News
- Metal Gear Solid Delta: Snake Eater's multiplayer mode, Fox Hunt, won't support cross-play
- Konami confirmed this in a social media post, saying cross-play won't be supported between PC and console
- Fans have expressed disappointment and are urging the developer to delay the mode until crossplay gets added
Konami has confirmed that Metal Gear Solid Delta: Snake Eater's multiplayer mode, Fox Hunt, won't support cross-play between console and PC.
This announcement comes from the Japanese Metal Gear X / Twitter account, just weeks before the game's release, where the developer said that "cross-play between different platforms will not be supported" (machine translated).
This means players on PC via Steam, PlayStation 5, Xbox Series X, and Xbox Series S won't be able to play the multiplayer together.
Following Konami's statement, fans have been expressing disappointment online, with many asking the studio to delay Fox Hunt until crossplay is ready to be added.
"Delay FOX HUNT, Please. Crossplay is necessary for players to have fun with their friends on other platforms!" one user said under the post.
"Even if it needs a bigger delay do not launch the mode without crossplay it's 2025 crossplay is not a feature anymore it's something needed," another fan wrote.
【ご案内】『METAL GEAR SOLID Δ: SNAKE EATER』のオンライン対戦モード 「FOX HUNT」につきまして、異なるプラットフォーム間でのクロスプレイは非対応となります。対応機種PlayStation®5、Xbox Series X|S、Steam®#MGSDelta #MetalGearSolid #メタルギアソリッド pic.twitter.com/sGsv3CjyhWAugust 13, 2025
The reactions over on the Metal Gear subreddit are similar, though fans are hopeful the crossplay will be added to the mode in a later update.
"Crossplay should have become the norm by now," one Redditor said.
Fox Hunt was announced in June and is described as a "completely original online battle mode" that will play differently from 2008's Metal Gear Online.
The multiplayer is being directed by series veteran Yu Sahara and takes place in the same world as the main game. It will also feature hide-and-seek mechanics, mixed with stealth and survival elements.
Metal Gear Solid Delta: Snake Eater arrives on August 28 for PS5, Xbox Series X|S, and PC.
You might also like...- The Nintendo Switch 2 is the company’s least ambitious console to date, but its improvements are astronomical
- Assassin's Creed Shadows devs tease the return of some past series lore, including the Assassin vs. Templar conflict, the Isu, and a modern-day storyline – 'You can expect it to be expanded at some point, in some fashion'
- Monster Hunter Wilds patch notes reveal a new endgame expansion, 9-star monsters and additional stability fixes for PC
- HTC has a new pair of smart glasses called the Vive Eagle
- They boast many of the same features as their rivals
- You can only get them in Taiwan right now
HTC has just announced a new Vive wearable, but it’s not another VR headset – instead it’s a pair of stylish AI glasses called Vive Eagle. And I’m feeling twinges of jealousy towards anyone that can get a pair.
That’s because these smart specs could put up a good fight against the Ray-Ban Meta smart glasses I’ve come to love (as well as the new Oakley HSTN specs that take the RayBans and make them a smidge better).
Hardware-wise, Vive goes beat for beat with the competition. It boasts a 12MP snapper as well as open-ear speakers, and a solid battery life of up to 36 hours of standby time and 4.5 hours of music playback while still boasting a sleek 49g design.
Where I think it truly shines however is its AI.
(Image credit: HTC)That’s because while Vive’s own assistant can help you with tasks it can also team up with your choice of ChatGPT or Google Gemini rather than locking you to one service – like how Meta’s glasses are entirely dependent on Meta AI.
As we’ve come to expect from AI companions, Vive also makes clear that “all user data is stored locally” on its glasses meaning it isn’t used for model training. When using third-party AI services your data is also anonymised to provide improved privacy.
Rounding off the AI upgrades, Vive’s specs are better with languages. Meta’s AI currently lets its smart specs translate between English, French, Spanish, Italian and German, Meanwhile Vive’s assistant supports Arabic, Traditional Chinese, English, French, German, Greek, Italian, Japanese, Portuguese, Spanish, Korean, Thai, and Turkish.
