TechRadar News

Looking for a different day?

A new NYT Connections puzzle appears at midnight each day for your time zone – which means that some people are always playing 'today's game' while others are playing 'yesterday's'. If you're looking for Monday's puzzle instead then click here: NYT Connections hints and answers for Monday, October 12 (game #1219).

Good morning! Let's play Connections, the NYT's clever word game that challenges you to group answers in various categories. It can be tough, so read on if you need Connections hints.

What should you do once you've finished? Why, play some more word games of course. I've also got daily Strands hints and answers and Quordle hints and answers articles if you need help for those too, while Marc's Wordle today page covers the original viral word game.

SPOILER WARNING: Information about NYT Connections today is below, so don't read on if you don't want to know the answers.

NYT Connections today (game #1220) - today's words

(Image credit: New York Times)

Today's NYT Connections words are…

• EYE
• LASH
• LIPS
• JACKET
• TIE
• SENSE
• CHANGE
• BELT
• RING
• TASTE
• CUP
• HITCH
• CANNON
• TETHER
• FLAIR
• ENDS

NYT Connections today (game #1220) - hint #1 - group hints

What are some clues for today's NYT Connections groups?

• YELLOW: Very particular
• GREEN: Put together
• BLUE: What winners get
• PURPLE: Begin with the opposite of tight

Need more clues?

We're firmly in spoiler territory now, but read on if you want to know what the four theme answers are for today's NYT Connections puzzles…

NYT Connections today (game #1220) - hint #2 - group answers

What are the answers for today's NYT Connections groups?

• YELLOW: DISCERNMENT
• GREEN: FASTEN
• BLUE: SPORTS CHAMPIONSHIP PRIZES
• PURPLE: LOOSE_____

Right, the answers are below, so DO NOT SCROLL ANY FURTHER IF YOU DON'T WANT TO SEE THEM.

NYT Connections today (game #1220) - the answers

(Image credit: New York Times)

The answers to today's Connections, game #1220, are…

• YELLOW: DISCERNMENT: EYE, FLAIR, SENSE, TASTE
• GREEN: FASTEN: HITCH, LASH, TETHER, TIE
• BLUE: SPORTS CHAMPIONSHIP PRIZES: BELT, CUP, JACKET, RING
• PURPLE: LOOSE_____: CANNON, CHANGE, ENDS, LIPS

• My rating: Hard
• My score: 1 mistake

Seeing LIPS, LASH and EYE on the top line of the grid immediately made me think about makeup, but I resisted the temptation to cobble a group together and instead waited to see if I could find anything more concrete.

I stumbled on a Purple First after trying to think about what possible group could contain CANNON and could only come up with weaponry and the phrase “loose cannon”, which chimed with CHANGE, ENDS, and LIPS.

Sadly I ruined my clean run with eight tiles left, as I thought we were looking for ways of strapping items of clothing, so I had BELT instead of TETHER in the group that became FASTEN.

Yesterday's NYT Connections answers (Monday, October 12, 2026, game #1219)

• YELLOW: STATUS: CONDITION, SHAPE, SITUATION, STATE
• GREEN: TOLERATE: BEAR, HANDLE, STAND, WEATHER
• BLUE: WORD SHORTENINGS, OFTEN: ABBREVIATION, CONTRACTION, ELISION, NICKNAME
• PURPLE: WORDS MADE USING U.S. POSTAL ABBREVIATIONS: ALKYLAMIDE, CALAMARI, MANDARIN, MELAMINE

What is NYT Connections?

NYT Connections is one of several increasingly popular word games made by the New York Times. It challenges you to find groups of four items that share something in common, and each group has a different difficulty level: green is easy, yellow a little harder, blue often quite tough and purple usually very difficult.

On the plus side, you don't technically need to solve the final one, as you'll be able to answer that one by a process of elimination. What's more, you can make up to four mistakes, which gives you a little bit of breathing room.

It's a little more involved than something like Wordle, however, and there are plenty of opportunities for the game to trip you up with tricks. For instance, watch out for homophones and other word games that could disguise the answers.

It's playable for free via the NYT Games site on desktop or mobile.

Looking for a different day?

A new NYT Strands puzzle appears at midnight each day for your time zone – which means that some people are always playing 'today's game' while others are playing 'yesterday's'. If you're looking for Monday's puzzle instead then click here: NYT Strands hints and answers for Monday, October 12 (game #953).

Strands is the NYT's latest word game after the likes of Wordle, Spelling Bee and Connections – and it's great fun. It can be difficult, though, so read on for my Strands hints.

Want more word-based fun? Then check out my NYT Connections today and Quordle today pages for hints and answers for those games, and Marc's Wordle today page for the original viral word game.

SPOILER WARNING: Information about NYT Strands today is below, so don't read on if you don't want to know the answers.

NYT Strands today (game #954) - hint #1 - today's themeWhat is the theme of today's NYT Strands?

• Today's NYT Strands theme is… Well, this is awkward

NYT Strands today (game #954) - hint #2 - clue words

Play any of these words to unlock the in-game hints system.

• SOME
• BREWS
• PART
• PATH
• FROM
• FROGS

NYT Strands today (game #954) - hint #3 - spangram lettersHow many letters are in today's spangram?

• Spangram has 12 letters

NYT Strands today (game #954) - hint #4 - spangram positionWhat are two sides of the board that today's spangram touches?

• First side: left, 1st row

• Last side: right, 7th row

Right, the answers are below, so DO NOT SCROLL ANY FURTHER IF YOU DON'T WANT TO SEE THEM.

NYT Strands today (game #954) - the answers

(Image credit: New York Times)

The answers to today's Strands, game #954, are…

• EMBARRASS
• APPALL
• NONPLUS
• CHAGRIN
• MORTIFY
• SPANGRAM: CRINGEWORTHY

• My rating: Hard
• My score: Perfect

I’m classing this as tricky because despite getting by without hints it still took me longer than usual to complete the board.

After getting the spangram fairly quickly I was confident I’d roar through the board at speed, chuckling at memories of CRINGEWORTHY scenes from Something About Mary and Meet The Parents.

Alas, after getting EMBARRASS I struggled to think of any other synonyms for awkward; I thought NONPLUS meant you were indifferent. Ah well, every day’s a school day.

Yesterday's NYT Strands answers (Monday, October 12, game #953)

• PARLANCE
• JARGON
• VERNACULAR
• LINGO
• DIALECT
• SPANGRAM: TURNOFPHRASE

What is NYT Strands?

Strands is the NYT's not-so-new-any-more word game, following Wordle and Connections. It's now a fully fledged member of the NYT's games stable that has been running for a year and which can be played on the NYT Games site on desktop or mobile.

I've got a full guide to how to play NYT Strands, complete with tips for solving it, so check that out if you're struggling to beat it each day.

• Kioxia's upgraded GP drive exceeds 20 million random-read IOPS
• Third-gen XL-FLASH improves memory access through shorter wiring and additional planes
• Kioxia LD4 launches with 15.36 TB and 30.72 TB capacities

Kioxia has announced a faster edition of its GP series drive that exceeds 20 million IOPS when reading 512-byte data blocks at random.

The GP1 unveiled in August 2026 delivered about 10 million IOPS, so the new version roughly doubles the speed of the earlier model.

The faster model relies on third-generation XL-FLASH memory, which delivers three times the reading speed and roughly 150% higher writing speed than before.

