rooms, so I was basically Batman. I was stuck in a dark room," Smith shared in a recent post on X. Bright environments would disrupt the system, making communication slow and sometimes impossible. Now, Smith says, "Neuralink lets me go outside and ignore lighting changes."PARALYZED MAN WITH ALS IS THIRD TO RECEIVE NEURALINK IMPLANT, CAN TYPE WITH BRAIN ALS patient Brad G. Smith. (Bradford G. Smith/X)How the Neuralink brain implant worksSmith is the first non-verbal person and only the third individual worldwide to receive the Neuralink Brain-Computer Interface (BCI). The device, about as thick as five stacked coins, sits in his skull and connects to the motor cortex-the part of the brain that controls movement.Tiny wires, thinner than human hair, extend into Smith's brain. These pick up signals from his neurons and transmit them wirelessly to his MacBook Pro. The computer then decodes these signals, allowing Smith to move a cursor on the screen with his thoughts alone.As Smith explains, "The Neuralink implant embedded in my brain contains 1024 electrodes that capture neuron firings every 15 milliseconds generating a vast amount of data. Artificial intelligence processes this data on a connected MacBook Pro to decode my intended movements in real time to move the cursor on my screen. Neuralink does not read my deepest thoughts or words I think about. It just reads how I wanna move and moves the cursor where I want."WHAT IS ARTIFICIAL INTELLIGENCE (AI)? Neuralink brain implant. (Bradford G. Smith/X)Training the brain-computer connectionLearning to use the system took some trial and error. At first, the team tried mapping Smith's hand movements to the cursor, but it didn't work well. After more research, they discovered that signals related to his tongue were the most effective for cursor movement, and clenching his jaw worked best for clicking. "I am not actively thinking about my tongue, just like you don't think about your wrist when you move a mouse. I have done a lot of cursor movements in my life. I think my brain has switched over to subconscious control quickly so I just think about moving the cursor," Smith said. ALS patient Brad G. Smith with his wife and child. (Bradford G. Smith/X)Everyday life: Communication, play, and problem-solvingThe Neuralink implant has given Smith new ways to interact with his family and the world. He can now play games like Mario Kart with his children and communicate more quickly than before. The system includes a virtual keyboard and shortcuts for common actions, making tasks like copying, pasting and navigating web pages much easier.Smith also worked with Neuralink engineers to develop a "parking spot" feature for the cursor. "Sometimes you just wanna park the cursor and watch a video. When it is in the parking spot, I can watch a show or take a nap without worrying about the cursor," he explained. ALS patient Brad G. Smith and his child. (Bradford G. Smith/X)AI assistance: Keeping up with conversationTo speed up communication even more, Smith uses Grok, Elon Musk's AI chatbot. Grok helps him write responses and even suggests witty replies. "We have created a chat app that uses AI to listen to the conversation and gives me options to say in response. It uses Grok 3 and an AI clone of my old voice to generate options for me to say. It is not perfect, but it keeps me in the conversation and it comes up with some great ideas," Smith shared. One example? When a friend needed a gift idea for his girlfriend who loves horses, the AI suggested a bouquet of carrots. ALS patient Brad G. Smith and his family. (Bradford G. Smith/X)The human side: Family, faith and perspectiveSmith's journey has been shaped by more than just technology. He credits his wife, Tiffany, as his "best caregiver I could ever imagine," and recognizes the support of his kids, friends and family. Despite the challenges of ALS, Smith finds meaning and hope in his faith. "I have not always understood why God afflicted me with ALS but with time I am learning to trust his plan for me. I'm a better man because of ALS. I'm a better disciple of Jesus Christ because of ALS. I'm closer to my amazing wife, literally and figuratively, because of ALS," he said. ALS patient Brad G. Smith and his family. (Bradford G. Smith/X)Looking ahead: What does this mean for others?Neuralink's technology is still in its early stages, but Smith's experience is already making waves. The company recently received a "breakthrough" designation from the Food and Drug Administration for its brain implant device, which hopes to help people with severe speech impairments caused by ALS, stroke, spinal cord injury and other neurological conditions.Neuro-ethicists are watching closely, as the merging of brain implants and AI raises important questions about privacy, autonomy and the future of human communication. ALS patient Brad G. Smith and his family. (Bradford G. Smith/X)Kurt's key takeawaysSmith's story is about resilience, creativity and the power of technology to restore something as fundamental as the ability to communicate. As Smith puts it,CLICK HERE TO GET THE FOX NEWS APPIf you or a family member lost the ability to speak or move, would you consider a brain implant that lets you communicate with your thoughts? Let us know by writing to us atCyberguy.com/ContactFor more of my tech tips and security alerts, subscribe to my free CyberGuy Report Newsletter by heading to Cyberguy.com/NewsletterAsk Kurt a question or let us know what stories you'd like us to cover.Follow Kurt on his social channels:Answers to the most-asked CyberGuy questions:New from Kurt:Copyright 2025 CyberGuy.com. All rights reserved.   Kurt "CyberGuy" Knutsson is an award-winning tech journalist who has a deep love of technology, gear and gadgets that make life better with his contributions for Fox News & FOX Business beginning mornings on "FOX & Friends." Got a tech question? Get Kurt’s free CyberGuy Newsletter, share your voice, a story idea or comment at CyberGuy.com.