And I can’t ignore the Eagle’s striking design. I’m boring enough to want to grab the translucent black pair, but the slightly see through berry red, blueish gray, and coffee brown all look beautiful and each perfectly highlight the specs’ blend of fashion and technology.
Though with a Wayfarer-like shape, these glasses perhaps don’t create enough of their own personality from a silhouette perspective.
(Image credit: HTC)Unfortunately they won’t be easy to get your hands on.
Firstly, the Vive Eagle specs are currently exclusive to Taiwan. Specifically you can find them at 2020EYEhaus premium eyewear locations and designated Taiwan Mobile OP Experience Stores.
Secondly, they are pricier than some of their rivals, such as the Meta Ray-Bans and their new Oakleys collaboration.
Viva's glasses are currently up for pre-order, and when they launch on September 1 they’ll cost NT$15,600 (New Taiwan dollars).
That comes to about $520 / £385 / AU$795 – which is higher than the most expensive Meta Ray-Bans at $379 / £379 / AU$539, and the $499 / £499 / AU$789 Oakley smart glasses with PRIZM lenses (ignoring the UK where the Oakleys are over £100 more).
I’ll need to try the Vive Eagle glasses out before passing judgement, but there is seemingly a lot to love here, and as the AI glasses competition heats up – with Meta Connect promising next-gen specs, and Android XR due next year – Vive’s Eagle wearable is a welcome addition that I look forward to seeing more of.
You might also like- Meta CTO teases ‘big wearables announcements’ for later this year but makes no VR headset promises
- These smart glasses promise to remember your life better than you can, and we simply aren’t ready for these next-gen wearables
- 4 key changes Meta is making with its Oakley smart glasses to make them stand out from the Ray-Ban AI specs
- Proton VPN has expanded its split tunneling feature to the Linux app
- Split tunneling allows users to selectively route their internet traffic
- Proton plans to bring split tunneling to its macOS app next
Proton VPN has expanded its Linux capabilities with the addition of a new split tunneling feature, currently in beta.
Split tunneling for Linux is available to subscribers using the official Ubuntu and Fedora apps for Proton VPN, which is already one of the best VPN services available according to TechRadar's testing. Though not currently offered for the unofficial Flatpak version of the app, Debian 12 support is on the way.
The latest update brings Proton VPN’s Linux app in line with its Windows and Android apps in offering the feature. Mac users needn’t feel left out with split tunneling for macOS also on Proton VPN’s summer roadmap.
What is split tunneling and why does it matter?(Image credit: Proton)Split tunneling is a popular VPN feature, providing users with a greater degree of control and flexibility for managing their internet connections. When you connect to a Virtual Private Network (VPN) server, all of your internet traffic is encrypted and routed through a secure tunnel between your device and the VPN server.
Although it’s undoubtedly beneficial to have all of your data encrypted for the best possible security, split tunneling lets you balance privacy with performance. For example, if you want to browse securely but have the fastest speeds for gaming – even the best gaming VPN can slow things down at times – you can use Proton VPN’s split tunneling feature to have the gaming app bypass the VPN connection.
Likewise, you may have difficulties accessing local network devices, such as a printer, when connecting to a VPN. With split tunneling, you can access these devices through your local connection without disrupting your VPN connection and the protection it provides to the rest of your traffic.
The same goes if you need to access local websites like your online banking or government sites.
How to use split tunneling on Proton VPN for LinuxIf you’re a Linux user with a premium Proton VPN subscription and want to take advantage of this new feature, you'll need to update to the latest Linux app version.
The beta feature is currently being rolled out gradually, but you can soon find it via the app’s settings page. Simply navigate to features and toggle on the split tunneling option.
(Image credit: Proton)For the time being, it’s only possible to select apps to exclude from the VPN connection via the exclude mode. To do this, click “Add” and check the boxes of any apps whose traffic you’d prefer not to route via the VPN tunnel (see image above). Note that you’ll then need to reconnect to the VPN and restart any excluded apps that were opened before connecting to the VPN server.