How flash is meant to rival DRAM

The company is pursuing a plan to take on DRAM in AI markets by offering flash storage that approaches memory-class speed.

“Kioxia is making game-changing high-performance flash storage solutions for AI infrastructure,” said Neville Ichhaporia, SVP and GM of the SD business unit at Kioxia America.

“As AI drives new performance demands, scalable, efficient storage becomes the foundation for innovation.”

Its XL-FLASH technology is designed to deliver higher operation rates, finer 512-byte access and lower power per operation than conventional SSDs.

The third-generation XL-FLASH shortens its word and bit wiring to cut RC delay and fits more planes in each die than the 16 of Gen 2.

Its efficiency is forecast to rise roughly 40% for reads, and 80% for writes, yet energy used by each operation is expected to keep climbing.

The first-generation GP1 uses MLC cells storing 2 bits, a faster SLC mode, a 16-plane layout, and read latency below 5 microseconds.

It is a PCIe 6.0 NVMe 2.2 drive meant for GPUs to reach storage directly, with endurance rated at up to 50 DWPD.

Kioxia has not disclosed the upgraded model's endurance rating or confirmed whether it keeps the PCIe 6.0 interface, but it will most likely follow the path of its predecessor.

LD4 arrives without performance figures

Kioxia also exhibited the LD4, a QLC drive with one port that stores 4 bits per cell in E1.L format using a four-lane PCIe 5 link.

It comes in 15.36 TB and 30.72 TB capacities, though its design can scale to 122.88 TB for hyperscalers, which is still modest compared to the Kioxia LC9’s 246 TB capacity.

The 15.36 TB model fills a quarter of the E1.L enclosure, with a 61.44 TB version likely at the three-quarter point.

The LD4 uses 218-layer BiCS 8 flash and appears to succeed the October 2023 LD2, which ran on PCIe 3 with identical capacities.

It also supports NVMe 2.0, OCP specification 2.6, and full-path safeguards against data loss.

Kioxia skipped an LD3, so the LD4 follows the LD2 after roughly three years and swaps in a faster PCIe 5 bus.

Rival E1.L drives with a PCIe 5 interface include the 6550 ION from Micron, the R6060 from DapuStor and the Echo from FADU.

As of the time of writing, the company has not released the speed or durability figures for the LD4.

Via Blocksandfiles

The future of AI lies squarely in the physical realm, according to many industry figures such as Nvidia CEO Jensen Huang. Much of the evolution of AI in recent years has, as expected, been in aid of building machines that can learn quicker, think faster, and interact with the physical world in a more compelling way.

Checkmate

Nearly 40 years ago, the scientist Rodney Brooks authored a significant robotics paper that argued real-world robotics should focus on simpler, layered behavioral systems to understand the world rather than complex internal models.

Quote of the day

This article is part of TechRadar Pro's QOTD project to provide an insight into the minds of the brightest and most recognized figures in the technology industry today and in years gone by. Read the full series here.

His comments in the study 'Elephants Don't Play Chess' signalled a major paradigm shift in the way scientists saw the training and learning process for robots then and in the future. It challenged the established status quo for how scientists considered physical AI, and proved fairly divisive at the time.

The old sense-model-act-plan (SMPA) approach to AI – in which robots would scan a room to build a 3D digital map inside their memory and plan the optimal path within that map – was out. Instead, Brooks proposed that when a robot changes the world around itself, it doesn't need to compute the impact of everything – just the most relevant parts.

One man saw it coming

Brooks co-founded the company now known as iRobot alongside two colleagues from MIT's AI lab in the same year as he published his seminal paper, although back then it was called IS Robotics.

The first product was Geghis, an insect-like space explorer that weighed 1kg and served as a prototype for low-cost space exploration. Then, more than 10 years later in 2002, the company launched its first Roomba robot vacuum cleaner.

Despite enjoying success through the decades, iRobot filed for bankruptcy in December 2025 and the company was eventually taken private and absorbed into the operations of Chinese company Shenzhen Picea Robotics.

• Blizzard Entertainment says it learned plenty of lessons from World of Warcraft Classic: Season of Discovery
• It acknowledged that "we were going too far" with class changes in Season of Discovery and "Forever will be a lot more careful and cautious with slower class changes"
• The studio also learned "what levers we could create to help us manage the in-game economy" in Forever

Blizzard Entertainment has learned a lot from World of Warcraft Classic: Season of Discovery, but it's taking care not to overindulge in class experimentation for World of Warcraft: Forever.

In the same lengthy interview with TechRadar Gaming, where Blizzard confirmed it doesn't plan to modernize Forever's classes, as it wants to preserve the Classic experience, the team shared what it learned from Season of Discovery, a seasonal release for Classic that reimagines progression systems, adds new abilities, unconventional class roles, and redesigned instances, and how it will apply those lessons to Forever.

The team said that it had fun exploring class changes, but admitted it went "too far" and is taking steps to be more considerate with how it applies updates in Forever.

"We pushed every class to the limit and explored every power fantasy on every class," Blizzard said. "But we knew we were going too far. It’s like eating all your Halloween candy on October 31. It’s fun and amazing, but afterward you are sick to your stomach.

"Forever will be a lot more careful and cautious with slower class changes."

Another lesson it learned from Season of Discovery was "what levers we could create to help us manage the in-game economy," something that sometimes suffers from botting and real-money trading (RMT).

It told us that it's actively working to combat RMT in Forever and has already devised penalties that, depending on the terms of service infraction, could result in permanent bans.

Blizzard continued, saying, "We built several systems both under the hood and player-facing that allowed us to control the inflow and outflow of key resources, which led to a much healthier economy, with much less problematic behavior such as RMT."

RMT has always been a huge issue in WoW, whether it's retail or Classic, and Blizzard confirmed it has "a number of new systems in place" to support detection.

"It’s a cat and mouse game, so talking too much about our plans here could tip our hand," it said. "Suffice to say, we are watching very closely, and engaging in RMT is not worth putting your account at risk."

Blizzard has also addressed WoW: Forever's controversial Skyborne race and explained why the Elf race is available to both Alliance and Horde. It's also shared why Summoning Stones won't be in the upcoming game.

• US Army awarded $4.15 billion in counter-drone contracts to 10 companies
• Seven of the ten winners each received a $500 million base award
• The framework carries a maximum ceiling of $7 billion across all awards

The US Army has awarded contracts worth $4.15 billion to 10 companies developing technologies against cheap drones threatening military installations and critical infrastructure across the country.

The awards fall under Domestic Shield, a Department of Defense initiative intended to strengthen protection against unmanned aircraft without restricting procurement to individual suppliers.

Seven companies will receive $500 million each, while three others secured smaller awards, establishing a multi-vendor arrangement with a maximum value of $7 billion.

Ten suppliers share billions in counter-drone contracts

Allen Control Systems, Digital Force Technologies, DroneShield, L3Harris WESCAM, Napatree Technology, RADA Technologies and SRC each received contracts with $500 million base values under the initiative.

Echodyne and PVP Advanced EO Systems secured awards worth $250 million apiece, while Smart Shooter received a contract valued at $150 million.

The contracts allow military services, federal agencies and allied partners to obtain commercially available counter-drone equipment through a common procurement framework rather than relying on separate purchasing arrangements.

Officials expect further awards below the established contract threshold before the end of October 2026, keeping additional suppliers eligible to participate as procurement continues.