WTF?! Some people prefer to hide their PCs to save desk space or maintain a minimalist aesthetic, but a few modders have taken extreme measures to conceal their rigs inside furniture. The latest example successfully crams a mid-range gaming PC into an unassuming office chair. A recent video from YouTuber and modder "Basically Homeless" showcases one of the most unusual methods for conserving space in a PC gaming battle station: turning the chair into a case mod. Instructions for 3D printing an enclosure to fit inside a FlexiSpot office chair are available to subscribers of the YouTuber's free-tier Patreon. Related reading: FlexiSpot C7 Ergonomic Office Chair Review Luckily, the chair FlexiSpot donated for the video has an opening between the seat cushion and the chair mechanism just wide enough to accommodate a mini-ITX motherboard equipped with a Ryzen 7 9800X3D and 64GB of RAM. Inserting 50mm aluminum standoffs provides enough space for a low-profile cooler, a flex power supply unit (normally used in server racks), and a mini-ITX Nvidia RTX 4060. After some trial and error, Basically Homeless designed and 3D printed a custom enclosure to conceal the PC components between the seat and cylinder with sufficient ventilation and several I/O ports. The I/O port openings support keystone modules, allowing the modder to hot-swap various ports such as HDMI outputs, USB ports, and headphone jacks. However, when using the chair PC normally, the only cable that is partially visible at the bottom is the power cord, which Basically Homeless ran through several holes he cut into the base of the chair. This leaves the display as the last component that normally requires wires, which prompted the most unorthodox step of the entire project. A wireless monitor that receives a video signal over Wi-Fi is one option, but it adds about 10 milliseconds of input lag, which Basically Homeless couldn't accept. So he ran another cable through the lumbar and headrest, which connects to a Bigscreen Beyond VR headset. Even when playing non-VR games, it (or, alternatively, a Meta Quest) can project a virtual 1080p screen in front of the viewer. Impressively, the PC components remain undamaged when reclining, and Basically Homeless doesn't feel them through the seat. However, he might have inadvertently turned it into a heated seat. // Related Stories In principle, the project resembles the Endgame Invisible PC, which modder and YouTuber Matthew Perks installed inside a desk last year. It includes a fold-out monitor, two PSUs, an RTX 4090, and liquid cooling.

the fastest one 2025 Chevrolet Corvette ZR1 first drive: Engineered for insane speed Now that Chevrolet can fit turbos to the Corvette, it's gone and done just that. Michael Teo Van Runkle – May 30, 2025 10:00 am | 0 Chevrolet has given its latest Corvette variant a four-figure power output to go with a six-figure price tag. Credit: Michael Teo Van Runkle Chevrolet has given its latest Corvette variant a four-figure power output to go with a six-figure price tag. Credit: Michael Teo Van Runkle Story text Size Small Standard Large Width * Standard Wide Links Standard Orange * Subscribers only   Learn more Chevrolet provided flights from Los Angeles to Austin and accommodation so Ars could drive the ZR1. Ars does not accept paid editorial content. AUSTIN, Texas—By just my third lap in the top-spec 2025 Chevrolet Corvette ZR1, I glanced down at the speedometer toward the end of the Circuit of the Americas' long back straight and spied 181 mph (291 km/h) displayed for a split second. Not bad for Chevy’s newest flagship sports car, especially given that the $174,995 ZR1’s twin-turbocharged V8 pumps all 1,064 horsepower to the rear wheels only. The US’s only purpose-built F1 track made for an excellent setting to taste Corvette’s latest; the ZR1 also commanded your attention while conquering the steep uphill toward the first corner, then winding through a series of challenging corners with plenty of elevation change. Luckily, the car itself is an engineering marvel, and Chevy brought along a team of engineers to explain exactly how the total package comes together to enable such a breathtaking pace, as well as how Chevy can responsibly sell such a powerful car to the general public at all. The entire point of switching the Corvette’s eighth generation to a mid-engine layout was to improve weight distribution and allow the Corvette to compete against much more exotic competition from European OEMs like Ferrari. The front-engined car's engine bay also lacked the width to add a pair of turbos, due to the suspension and tire orientation, which dictated the use of a supercharger that kept peak power to “just” 755 hp (563 kW) in the last Corvette to wear the ZR1 badge. It's a tight fit in there. Credit: Michael Teo Van Runkle COTA reveals the ZR1's excellent balance, especially when specced with the optional aero package, carbon fiber wheels, and Michelin’s Cup 2 R tires. The tires—in effect, grooved slicks—allow for improved lateral acceleration but also the ability to consistently put the four-figure horsepower down to the asphalt. Yet Chevy’s engineers readily admitted the original target for ZR1 was just 850 hp (634 kW), until 1,000 came into sight and required some serious creativity to reach reliably. Biggest turbos ever The ZR1’s engine, dubbed LT7, shares much with