An include mode, to enable VPN protection only for selected apps, is coming soon, as are IP-based rules for the exclude and include modes.
Proton VPN is one of the best Linux VPNs we’ve reviewed, but it’s not the only VPN with split tunneling for Linux. ExpressVPN has also been busy and released split tunneling for its Linux app in March 2025.
You might also like- Dice has announced new changes for the Battlefield 6 Open Beta Weekend 2
- Playlists will receive game mode rotation adjustments, including to the All-Out Warfare, and a Custom Search feature for maps and mode selection
- The second Open Beta starts today and ends on August 17
After the massive success that was the Battlefield 6 Open Beta Weekend 1, Dice has revealed new changes it will be adding for Weekend 2.
The studio revealed the details in a new blog post, where it confirmed that it will be making playlist adjustments in Weekend 2 for more variety and choice, including to the All-Out Warfare playlist, which will include the Breakthrough, Conquest, and Rush game modes. You can check out the updated playlists below.
"Rotating through multiple modes within playlists helps us understand what our community enjoys as we shape the launch experience, and enables us to keep more fan-favorite modes available at the same time," it explained.
The second Open Beta will also add a new Custom Search feature to give players more control over map and mode selection.
"This feature allows you to choose your favorite combinations of maps and modes," the post said. "For example, if you want to play Conquest on Siege of Cairo, then matchmaking will prioritize putting you in a match with this combo, if available. Note that this is not a Server Browser (which is not being tested this week).
During Open Beta, Custom Search will be available for Open Weapon mode playlists as well, allowing players to mix and match maps and modes as they like.
In addition, Training Grounds will continue to offer the Firing Range and gameplay introductions for classes, Breakthrough, and Conquest, and players at career rank 14 and below will be able to access Initiation Mode, which includes AI Soldiers.
"This initiation experience is tailored for new players and will shift from Breakthrough to Conquest. Separately, there will be a non-playable introduction to Rush as well," the developer said.
Dice also confirmed that Open Beta Weekend 2 will kick off today, Thursday, August 14, at 8 AM UTC / 1 AM PT / 9 AM BST.
The full game launches on October 10, 2025, for PS5, Xbox Series X, Xbox Series S, and PC.
Open Beta Weekend 2 - Playlists OverviewDay 1 - Thursday, August 14:
- Conquest
- Rush
- Attack & Defend (Breakthrough, Rush)
- Close Quarters; Fast-paced playlist on smaller maps (Domination, King of the Hill)
- Closed Weapons All-Out Warfare (Conquest, Breakthrough)
Day 2 - Friday, August 15:
- Conquest
- Squad Deathmatch
- Close Quarters; Fast-paced playlist on smaller maps (Domination, King of the Hill, Squad Deathmatch)
- All-Out Warfare; Epic battles across expansive maps (Conquest, Breakthrough, Rush)
- Closed Weapons All-Out Warfare (Conquest, Breakthrough)
Day 3-4 - Saturday, August 16 - Sunday, August 17:
- Conquest
- Close Quarters; Fast-paced playlist on smaller maps (Domination, King of the Hill, Squad Deathmatch).
- All-Out Warfare; Epic battles across expansive maps (Conquest, Breakthrough, Rush)
- Closed Weapons All-Out Warfare (Conquest, Breakthrough)
- The Nintendo Switch 2 is the company’s least ambitious console to date, but its improvements are astronomical
- Assassin's Creed Shadows devs tease the return of some past series lore, including the Assassin vs. Templar conflict, the Isu, and a modern-day storyline – 'You can expect it to be expanded at some point, in some fashion'
- Monster Hunter Wilds patch notes reveal a new endgame expansion, 9-star monsters and additional stability fixes for PC
- Kaspersky warns AI is used to craft convincing emails, generate deepfakes, and more
- The number of clicks on phishing links grows quarter-on-quarter
- Users warned to stay alert to be safe from attacks
Generative Artificial Intelligence (GenAI) is making phishing attacks smarter, harder to detect, and more widespread, new research from Kaspersky has warned.