The Army is working with Joint Interagency Task Force 401, the Pentagon unit responsible for coordinating efforts against unmanned aerial threats, to administer the initiative.

The arrangement uses multiple suppliers to maintain competition and reduce dependence on technologies controlled by individual vendors.

Open architecture allows different counter-drone systems to work together

The contracted technologies include robotic weapon stations, optical rifle sights, compact radars, long-range tracking cameras, and command-and-control software.

These systems handle different tasks, including detecting aerial threats, tracking their movements and supporting responses, making coordination between equipment an important consideration.

To support that coordination, the Army is requiring open architecture standards that allow equipment from different suppliers to connect with existing defense networks.

“This multi-vendor marketplace enforces strict open architecture standards, breaking proprietary vendor lock, and ensuring new technologies can immediately integrate into our existing defense networks,” said Brent Ingraham, assistant secretary of the Army for Acquisition, Logistics and Technology.

The need for these systems became more apparent during the 2026 US–Iran war, when Iranian drone attacks damaged American military facilities and equipment across the Middle East.

On March 3 2026, two Iranian drones struck the US Embassy in Riyadh, Saudi Arabia, causing physical damage estimated at approximately $11.5 million.

Iranian drones also struck a fuel tank and damaged a US Army asset in Kuwait, while attacks against power stations and desalination plants continued after the ceasefire.

In Saudi Arabia, Iranian drones struck a CIA station in Riyadh, while attacks within a week left 29 American servicemen wounded.

Congressional Research Service estimated that 81 American aircraft and drones were destroyed or damaged during the conflict, with potential losses reaching $3.3 billion.

Iran carried out many of these attacks with swarms of cheap drones and bombs, necessitating affordable counter-drone systems that can intercept incoming threats without exhausting expensive military assets.

Via The Drone Front

There’s little else quite so satisfying in gaming as a really chunky parry sound effect. When you read your opponent precisely and time a deflection to perfection, hearing that impactful audio cue as a reward is about as good as it gets — especially if you can follow it up with a devastating finishing move. Valor Mortis nails this.

I never tired of hearing the reverberated twang of blades clashing in my time with the first-person ‘Souls-like’ from One More Level. And given how much time is spent duelling with the reanimated corpses of Napoleonic soldiers, body horror monstrosities, and a few unique creatures I won’t spoil, it’s really something the developers had to nail.

So, I’m pleased to say that they have, and it’s one reason why I stuck with Valor Mortis through several of its other less successful aspects.

Review info

Platform reviewed: PC
Available on: PS5, Xbox Series X|S, PC
Release date: October 13, 2026

The opening, however, is strong. You awaken on a battlefield, surrounded by corpses, the rousing voice of Vincent Cassel as Napoleon in your ear encouraging you to move forward. But this conflict has turned into something far worse than just blades and cannon fire. A plague of sorts has torn through the land, transforming humans into monsters and unlocking mystical powers for you to use in your quest to understand exactly what’s happened while you’ve slumbered.

It’s all a setup that fits the Dark Souls-inspired core of the game, with its dangerous, foreboding, war-torn world to explore. This time, lanterns serve as the safe spaces to save your progress, level up using the blood from dead enemies, tweak your build with several skill-tweaking gems, and restart from when you inevitably die.

Your power to fight back against the horrors builds slowly and doesn’t ever get massively expansive. First, a sabre. Then, a pistol. After that, well, it’s the ability to shoot fire from your hands and put enemies in stasis. You can switch between these and a handful more melee, ranged, and magical powers at will — and will often need to do so to tackle different enemy types most effectively.

Bonjour

(Image credit: Future / One More Level)

For example, basic soldiers can be taken out with a few sword slashes. A more imposing axe-wielding officer is better handled by keeping your distance as you burn it with fire to set off the explosive keg on its back. There’s a malformed grenadier who protects its sack of explosives with an engorged hand, but when it moves to lob a grenade at you, a well-timed shot can cause a massive explosion.

As you proceed forward, each encounter requires you to assess whichever threats you’re facing and break down the best way to take them out or exploit a weakness with the tools at your disposal. Of course, you can always go wailing in with a sword, but it’s rarely the smartest move. You have limited stamina, and the aggressive enemies can obliterate you in just a few hits.

As is the trial-and-error nature of these types of games, it takes a few encounters with some new foes to get the better of them, but I wish Valor Mortis wasn’t so obsessed with the trickery of it all. Long delayed attacks that are preceded by multiple feints and five seconds of interpretive dance before a sudden blow pulverises you is kind of funny the first couple of times, but by the dozenth time I was getting irate.

(Image credit: Future / One More Level)

The same goes for some aspects of the level design, which is regularly full of traps and “gotcha” moments. Enemies pop up out of bushes, others surprise you from behind doors, items are placed for you to be ambushed, or trigger rocks falling on you, or make you step on sections of floor that fall out from under you. I get that these moments are meant to keep you on your toes, but when they happen repeatedly in quick succession, they start to feel cheap rather than challenging.

Best bit

(Image credit: Future / One More Level)

There are a couple of boss fights that I could call out here that feel awesome to emerge from victorious because you perfectly parry all attacks to deliver a satisfying riposte or are wowed by a piece of clever inventiveness, but I wouldn’t want to spoil the specifics. Instead, I’ll keep it vague and just say that when a set piece cleverly mixes combat, puzzle, and platforming elements well, that’s when Valor Mortis hits its peak.

What I do like is that at least Valor Mortis mixes up linear gauntlets with open areas that give you more freedom in how you approach. There are opportunities to sneak up or around to get better vantage points and backstab enemies for an instant kill or significant damage. Thinning out larger groups is definitely something you’ll need to do, as even facing just two enemies at once puts you at a massive disadvantage in the game’s first-person view.

Variety is added further with some fantastic sections that combine the frenetic platforming of the team’s previous Ghostrunner series with the tight combat mechanics to create some entertaining and inventive set pieces. A section across broken-up ships that are suspended in time above the sea is a highlight, as you hop and swing between the splintered decks, take out snipers with your musket, and slash through any remaining soldiers prowling around.

Finding its footing

(Image credit: Future / One More Level)

On the other hand, a diversion into a mine is perhaps one of the game’s most dreary and aggravating sections. Here, lots of tight, dark, and windy corridors are not the easiest to see or fight in. Then, while you’re also trying to dodge and parry different enemies, the game decides to throw freewheeling minecarts that can almost kill you in one hit down these same small alcoves at the same time. A fun idea, in theory, but there’s a point where a situation becomes too chaotic for you to assess it and react to it quickly enough, and so my approach changed to brute force and luck rather than skill.

But then I was quickly brought back again by another thrilling on-rails segment that makes full use of all abilities and weapons you’ve gathered until now, before culminating in a surprising out-of-this-world boss fight that puts the rewarding parrying mechanics front and centre.

When that works, it really works. And the flourishes One More Level adds with ranged weapons and magical abilities are welcome tactical options — even if you will have to rely on swordplay more than anything else. The fact I kept getting beat up but wanted to keep coming back to nail a parry combo shows where the strengths of Valor Mortis lie. It’ll likely scare many away, but if challenging Souls-likes are what you enjoy, here’s another with an intriguing premise and a neat twist to add to your list.

Should you play Valor Mortis?

(Image credit: One More Level)Play it if…

You’re looking for your next Souls-like
As far as Souls-likes go, Valor Mortis is a good one, especially if you prefer the parry-focused combat seen in the likes of Lies of P and Sekiro.