the 5.5 L naturally aspirated LT6 engine in the less-powerful, cheaper Z06. It’s still a flat-plane crank with dry-sump oiling, even if clever eyes inspecting an LT6 might have noticed that the dry-sump oil tank allowed for the placement of turbos all along. The dual 74-millimeter turbos, the largest ever fitted to a production car, required new intake routing, and computer control of the wastegate actuation maintains an anti-lag boost of 6 to 7 psi even under a closed throttle. Turbo speed sensors allow the turbines to spin closer to maximum speed before the vanes physically break apart—a mechanical system typically needs to maintain a 7 percent margin of error, but the ZR1’s is more like 2–3 percent. That's a massive turbocharger, and there are two of them. Credit: Michael Teo Van Runkle The eventual power output actually wound up breaking two of Chevrolet’s dynos during early testing, we're told. So the C8’s eight-speed dual-clutch transmission also needed beefing up with physically wider gears that were shot-peened for additional strength, plus a revised lubrication system. The engine, meanwhile, creates enormous cooling demands when running at full throttle, which plays hand in hand with the downforce requirements of hitting such high speeds. Consequently, the ZR1 sacrifices its usable frunk in favor of a massive radiator, while the hood’s heat extractor also releases trapped air and feeds it over the roofline. This freed up more space for additional cooling via the front fascia, which further benefits from canard spat dive planes. On the sides, an additional inlet on the side strakes complements the enormously wide scoops that debuted on the Z06. Coupes then get a split rear window—which harks back to Corvettes of old, while releasing hot air from the engine bay—plus new shoulder NACA intakes that directly feed the air box with cooler oxygen that even creates a ram air effect akin to mild supercharging. Cooling for the ZR1 became an even higher priority, because the LT6 and LT7 employ extremely tight tolerances between the crankshaft and connecting rods, which mandates keeping the 5W-50 oil below 120° C (248° F) at all times. And the system simply works, as even on a hot and humid Texas day, I only noticed oil temperatures cresting above 104° C (220° F) occasionally. The interior is better than any prior generation of Corvette, but it feels prosaic compared to the cockpits of its more exotic mid-engined rivals. Michael Teo Van Runkle The interior is better than any prior generation of Corvette, but it feels prosaic compared to the cockpits of its more exotic mid-engined rivals. Michael Teo Van Runkle Lightweight carbon-fiber wheels are mounted with the stickiest road-legal tires Chevy could fit. Michael Teo Van Runkle Lightweight carbon-fiber wheels are mounted with the stickiest road-legal tires Chevy could fit. Michael Teo Van Runkle The ZR1 gets added cooling and more wings. Michael Teo Van Runkle The ZR1 gets added cooling and more wings. Michael Teo Van Runkle Lightweight carbon-fiber wheels are mounted with the stickiest road-legal tires Chevy could fit. Michael Teo Van Runkle The ZR1 gets added cooling and more wings. Michael Teo Van Runkle The hardtop convertible ZR1 lacks the split-engine venting and shoulder intakes, while cutting into headroom so much that I skipped out while wearing a helmet. Other journalists noticed a drop-off in performance for the convertibles, and probably more so than the mild weight gains of just about 100 lbs (45 kg) might suggest. Instead, temperatures probably came into play, as the ECU drew back timing and instead allowed mild overboost of 24–25 psi to compensate for the Texas day. Even so, an engineer admitted he thought the engine was probably down 5–10 percent on power. The fact that I hit my highest-ever top speed despite the ZR1 potentially giving up somewhere between 53 to 106 hp (40–80 kW) only makes this Corvettes sound even more insane. But I essentially wound up driving the turbos, since the DCT’s gear ratios carry over from the Stingray and therefore drop out of peak power when shifting from second to third and third to fourth. I suspect nothing short of an F1 racecar feels this fast on a circuit of this size. A track designed for corner exit speeds double my pace in the ZR1 helps explain why Chevrolet declined to set us loose on public roads behind the wheel. We drove it on track—will owners cope with this much power on the street? Credit: Michael Teo Van Runkle That’s a concern for potential buyers, though, and why the ZR1’s electronics undoubtedly ratchet back the insanity. Chevy still uses Bosch’s ninth-generation traction control, which debuted on C7 and operates on a 10-millisecond loop, even if the ABS runs at 5 milliseconds—while the ESC is at 20 milliseconds. I suspect this computerized nannying slowed me down a fair amount, in addition to the torque-by-gear restrictions in first and second that purposefully protect driveline components. We’ve probably reached peak internal-combustion Corvette, which is something of a hint about the all-too-real question of where Chevy can go from here. If so, this car reaches a new level of unfathomable American ingenuity, combined with a newfound level of refinement and traction management that attempts to belie the undeniable absurdity to a minimal, arguably necessary, extent. 0 Comments