Its findings claim in the second quarter 2025, its products detected and blocked more than 142 million clicks on phishing links, representing a 3.3% increase over the first quarter.
While it can mean that there were more phishing attacks over the summer, it can also mean that the volume remained the same - but the actual attacks grew more convincing.
Generating deepfakes“AI has elevated phishing into a highly personalized threat. Large language models enable attackers to craft convincing emails, messages and websites that mimic legitimate sources, eliminating grammatical errors that once exposed scams,” the researchers said.
“AI-driven bots on social media and messaging apps impersonate real users, engaging victims in prolonged conversations to build trust. These bots often fuel romantic or investment scams, luring victims into fake opportunities with AI-generated audio messages or deepfake videos.”
The researchers also said that GenAI isn’t just being used to eliminate spelling and grammar errors - arguably the biggest red flags in phishing emails.
It is also used to create realistic audio and video deepfake impersonations of trusted individuals, including colleagues, celebrities, and bank officials.
These deepfakes are then used to promote fake giveaways, or steal sensitive information such as multi-factor authentication (MFA) codes, passwords, and similar.
“AI-powered tools analyze public data from social media or corporate websites to launch targeted attacks, such as HR-themed emails or fake calls referencing personal details,” Kaspersky added.
How to stay safeAs the risk of phishing grows, staying vigilant online remains the best way to be safe.
Users should always be skeptical of unsolicited incoming messages, especially those that demand urgent action or threaten with a disaster.
These are, and will continue to be, the biggest red flag in phishing attacks.
You might also like- This Microsoft 365 phishing campaign can bypass MFA - here's what we know
- Take a look at our guide to the best authenticator app
- We've rounded up the best password managers
Comedy Central did not air South Park season 27 episode 3 on its expected release date (August 13), with the series now thought to continue next week instead. This also means it now won’t be available on Paramount+ either. However, unlike the season’s previous delay, the episode 3 delay was scheduled in advance. Instead, the network declared the day to be “South Park day,” running a marathon of fan-favorite episodes followed by the sitcom’s pilot episode ‘Cartman Gets an Anal Probe’ in the typical 10pm ET timeslot.
Episode 2 was delayed after the show’s controversial premiere, which featured the likes of a deepfaked Donald Trump with no clothes on and ICE (Immigration and Customs Enforcement) agents shooting puppies and driving around ominously in large groups. Screenshots from the show were quickly used on both the White House and US Secretary of Homeland Security Kristi Noem’s social accounts in order to try and recruit more people to join ICE.
The Department of Homeland Security (NHS) additionally told Newsweek: “We want to thank South Park for drawing attention to ICE law enforcement recruitment. We are calling on patriotic Americans to help us remove murderers, gang members, pedophiles, and other violent criminals from our country.”
It’s for this reason that I can’t help but think the delay of South Park season 27 episode 3 has something more attached to it. It’s certainly a conveniently timed break in the schedule, but given that season 27 has only been on for three episodes yet has suffered two delays already. Surely things aren’t operating smoothly behind the scenes? We always know what we get into with the hit TV show, but that might no longer be to everybody’s tastes.
Has South Park season 27 hit a political snag after its season 3 delay? I wouldn’t be surprisedKristi Noem's parody in South Park season 27. (Image credit: Paramount)I don’t need to spell out that the political and cultural landscape has remarkably changed in the last few years, and that’s possibly to South Park’s detriment. While real-world political and cultural affairs are offering a smorgasbord of inspiration, it also comes with a more critical lens. We’ve already had the comments from the NHS, above, with Kristi Noem also weighing in on her portrayal as well. Before her secret post-credits scene was revealed, she responded during an interview on the Glenn Beck Program podcast, slamming her character’s appearance: “It’s so lazy to just constantly make fun of women for how they look. If they wanted to criticize my job, go ahead and do that. But clearly they can’t – they just pick something petty like that.”