You enjoyed Ghostrunner but wanted more involved combat
Valor Mortis is far from the speedy pace of the Ghostrunner series, but some of the developer’s inventive platformer and set piece design is in here, just with more emphasis on combat.

You’re a fan of horror
The world, environments, and monsters of Valor Mortis are tinged with horror, while some of the side stories you come across are equally unsettling or downright devastating.

Don’t play it if…

You get frustrated repeating encounters
You’d probably expect this based on the genre alone, but if quick deaths and repeating encounters until you master them are not what you enjoy, then Valor Mortis is best avoided.

You like a lot of weapons and build variety
A trio of swords, a trio of guns, and a few magical skills are all you really get in Valor Mortis. These all work fine and have different strengths and weaknesses, but it doesn’t make for a whole lot of variety in the long run.

Accessibility features

Valor Mortis offers very few accessibility features. In terms of dialogue, there is only an option to adjust the subtitle size between small, medium, and large. Mouse and keyboard, and controller options are available, with full key rebinding possible on the former and several control schemes available to choose from on the latter. You can also adjust the field of view, turn off or tone down camera shaking, as well as turn blood splatter and dismemberment on or off if you prefer a less gory experience.

How I reviewed Valor Mortis

I played over 10 hours of Valor Mortis on PC using an Xbox Wireless Controller and playing audio through my Bose QuietComfort Headphones. I’m using a fairly mid-range rig with a Ryzen 5 9600X, an AMD Radeon RX 9060XT, 48GB of RAM, and a Samsung 990 EVO Plus SSD. My monitor is the MSI MAG 27CQ6F.

With some tweaks to the settings, I was able to play the game on High at 1440p with frame generation on and maintain a comfortable 100+ fps at almost all times. It may not be a hugely demanding game, but it looks good and runs smoothly, which is ideal for reaction-based combat.

First reviewed October 2026.

Where were you when the first iPhone went on sale back in 2007? I remember clearly that I was working for Phones4U — a third-party phone reseller in the UK — and the iPhone launch completely decimated the business, as suddenly Phones4U wasn't a reseller of the most popular phone on the market.

The same existential threat applied to all carriers except AT&T and O2, which had exclusive launch rights to the iPhone in the US and UK, respectively.

Fast-forward 19 years, and Apple has repeatedly proven that the iPhone's popularity gives it considerable power over its carrier partners — customers are no longer loyal to the carrier providing the mobile service, but to the brand behind their favorite phone.

Apple's newest commercial venture, the Apple Upgrade program — which was announced earlier this year and is, at the time of writing, still only available in the US — is the latest dagger to the heart of major MNOs (Mobile Network Operators).

But despite being a fantastic deal for customers and Apple itself (and make no mistake: I believe it's a net positive move), the Apple Upgrade program could yet prove a double-edged sword for the mobile industry as a whole. Let me explain.

Carriers used to have ALL the power...

The Motorola Droid Ultra is an example of a carrier-branded smartphone

To understand what the Apple Upgrade program is, you have to first understand how the industry worked before the original iPhone. To say that Apple has irrevocably changed the carrier business model on numerous occasions may seem like hyperbole, but it’s hard to overstate the company's lasting effect.

Before the original iPhone, carriers were integral to a phone's success. Carriers controlled which phones were sold in stores, which phones received the bigger marketing pushes, and which were “certified” to work with that carrier.

This meant that all phone makers had to focus more on acquiescing to carrier requests — including shipping customized devices with several preloaded apps and even carrier-specific branding — than meeting their customers' needs.

In many ways, carriers were the customer rather than the end-user itself, and they proved even more integral, as each carrier also controlled when software updates were released to its phones. I remember clearly how customers would blame the phone maker for software-related shortcomings rather than the carrier, even if the software update was delayed by the carrier’s own internal testing.

The power of carriers was inevitably going to be replaced by a different model, but the iPhone had a marked effect on the power wielded by carriers in more ways than one.

... until Apple launched the original iPhone

(Image credit: TechRadar)

If you’ve seen the BlackBerry movie, you’ll remember the line used to describe the original iPhone before it went on sale: it had the "highest interest of any consumer electronic item in history". To go further, its original $499 / £269 price also made it the most expensive phone in history at the time.

But contrary to industry standards at the time, the exceptional consumer interest in the original iPhone meant carriers were clamoring to stock it, rather than Apple having to entice them to sell it. This complete reversal of established norms also meant Apple was able to stamp out many of the key controls carriers had, and the effect was stark.

The iPhone had a marked effect on the power wielded by carriers in more ways than one.

Nirave Gondhia

Let’s consider each of the previous requests that carriers would make of a phone maker. Branding on the phone itself or in the software? Gone for the iPhone. Preloaded carrier apps and other bloatware? Gone for the iPhone. A software update schedule controlled by carriers? Apple releases updates to all phones at the same time globally. Customized models? Carriers should be grateful they get to stock the iPhone at all.

However, Apple's unique position went even further. Phone makers used to have to discount phones and offer incentives, but for many years, Apple did not. Granted, we are now in an era in which Apple competes through trade-in promotions and works closely with its carrier partners, but it was the last major phone maker to embrace promotional pricing of any kind.

It's not just Apple that has benefited from these changes; as Apple established new ground rules, companies like Samsung, Google, and LG have all benefited from a different playbook than in the pre-smartphone era.

Now, with the Apple Upgrade program, Apple again holds outsized control over the customer journey. For many, this is a great program, but for other stakeholders, it’s decidedly not. Here’s why.

The Apple Upgrade program is a great deal for consumers

Customers can now lease a new iPhone, Mac, iPad, or Apple Watch from Apple directly and pay for it monthly over a fixed lease period (Image credit: Lance Ulanoff / Future)

The Apple Upgrade Program is a leasing program provided by Apple, with financing by Klarna. It lets you lease a new iPhone, Mac, iPad, or Apple Watch and pay for it monthly over a fixed lease period (between 12 months and 36 months). At the end of the lease, you can opt to return the item to Apple at no charge or buy it out for the remaining value; the latter is a particularly attractive offer for customers.

Most lease programs have you spending more money over the course of the lease compared with buying the item outright. This arbitrage is how the financing lender makes their money, but with the Apple Upgrade program, you’ll never pay more in total than the outright purchase value.

This is crucial: Apple is effectively offering 0% financing across its range of products, and while it likely pays Klarna directly for providing these services, it does not pass any additional costs on to the customer. In essence, you can still buy any Apple item you plan to, but you now have a more flexible way to pay for it, with no credit card or specific credit limit needed.

... and also great for Apple

(Image credit: Photo by Rasid Necati Aslim/Anadolu via Getty Images)

The current state of RAMageddon — aka the soaring cost of all technology due to the shortage of RAM modules — meant it was inevitable that Apple would have to increase its product prices, which it did by a considerable margin in both June and September this year.

Despite Apple customers traditionally being less price-sensitive than Android customers, rising prices during a global cost-of-living crisis would inevitably have impacted its financial statements and, therefore, its share price.

The Apple Upgrade program changes this reality in a big way: in the US at least, it’s now far easier for customers to buy Apple products (and more of them) than they used to be able to, and Apple itself gains far more valuable numbers to present to its shareholders: the number of customers on subscriptions; the dollar value of those sold; and the number of customers.

More than that, the Upgrade Program offers Apple predictable revenue numbers for quarters and years into the future, and allows it to launch more products across the year without worrying as much about established launch cycles and windows.