White House spokesperson Taylor Rogers also told Variety about the events of episode 1: "Just like the creators of South Park, the Left has no authentic or original content, which is why their popularity continues to hit record lows. This show hasn't been relevant for over 20 years and is hanging on by a thread with uninspired ideas in a desperate attempt for attention."
This won’t be the last round of comments from political figures about what’s happening on the show, and I wouldn’t be surprised if we see public conversations between the two sides on a weekly basis. I also wouldn’t be surprised if there’s resistance to what’s being shown behind the scenes, given nobody exactly comes across particularly well if they’re being featured on South Park. We’ve seen critics of the current US Government coming under fire or having their shows canceled (The Late Show with Stephen Colbert’s cancellation is a great example of this), so discussions about what should and shouldn’t be parodied could be holding up episode releases.
Of course, this is all speculation. We’ve got no idea what the rest of South Park season 27’s 10-episode run has in store for us, and I really hope Paramount and the show’s creators can hold firm to delivering the creative vision they want to. I can’t deny that I’m worried about episodes 4-10, but I’m also holding out hope. After all, South Park wouldn’t be South Park if it wasn’t near the knuckle and beyond, right?
You might also like- Beth Dutton and Rip Wheeler spin-off: Everything we know about the sequel series to Yellowstone on Paramount+
- Mayor of Kingstown season 4: everything we know so far about the hit Paramount+ show's return
- New Paramount+ series Avatar: Seven Havens will succeed where the Last Airbender live-action adaptation couldn't, and that should make Netflix panic
Geoffrey Hinton, scientist, former Google employee, and widely recognized 'Godfather of AI,' has made a late-stage career of criticizing his godchildren. And now he's taken it all a step further, insisting we need "AI Mothers," not AI Assistants.
Speaking at the AI4 Conference in Las Vegas this week, and as first reported by Forbes, Hinton again sounded the alarm on the impending advent of Artificial General Intelligence, which he now believes will arrive in a few years, a notion that syncs with recent comments from OpenAI CEO Sam Altman.
That acceleration from what was once thought to be decades to a few orbits around the sun is, perhaps, what prompted Hinton to argue that we need something other than AI Assistants.
"We need AI mothers rather than AI assistants," Hinton said, according to Forbes. The idea, Hinton posits, is that AI's with "maternal instincts" are a sort of protection system. After all, mothers generally don't harm and usually protect their children.
If AI systems like ChatGPT, Claude AI, and Gemini truly become smarter than us in a matter of years, having them in some way feel as if it's their job to look out for us might prevent them from harming us or society.
Hinton, who recently won a Nobel Prize and helped develop the technological foundation that arguably made all this AI possible, left Google in 2023 and immediately started warning people about a dire AI future. Imagine a parent disowning their child, and you get the idea.
I don't think Hinton is turned off from AI. After all, he can't stop talking about it, and appears to recognize its potential, but it's also clear it scares him.
He previously told The New York Times in 2023 that
- Increased competition is leading to a less cautious approach
- He presciently warned about the flood of fake online content
- He warned about AI taking on all the jobs we don't want to do (it may be taking others, as well)
- He worried about AI that can both program and then run programming (a very dangerous closed loop)
- And he was talking almost immediately about AI outsmarting us
So, sure that day is now fast approaching, but is a motherly AI what we want or need? I don't think so.
The minute we start training "Mom Instincts" into AI, it will start to act like a mother and slip into that creepy, uncanny valley where you can no longer tell if you're talking to a program or a person. Motherly instincts imply warmth, compassion, caring, understanding, and love. I don't want those things from an AI.
What I think we need, though, is for AI assistants to understand what it means to be human. Put another way, if AI chatbots can at least understand humanity, they can serve us better. They can also recognize our propensity for trust and perhaps finally stop presenting us with false narratives and fake friendliness and interest.
We shouldn't want companionship out of our super-intelligent AI systems. Instead, we need utility and trust, an ability to carry out our wishes in a way that best serves our interests.