It also makes all of its new products far more accessible to the masses, which is particularly prescient given the now-confirmed $1,999 / £1,999 / AU$3,499 launch price of the iPhone Duo, which begins shipping next week.

However, the Apple Upgrade program is bad for carriers...

(Image credit: Shutterstock)

For all its positives, there are two areas where the Apple Upgrade program could have a negative effect — and both stem from its potential impact on carriers.

Like the now-discontinued iPhone Upgrade program before it, an iPhone purchased via the Apple Upgrade program is seemingly unlocked for use on any network. It still requires activation with one of the Big Three carriers — Verizon, T-Mobile US, or AT&T — at the point of sale, but once you complete said activation, you can switch to a different carrier of your choosing.

This undoubtedly reduces the need to be locked into a long-term, multi-year contract with your carrier and, crucially, removes a key source of revenue from locked devices: roaming fees. Together, they will impact carriers’ revenue, inevitably leading to several knock-on effects for the consumer.

The Apple Upgrade program is a deal that's too good not to take advantage of.

Nirave Gondhia

First, customers will inevitably end up paying more for their service plans, whether through higher plan costs, higher add-on costs, or even increased out-of-bundle charges.

Second, the above will lead to an even more booming MVNO (Mobile Virtual Network Operator) market, but as we’ve seen with Mint Mobile, these MVNOs will also inevitably be acquired by the aforementioned big carriers they rely on.

The biggest impact will be felt in network technology and retail footprint. Verizon, T-Mobile US, and AT&T are publicly traded and have a fiduciary responsibility to their shareholders. They might pay even more for future spectrum holdings, but the adoption and rollout of future technologies could be much slower if the Apple Upgrade program takes off in the way Apple clearly hopes. This isn’t just hyperbole; we’ve already seen this happen with the 5G rollout and the LTE rollout before it.

The impact will be even more acute for those who like to frequent carrier retail stores. T-Mobile has already made some services fully digital, and, like other industries, phone retail seems to be undergoing a managed decline. Further pressure on carriers’ revenue sources will simply accelerate this shift, leaving vulnerable customers most likely to feel the effects.

So, yes, the Apple Upgrade program is a fantastic deal for most people involved, but I am concerned about the impact it will have on carrier operations, and specifically how these operations might impact consumers in the long term.

Having said all that, my next Apple purchase will almost certainly be through the Apple Upgrade program — it’s a deal that’s too good not to take advantage of, especially if you’re buying one of Apple's higher-spec models.

• OpenAI has solved or partially solved 372 outstanding mathematical challenges
• Some of the solutions, while clear, cannot be understood by some humans, and were produced with an internal, unreleased AI model
• Renowned mathematician Terence Tao has called for colleagues working with OpenAI to abandon their projects

OpenAI has stunned the world of advanced maths by releasing solutions and partial solutions to 372 outstanding problems that have puzzled the world’s greatest minds for decades.

Just two years ago, AI was finding it tricky to handle GCSE-level mathematical challenges. Today, it is solving problems that have taken humans years to even establish the existence of. The solutions provided, meanwhile, are considered almost as complex as the problems, and the scale – published in 722 papers – make it difficult for human mathematicians to understand and verified.

The impact of the release has prompted one of the world’s greatest mathematicians, Terence Tao, to call for a return to a traditional vision of science.

Demonstration of power

OpenAI’s publication of its work on these advanced mathematical problems is the latest demonstration of how far the technology has come. Just a few weeks ago, the company published its proposal for the Navier-Stokes equation (concerning the behaviour of viscous fluids), generated by its unreleased specialized math model.

Tao is leading the opposition, as one of the field’s leading voices, and a man considered to be the greatest living mind in mathematics.

Following OpenAI’s contribution, he shared a post from the Association for Human Mathematics, which concludes:

“Releasing over 700 files at once is not a demonstration of scholarship, but a demonstration of power. We urge mathematicians to discontinue their work with OpenAI and to return to a vision of science that centers human understanding.”

However, as powerful as that conclusion is, their response to OpenAI’s intervention as a whole highlights that the company may have overstepped.

Questions of legitimacy

The Association for Human Mathematics states that “Mathematicians did not ask for this work to be done,” continuing that “The Advisory Group on Mathematics and Artificial Intelligence, from whom OpenAI has claimed to derive its legitimacy, opened their initial advisory statement by saying that frontier AI corporations should not test advanced mathematical problems on internal models.”

By ignoring the position of the advisory group, they state that OpenAI has essentially gone beyond the pale, and “indicated total disregard for the norms of scientific research — norms that guarantee that mathematics remains trustworthy, ethically researched, and in the public interest.” Furthermore, they encourage mathematicians to treat the publication with “due skepticism.”

OpenAI has responded by stating "we want this progress to push the frontier of human knowledge and enable further progress in mathematics." However, its actions are being interpreted as reckless, and this showboating may well prove to be damaging for OpenAI and the wider AI industry.

Light spoilers follow for VisionQuest.

No Marvel movie trilogy is perfect. That might sound like a loaded statement, but ask any Marvel Cinematic Universe (MCU) fan and they'll concede that even the franchise's gold standards — the original Captain America films, or Guardians of the Galaxy movie series, to name just two — have flaws.

So, it falls to Marvel's TV show department to show its big-screen counterpart how to deliver a truly complete arc. And, with two acclaimed entries — WandaVision and Agatha All Along — under its belt, the Wanda/Vision TV trilogy only needed to stick the landing with a strong, final act to stake its claim as the MCU's best three-part story.

Thankfully, VisionQuest does exactly that, delivering a scintillating slice of superhero science fiction, and a surprisingly deep and emotionally resonant conclusion to the Wanda/Vision Saga.

Paranoid android

VisionQuest reconnects us with Paul Bettany's synthezoid four years after WandaVision's finale (Image credit: Marvel Television/Disney+)

Set some time after WandaVision's finale, VisionQuest reunites us with White Vision (Paul Bettany), the reanimated, blank-slate synthezoid rebuilt by S.W.O.R.D after Thanos destroyed the original. Now laying low in Edinburgh as a bartender named Sherwood, he's grappling with a severe identity crisis, caught between his vacuous programming and the restored memories of his past life.

Searching for purpose, Vision has turned inward and now consults the AI programs embedded in his programming, including Ultron (James Spader), whose leftover code lingers in Vision's system.

Bettany thrives on the material he has to work with

Unfortunately, his peaceful existence is about to be interrupted.

Hunted by Paladin (Todd Stashwick), a mercenary hired by a mysterious client who wants his highly-advanced tech, Vision's quiet life is further upended when he meets Thomas Shepard (Ruaridh Mollica), a thief claiming to be one of Vision and Wanda's twin sons. Forced to go on the run, Vision must evade capture, uncover the truth about Thomas, and resist Ultron's insidious influence.

There are clear parallels between the imaginary worlds that Vision and Wanda create in their self-titled shows (Image credit: Marvel Television/Disney+)

After 12 years portraying Vision, Bettany has more than earned his shot at leading a solo project that not only lets him shine, but add profound substance to the synthezoid.

Armed with his obvious talent, and clever scripts from showrunner Terry Matalas and his writing team, Bettany thrives on the earnest, amusing, mind-bending, and occasionally sinister material he has to work with.