The last thing we need is an AI full of maternal instincts, which then makes its own choices and, when things go awry, insists, "Well, dear, mother knows best."
You might also like- Finance-themed phishing uses personalized subjects and file names to deliver malware
- Travel and response phishing also use personalization to push information stealers and RATs
- Cofense urges verification of unexpected emails and updates to security tools
Attackers are increasingly personalizing phishing emails to deliver malware, experts have warned, with criminals reaping in huge gains.
In adding the recipient’s name, company and other details into subject lines, file names, and message content, threat actors seek to make the messages appear more legitimate, increasing the chances that recipients will open malicious attachments or click links, researchers at Cofense have revealed.
Cofense analyzed a year’s worth of data and found that while several campaign themes use this tactic, finance-themed phishing was the most worrying due to both its frequency and impact.
Stay safeNearly 22% of subject-redacted emails fell into this category, often posing as invoices, tenders, or payment summaries.
Many of these emails carried jRAT, a cross-platform remote access trojan that can give attackers full control of a system, steal files, and install more malware.
Finance-themed phishing is particularly effective because it blends seamlessly with normal workplace communication, as employees often expect emails about contracts or payment updates.
While finance-themed phishing accounted for 21.9% of personalized subject cases, other themes also made heavy use of this approach.
Travel Assistance was the largest category at 36.78%, often used to deliver Vidar Stealer under the guise of reservation or itinerary updates.
Response-themed emails followed at 30.58%, frequently carrying PikaBot in messages disguised as meeting cancellations or order confirmations.
Tax-themed campaigns made up 3.72%, commonly involving Remcos RAT in password-protected archives, while Notification-themed phishing also represented 3.72%, delivering various malware families including WSH RAT and jRAT.
To counter these threats, Cofense advises verifying unexpected email requests through trusted channels, keeping antivirus and malware removal tools up to date, and limiting public exposure of staff details to make targeting harder.
Summing up, Cofense says, “While customized subject lines are not used in all malware email samples, it is a strong tactic to make the recipient feel a higher sense of urgency that may lead to a successful infection. Particularly targeted emails delivering RATs or Information Stealers can be notable for potentially providing remote access or login credentials that can be brokered to ransomware threat actors.”
You might also like- Check out these 5 ways to improve email security
- And look at our round up of the best secure email providers
- Fragmented security: the hidden threat undermining your cyber defenses
LG and Panasonic are just two of the top brands in the OLED TV market, but unlike rivals such as Samsung, Sony and Philips, their TVs often share a lot in common, right down to using the same OLED display panels.
The LG G4 and Panasonic Z95A, two of the best OLED TVs launched in 2024, both used micro-lens-array (MLA) OLED panels manufactured by LG. In 2025, it’s a similar story as the LG G5 and Panasonic Z95B both use LG’s new Primary Tandem RGB OLED panel (also known as four-stack OLED).
That means you should expect a similar picture from both, right?
As I’ve discovered when testing combinations of some of the best TVs, models that use the same display panel don’t always look the same. For example, the Samsung S95F and Sony Bravia 8 II flagship OLED TVs use Samsung’s QD-OLED panels, but when I tested them side by side, I got very different results.
The LG G5 and Panasonic Z95B were featured in our recent OLED four-way showdown with a casual judging panel, with the aforementioned S95F and Bravia 8 II rounding out the group, but I thought it would be illuminating to compare the LG and Panasonic on their own.
Brightness Image 1 of 2In default Filmmaker Mode, the LG G5 (right) clearly has the brightness advantage over the Panasonic Z95B (right) (Image credit: Future)Image 2 of 2But adjust the Z95B's brightness, and it gets much closer to the G5 (Image credit: Future)With both the Z95B and G5's default Filmmaker Mode settings active, it was clear that the G5 had a huge brightness advantage – an odd result considering both use the same OLED panel. When viewing some demo footage from the Spears & Munsil UHD Benchmark 4K Blu-ray in HDR10 format, scenes with snow looked far brighter and more vibrant on the G5, with much more brilliant whites and highlights. The Z95B’s picture still looked very accurate, but dim in comparison.
Leaving both sets in Filmmaker Mode, I measured the Z95B’s peak HDR brightness on a 10% white window pattern at 1,028 nits, a significantly lower result than the G5’s 2,268 nits on the same pattern.
After a quick check of the Z95B’s picture settings, I found its Luminance Level (brightness) was set to 70/100 by default, while the G5’s was set to 100/100. After upping the Z95B’s brightness to 100, it hit 2,355 nits, making it the brightest OLED I’d measured to date on this test.
Going back to the same Spears & Munsil footage with the brightness on the Z95B set to 100, the snow scenes now showed much more dazzling whites. It looked very similar to the G5, with the only real difference being the color temperature. I did, however, feel that despite the boost in brightness, the Z95B lost a little of the picture accuracy I’d noted previously.
Color and contrast With Dolby Vision sources, such as Wicked (pictured) the Panasonic Z95B (left) and LG G5 (right) have very similar colors. (Image credit: Universal Pictures / Future )Both the Z95B and G5 had superb color in my comparison. Watching more Spears & Munsil footage in HDR10 format, a parrot’s yellow and green feathers looked vivid on both TVs, and a field of red flowers was crisp and refined, with plenty of punch. Once again, the G5’s brightness in default settings gave colors more pop, but upping the Z95B’s brightness brought it to a similar level of vibrancy. The Z95B demonstrated richer, deeper colors, though, thanks to its stronger contrast and more refined black levels.
Switching to Wicked on 4K Blu-ray in Dolby Vision, the default brightness settings were 100/100 in both the Dolby Vision Dark mode on the Z95B and Dolby Vision Filmmaker Mode on the G5. This made the colors on both TVs look very similar. Elphaba’s green skin and the pink flowers and blue details on a wall in the Wizard & I scene both had the same eye-popping, vibrant color, and they also looked true-to-life.
Where the TVs differed was that the Z95B's deeper black levels made colors appear bolder and more detailed, whereas they looked brighter and punchier on the G5. This was easy to see in Glinda’s pink outfits and the greens of the Emerald City in Wicked.
Both the Panasonic Z95B (left) and LG G5 (right) deliver excellent contrast and black levels, but the Z95B looked more accurate on The Batman (Image credit: Warner Bros. / Future)As you’d expect from two top-tier OLED TVs, both black levels and contrast are excellent. Watching Alien: Romulus in Dolby Vision on 4K Blu-ray, shots of space or dark tunnels within the ship showcased rich black levels on both TVs, with excellent contrast between dark shadows and bright highlights from stars, lights and torches. The Z95B had the edge of the two TVs here with its deeper blacks, but again, both looked great.
Switching to The Batman in Dolby Vision on 4K Blu-ray, I used the opening crime scene section I regularly use for testing contrast. For this movie, I had to view in dimmed or pitch black conditions as both the Z95B and G5 struggled with the overhead lighting in our testing lab.
Interestingly, although both showcased excellent black levels and contrast, with the torches carried by detectives balancing well against the dim surroundings, I noticed that in pitch black conditions, black levels were more elevated on the G5. The Z95B’s deeper blacks resulted in stronger contrast, although there was some loss of shadow detail, specifically Batman’s eye and the logo on his chest when he looked towards the camera. Of the two, I found myself drawn to the Z95B as it felt more accurate to the movie.
Bang for your buck
Both the Z95B (left) and G5 (right) are brilliant OLED TVs, but your choice may come down to personal preference and price. (Image credit: Future)It’s fair to say that in recent years, a drawback of Panasonic’s flagship OLED TVs has been their price, which is often hundreds more than that of its main rivals at launch. This year, however, Panasonic is being more aggressive with its pricing. The 55-inch Z95B is available for $2,399 / £2,299, while the 55-inch G5 is available for $2,199 / £2,299. The 65-inch Z95B, meanwhile, is available for $3,099 / £2799, while the 65-inch G5 is $2,899 / £2.899.