There's a strange delight in seeing this side of Vision fully unleashed

Crucially, as this Marvel Phase 6 show reminds us, this isn't quite the Vision we knew, but a repurposed and weaponized Re-Vision (if you will) of the original — and Bettany plays the uncanny shift brilliantly.

Indeed, he delivers a striking performance as a morally gray Vision, whose emotional turmoil gives rise to a colder, darker version of the character that strays dangerously into Ultron 2.0 territory more than once. Yet there's a strange delight in seeing this side of Vision fully unleashed on his foes, delivering some unexpectedly bloody, brutal beatdowns along the way.

Only highlighting his ruthless streak would be a disservice to the depth Bettany brings, though.

Given the performances of VisionQuest's terrific supporting ensemble, of which the spellbinding Spader is the clear highlight (more on him shortly), I initially worried that Bettany would be upstaged in his own show. Those fears were completely shattered by his mesmeric, commanding display in the penultimate episode — a chapter that, major spoilers notwithstanding, makes Bettany's infamous dual-acting scene in WandaVision's finale feel like small fry.

Father and son

Spader and Bettany are a delightful duo who light up the screen at every opportunity (Image credit: Marvel Television/Disney+)

While VisionQuest is naturally positioned as a Vision-first story, it wouldn't have worked as well without the scene-stealing Spader.

Spader is an utter delight from the outset

Reprising his role 12 years on from his only MCU appearance, in Avengers: Age of Ultron, Spader is an utter delight from the outset. Ultron's abrupt death was a sore point for fans wanting more of Spader's sassy, loquacious, and menacing android in the MCU, and VisionQuest gives him the chance to chew the scenery again, with a multifaceted performance that keeps you guessing about his true motives at every turn.

You'll get to see enough of Ultron in android form throughout VisionQuest (Image credit: Marvel Television/Disney+)

As the proverbial family misfit of Vision's surrogate AI clan, and the button-pushing devil on Vision's shoulder, Ultron's modus operandi remains grounded in manipulation. Yet, despite his destructive father-figure tendencies, there's far more to this version of Ultron than meets the eye — a depth that truly reveals itself whenever Spader and Bettany share the screen.

VisionQuest gives Bettany and Spader room to fully unpack the complexities of Vision and Ultron's paternal bond

Their brief, piercingly philosophical exchange in Age of Ultron already teased their sharp dynamic, but an eight-part series gives Bettany and Spader room to fully unpack the complexities of Vision and Ultron's paternal bond.

Indeed, there's a genuine earnestness to their paradoxically human conversations, which carry the weight of a two-hander Shakespearean tragedy. And, while their odd-couple partnership offers plenty of welcome levity, it's the soulful, quiet segments that ultimately left the biggest impression on me.

The evolution of Vision and Tommy's complex relationship is a real highlight (Image credit: Marvel Television/Disney+)

The bond between Bettany's Vision and Mollica's charismatic yet anguished Shepard receives a similarly compelling, if slightly less expansive, exploration.

Driven by striking parallels in their search for identity, their dynamic is equal parts melodrama, humor, charm, and raw sentiment. Bettany and Mollica capture the ill-equipped dad and parentally unanchored son with real nuance, fitting seamlessly into VisionQuest's overarching study of intergenerational paternal trauma. Watching them navigate this fledgling family connection, and the trust that slowly takes root, is easily another of the Disney+ show's biggest strengths.

This modern glitch

F.R.I.D.A.Y, J.A.R.V.I.S, U, and D.U.M-E have roles of varying importance in VisionQuest (Image credit: Marvel Television/Disney+)

VisionQuest's long-form storytelling format affords it the time and space to explore Vision's AI family, even if it's fairly selective in terms of who actually gets the spotlight.

VisionQuest is fairly selective in terms of which AI characters actually gets the spotlight

The bubbly E.D.I.T.H (Emily Hampshire) is an easy standout among Vision's digital siblings, though lacks the complete character arc given to the overly cautious F.R.I.D.A.Y (Orla Brady). The rest of of the AI ensemble serves as comic relief — a role capably filled by Henry Lewis' D.U.M-E and Jonathan Sayer's U occupy — while James D'Arcy's J.A.R.V.I.S, is criminally underused, for reasons I can't reveal here in order to avoid spoilers.

VisionQuest has its share of action sequences, including this duel between Vision and Jocasta (Image credit: Marvel Television/Disney+)

Fortunately, Jocasta (T'Nia Miller) doesn't suffer the same fate. Teased as one of four villains in VisionQuest's official trailer, this morally complex android plays a far more important role than I had initially anticipated.

I had concerns that separating Bettany and Spader's electrifying dynamic at VisionQuest's midpoint might impact its narrative impetus. However, Spader and Miller's captivating chemistry as misfits with a united goal quickly allayed those fears. Drawn together by a shared emotional void, their tragic, dysfunctional romance provides a bittersweet alternative to Bettany and Spader's dynamic that admirably fills the void.

Narratively, VisionQuest doesn't always hit the sweet spot

Narratively, VisionQuest doesn't always hit the sweet spot.

Todd Stashwick's cold and calculating Paladin makes for an intriguing threat early on, but he's disappointingly sidelined as the plot advances, which frustratingly leaves his backstory somewhat unexplored. Elsewhere, the show's first major plot twist lacks the heart-shattering impact it strives for, while a few tense midseason beats are undermined by jarring editing choices.

Lisa Molinari, played by Lauren Morais, has a semblance of a character arc in VisionQuest (Image credit: Marvel Television/Disney+)

By contrast, VisionQuest's tonal shifts rarely drag its story down. A multi-genre narrative in every sense, it effortlessly flits between light-hearted, clever, and innuendo-based humor, and dark, visceral moments (believe me, it's more violent that you'd expect) without letting its sci-fi elements overwhelm the human core.

And then there's the perpetual elephant in the room: the long shadow cast by Tony Stark. Marvel's habit of tethering stories to Robert Downey Jr's iconic character can often feel like a tired trope, yet here it works. VisionQuest isn't shy or subtle about its overarching exploration of the 'sins of the father', retracing the AI family tree through Stark, and — again I'm avoiding major spoilers here — even further back into the MCU’s past. Given how deeply that lineage intersects with Vision, Ultron, and the rest of this eclectic cast, Stark's looming presence feels entirely organic to the story being told.

My verdict

VisionQuest is a terrifically assembled, genre-bending conclusion to the Wanda/Vision TV trilogy. Equally enjoyable as a standalone tale as it is as part of a larger saga, it delivers an emotionally rich examination about what it means to be human — all wrapped up in a mind-melting, expectation-subverting package that's full of gusto, poise, and visual flair.

As its end credits rolled for the final time, I couldn't help but be reminded of Vision's famous quote from Age of Ultron: "A thing isn't beautiful because it lasts." As an eight-part miniseries, there's an obvious finality to VisionQuest, but that doesn't mean we can't appreciate it for the triumph it is, both now and for years to come — and that's the exact kind of beauty I can get behind.

VisionQuest makes its worldwide debut on Disney+ with a two-episode premiere on Wednesday, October 14.

• Mozilla accused Microsoft of using Windows, Edge, Bing, and Copilot to steer users toward Edge
• Firefox maker claims stricter EU regulations force fairer browser choice than elsewhere
• Browser competition increasingly overlaps with the fight for AI distribution and user reach

Mozilla has accused Microsoft of abusing its dominant position wherever it can to push people towards Edge, and away from other browsers.

The company framed the story as if it wants “better for the people”, while Microsoft only does it when it’s forced to, but is that really the case?