One thing that could swing things in the Z95B’s favor between these two TVs is sound. Panasonic’s TVs are always among the best TVs for sound, and the Z95B is no different.
Watching the Batmobile chase scene from The Batman, the Z95B stood head and shoulders above the G5 when it came to the built-in sound. The Z95B’s punchier sound and much more powerful bass better captured the rumble of the Batmobile’s engine and the crunching of car impacts and explosions. Speech was clear on both sets, but the Z95B delivered better clarity. With the Z95B, you can easily watch without using one of the best soundbars – a factor that saves you money – whereas a soundbar is more recommended with the G5.
Where the G5 has the edge is in its gaming features, which are up there with the best gaming TVs and include 4K 165Hz support on four HDMI 2.1 ports. The LG’s webOS 25 smart TV interface is also one of the best I’ve used and provides a superior overall experience to the Z95B’s Fire TV smart interface.
Both of these are class-leading OLED TVs, and both are in the conversation for TV of the year. But based on this comparison, I found myself favoring the Panasonic Z95B. Its default Filmmaker Mode brightness setting may be odd, but once adjusted, I found myself more drawn to its picture than the LG G5's picture.
You might also like- I tested LG, Samsung and Sony's elite 2025 OLED TVs side-by-side – here's the one I'd buy with my own money
- I tested the Samsung S95F and LG G5 OLED TVs side-by-side: here's which one you should get
- I tested Samsung and LG's cheapest OLED TVs side-by-side to see which TV comes out on top – here's what happened
- Dahua CCTV flaws identified by Bitdefender affect over 100 popular security camera models
- Vulnerabilities allow remote code execution without authentication over local or internet connections
- Company urges firmware updates and network isolation to prevent exploitation
Researchers at Bitdefender have announced two critical vulnerabilities affecting a large number of Dahua smart cameras.
The flaws, which were patched in the most recent firmware update, could allow unauthenticated attackers to take full control of affected devices.
Dahua has confirmed that a total of 126 models were affected, including multiple IPC, SD, and DH series devices, not just the Hero C1 model first reported.
Patch nowThe first of the vulnerabilities, CVE-2025-31700, is a buffer overflow flaw in Dahua camera firmware that can be triggered when the device processes specially crafted network packets. If exploited, it could cause the camera to crash or, in some cases, allow a remote attacker to run their own code on the device.
The second, CVE-2025-31701, is another buffer overflow issue also exploitable through maliciously crafted packets sent over the network. It too can be used to crash the camera or potentially gain full remote control depending on the target’s defenses.
Both can be exploited to run arbitrary code with root privileges.
Bitdefender privately reported the issues to Dahua on March 28, 2025. The Chinese video surveillance equipment manufacturer acknowledged the report the next day and validated the findings by April 1.
It requested some time to prepare a fix for the issues, with patches finally rolling out last month, followed by the agreed public disclosure.
The two vulnerabilities can be especially dangerous for devices accessible from the internet via port forwarding or UPnP, as no authentication is required for possible exploitation.
Bitdefender warns that successful attacks could bypass firmware integrity checks and deploy persistent malicious code, making cleanup difficult.
Dahua, the world’s second-largest CCTV manufacturer behind Hikvision, has faced scrutiny in several countries over cybersecurity issues and data privacy concerns, particularly related to potential vulnerabilities in its network-connected devices.
It maintains a Product Security Incident Response Team (PSIRT) to coordinate with researchers on reported flaws, such as in the case of these vulnerability disclosures.
It is urging all customers who have not yet done so to update their camera firmware as a matter of urgency.
For anyone unable to do so immediately, it advises disconnecting vulnerable devices from direct internet access, disabling UPnP, and isolating cameras on separate networks to reduce risk.
A detailed list of affected models is included in Dahua’s online advisory, along with links to patched firmware.
Both Dahua and Bitdefender stress unpatched internet-connected devices should be considered prime targets.
You might also like