In a newly released blog, Mozilla noted it has now been 100 days since it published Over The Edge 2.0, a report from “leading experts on deceptive design Dr. Harry Brignull and Cennydd Bowles on how Windows, Edge, Bing, and Copilot are designed to steer people away from the browser they choose.”

Better for the people

In the report, Mozilla said that in the European Economic Area (EEA), where the Digital Market Acts (DMA) is being enforced, Microsoft is forced to “play fair”.

Anywhere else around the world, including the US, UK, India, and other parts of the planet, it still uses its “shady tactics”. These include the Windows 10 message “You’re almost done setting up your PC” which allegedly uses “tricky wording” to pull people off the default they picked, Windows Search and Widgets opening links in Edge instead of the default browser, and Windows keeping Edge as the default for common browser file types like PDF and SVG, even after another browser were set up as the default.

Mozilla claims that even within the EU, Microsoft is doing all it can to steer people towards Edge. That includes pre-pinning the browser to the taskbar which, according to the company, “seems to directly violate the ‘principled approach to app pinning and app defaults’, Copilot opening web links in a side-panel instead of in the browser, and Windows Backup “silently” trying to set Edge as the default when it restores a PC and “fails to properly bring the user’s own browser along. “

“Most of the 2024 patterns are still there. The 2026 patterns are worse. Microsoft is using new surfaces like Copilot and the migration to Windows 11 to further undermine user choice,” Mozilla claims.

“We also shared the report with Microsoft and talked to them about the issues raised in the report. We knew that an immediate change was unlikely, but we hoped that, when faced with independent evidence of harmful design practices impacting Windows users, they would also want better for people,” Mozilla said.

“Instead, it became clear that Microsoft will respect user choice only when it is forced to. Microsoft’s standard for respecting people’s choices is legal compliance. Not what’s best for its users. Not its own published principles. Not what’s technically feasible. Not even what it already does for users in Europe.”

A story on AI

Mozilla said it shared the report with Microsoft but did not say if the company replied in any way. We’ve reached out to Microsoft with a request for comment and will update this article when we hear from them.

In the meantime, we believe it is also important to mention what Mozilla said in the blog’s introduction:

“As AI reshapes our online experiences, the stakes couldn’t be higher. Browsers are a key distribution layer for AI tools and services, while still determining many of the privacy and security functionality people rely on every day. Steering people toward one browser can therefore shape which AI services reach users and the protections that come with it.”

So the question becomes - is this really about wanting better for people, or wanting people to use one AI model instead of another AI model? If demanding equal browser opportunity is fair game, then demanding equal AI opportunity is also fair game, so why not just say so? Why hide behind alleged altruistic causes and principles?

In the blog, Mozilla says it has a “simple” ask - to expand the “fairer designs” currently being enforced in the EEA everywhere on the planet, and to “stop the remaining harmful design practices that persist in the EEA.”

The company concluded the blog by saying there will be no more mister Nice Guy. It says it “stepped up” public pressure on the company and has reached out to regulators and policy makers.

“We hope that regulators in the US, the UK, India, and Brazil are paying attention.”

The Sihoo B100 is one of the best budget ergonomic office chairs I've seen. It combines adaptive lumbar support, breathable mesh, and several adjustment options, and it's reduced in price right now.

• Best for: Office and remote workers and students spending long hours at a computer who wants an adjustable chair with lumbar support and breathable mesh.

• The deal: The Sihoo B100 ergonomic office chair is now $160 (was $220) at Amazon, saving you $60. In the UK, the B100 is now £150 (was £190).

• Why it matters: This is the chair our office furniture editor recommends for those looking for a cheap option. Most office chairs at this price have either fixed or adjustable lumbar support - but this one has an adaptive design that delivers consistent support as you move in your seat. I also like the three reclining positions, an adjustable headrest, and flip-up armrests, which let you customize your seating position throughout the working day.

B100 Mesh Ergonomic Office Chair: was $219.99 now $159.97
The Sihoo B100 ergonomic office chair features adaptive lumbar support, breathable mesh, an adjustable headrest, flip-up armrests, three reclining positions, adjustable seat height, a reinforced metal base, and 300lb capacity.

In the UK: now £150 (was £190)View Deal

Why we recommend it

If you're looking for a reasonably priced ergonomic office chair, the Sihoo B100 offers an adaptive lumbar support that's designed to follow your lower back as you change sitting positions. The support moves with you when you lean forward or sit upright.

The chair also features a breathable mesh backrest, which allows air to circulate and helps reduce heat buildup. There's an adjustable headrest for neck support, and a widened seat cushion that distributes weight across your hips and thighs.

You can adjust the seat height from 18.3 to 21.9 inches to suit different desk heights and sitting preferences and the B100 also offers three lockable reclining positions at 110, 125, and 135 degrees.

We haven't reviewed this specific model yet, but we've tested plenty of Sihoo office chairs over the years and they always impress us with their comfort, adjustability, and value.

For more of our favorites, see our guide to the best office chairs we've tested.

Price context & historical value

Amazon has reduced the Sihoo B100 from $219.99 to $159.97, or approximately 27% off. The B100 has sold for as low as $149.99 at Amazon previously, but at just $10 more than its record low, this is still a great deal for an ergonomic chair with adaptive lumbar support. In the UK, the lowest price we've seen is £140 during sales events - in this year's October Prime Day, it hit £143, so not an all-time low, but as with the US deal price, still good for the features on offer.

Should you buy it?

✅ Buy the Sihoo B100 Mesh Office Chair if...

You spend several hours a day working at a desk and want a chair that offers plenty of adjustment options. The adaptive lumbar support, adjustable headrest, and three reclining positions give you flexibility and the breathable mesh backrest will keep you cool when your home office gets warm.

❌ Skip the Sihoo B100 Mesh Office Chair if...

You prefer a heavily padded chair or need more control over the armrest positioning. The B100 has flip-up armrests rather than fully adjustable 3D or 4D versions. You should also check the seat dimensions and height range before ordering if you're taller or shorter than average.

• Shop all the office chairs on sale at Amazon

• Budget Android phones shipped with firmware-level malware preinstalled before reaching customers
• Malware generated fraudulent ad revenue and turned infected devices into residential proxies
• Victims spanned 150 countries, with cleanup often requiring firmware-level remediation

Imagine buying a brand new, sealed Android phone, and later realizing it came with preinstalled malware rooted deeply into the firmware and thus almost impossible to remove - generating fraudulent ad clicks for a malicious third party, and even turning the phones into residential proxies to be used for whatever sinister purposes.

Security researchers Bitdefender found an unidentified threat actor managed to compromise the supply chain of multiple smartphone manufacturers and added malware on a firmware level before the devices were even shipped to stores.

A number of brands were mentioned, including Doogee S200 X, Doogee Fire 3 Max, and Cubot KINGKONG X - both Chinese smartphone manufacturers building budget, rugged Android phones.

The reports also mention ripoff devices impersonating Samsung and Apple products. In general, these are devices using MediaTek chipsets, which were infected in a way that allowed the attackers to tamper with applications, grant themselves higher privileges, and even do remote code execution (RCE) attacks without the victim’s knowledge.

Victims across the world

Unfortunately, these devices were shipped all over the world, and Bitdefender claims “thousands” of people were victims, scattered across some 150 countries around the world, with the highest concentration found in Mexico, France, Italy, United States, Germany, Brazil, and Spain.

The malware comes in different shapes and sizes. So far, Bitdefender uncovered 32 malicious apps, including those pretending to be weather apps, file managers app lockers, OCR tools, and audio editors. All of these do the same thing - create an invisible overlay over different apps and register an ad being viewed, despite the fact that the user never actually saw it.

That way, whoever installed the malware scams the ad network into thinking the ad was served to thousands of users which, in turn, generates profit. We don’t know exactly how many devices are compromised this way, or how much money the crooks stole this way.

"The system app itself doesn’t register the fraudulent impressions and clicks," Bitdefender said. "The revenue engine is driven by the dropped cover apps, including real-looking weather, app-lock, note, and OCR apps, which load genuine ads through a legitimate ad SDK. The goal is simple: to load an invisible window on top of apps that registers ads being shown."

Spilling into the Play Store

Bitdefender also said that one of the uncovered applications (the app locker one), contained a TCP proxy element that registered the compromised phones with a remote command server. After successful registration, the attackers were able to forward traffic through the phone, essentially creating a residential proxy service that criminals can use for different things, from launching DDoS attacks to participating in spam and wire fraud.

They also found more than a dozen applications on the Google Play Store with the same fraudulent advertising code, as well as the same communications infrastructure. These, however, required manual installation and required users to grant them the requested permissions, which might have raised flags with the users.

Removing the malware isn’t as easy as simply uninstalling the app. The researchers said that many of the devices require cleanup on the firmware level, or disabling the malicious component using Android Debug Bridge (ADB), which could be too technical for many victims.

The easier and more elegant solution is also a more expensive one - don’t buy devices from brands with a shady reputation.

If you're looking to upgrade your desktop display, I've found a great deal on the Asus ROG OLED monitor that combines a glossy 1440p OLED panel, a 240Hz refresh rate, and excellent color reproduction, and it's on sale right now.

• Best for: Content creators and professionals who want excellent contrast, accurate colors, and a smooth display for photo and video editing.

• The deal: The Asus ROG OLED monitor is now $384 (was $599) at Amazon, saving you $215.

• Why it matters: You're getting a 1440p OLED panel, 99% DCI-P3 color coverage, DisplayHDR 400 True Black certification, and advanced burn-in protection features at a considerably lower price than normal.

Today's best Asus OLED monitor

ROG 27-inch OLED monitor: was $599 now $384
This monitor features a 27-inch 1440p glossy OLED display, 240Hz refresh rate, 0.03ms response time, DisplayHDR 400 True Black, 99% DCI-P3 coverage, and Delta E < 2. View Deal

Why we recommend it

The Asus ROG monitor's 26.5-inch viewable WOLED panel delivers a 2560 x 1440 resolution, providing sharp images and a useful amount of desktop space for everyday tasks.

For content creation, the TrueBlack Glossy finish produces clear images and rich colors, making photographs, videos, and other visual content look great. The monitor's 99% DCI-P3 color coverage and true 10-bit color support with Delta E < 2 color difference is useful for creators - especially at this price. For video editing, I especially like the 240Hz refresh rate and 0.03ms response time, too.

Another useful feature is Asus OLED Care Pro, which includes functions designed to reduce the risk of burn-in. You also get DisplayWidget Center software, allowing you to adjust monitor settings through your computer rather than relying entirely on physical controls.

Price context & historical value

Amazon has reduced the price of Asus' ROG 27-inch OLED monitor from $599 to $384, saving you $215. That's a discount of approximately 36%, and is the lowest price we've seen this model sell for on Amazon so far.

Should you buy it?

✅ Buy the ROG 27-inch OLED monitor if...

You want a high-quality desktop monitor for creative work. This screen offers deep blacks, excellent contrast, and wide color coverage backed with a 240Hz refresh rate, and vivid images.

❌ Skip the ROG 27-inch OLED monitor if...

You need a 4K display, regularly work with static content for extended periods, or use your computer in a brightly lit environment.

• Shop all the best monitors Amazon

If your internet plan or devices have outgrown your old router, TP-Link's Archer BE670 Wi-Fi 7 model has a good discount right now with all the features I recommend.

For me, the port layout sets it apart at this price: a 10Gbps port and a 2.5Gbps port sit alongside Wi-Fi 7, so it should keep up with fiber plans and fast wired devices as you upgrade.

Tri-band Wi-Fi 7 with Multi-Link Operation also lets compatible phones and laptops use more than one band at once, though your devices need Wi-Fi 6E or Wi-Fi 7 to benefit for the fastest speeds.

• Best for: Multi-gig or fiber households and home offices that want a fast tri-band Wi-Fi 7 router with room to grow.

• The deal: The TP-Link Archer BE670 Wi-Fi 7 router is $250 (was $300) at Amazon.

BE670 Wi-Fi 7 Router: was $299.99 now $249.99
The BE670 is a BE12000 tri-band router with speeds up to 5,765 Mbps on 6 GHz, 5,765 Mbps on 5 GHz, and 688 Mbps on 2.4 GHz. It adds 320MHz channels and Multi-Link Operation, plus one 10Gbps WAN/LAN port, one 2.5Gbps WAN/LAN port, three 1Gbps LAN ports, a USB 3.0 port, EasyMesh support, and HomeShield security. TP-Link says it supports up to 128 devices, and Amazon sells and ships it.View Deal

Price Context & Historical Value

At $249.99, this is $50 (17%) off Amazon's $299.99 list price. According to CamelCamelCamel, that's almost the lowest price new since March - although Keepa suggests it dropped to $230 in June. So, it's about what I'd expect to pay for a Wi-Fi 7 router outside of a sales event like Prime Day or Black Friday.

Should you buy it?

✅ Buy the TP-Link Archer BE670 Wi-Fi 7 router if... you have a multi-gig internet plan or lots of wired gear, and you want a Wi-Fi 7 router with a 10Gbps port.

❌ Skip the TP-Link Archer BE670 Wi-Fi 7 router if... you only have a gigabit plan and a handful of devices, since cheaper Wi-Fi 7 routers will do, or you want a tested pick, since we haven't tested the BE670 ourselves and it has only 125 Amazon ratings so far.

• Check out more network deals at Amazon

Our report on laptop price hikes says analysts expect more increases ahead even as PC sales fall, which is why a laptop with plenty of memory and storage is a sensible buy now.

We've had great experiences with HP's OmniBook line in the past - and the HP OmniBook 16 AI is a great example of the sort of 'future-proof' laptop with a configuration that should stay useful for years.

32GB of DDR5 RAM and a 1TB SSD give it headroom that cheaper 16GB configurations lack, and are the parts most likely to get pricier. And the 50 TOPS NPU covers Windows' Copilot+ AI features as they arrive.

• Best for: Remote workers and professionals who want a future-proof laptop now, before prices climb.

• The deal: The HP OmniBook 16 AI laptop is now $1300 (was $2400) at Amazon.

The OmniBook 16 AI has a 16-inch 2K (1920 x 1200) IPS touchscreen, an AMD Ryzen AI 7 445 with six cores and a 50 TOPS NPU, Radeon 840M integrated graphics, 32GB of DDR5 and a 1TB NVMe SSD. It has two USB-C ports, two USB-A ports, HDMI 2.1, a 1080p IR camera with a privacy shutter, Wi-Fi 6 and Windows 11 Pro.View Deal

Price Context & Historical Value

At $1,299.99, Amazon shows this as 46% off a $2,399.99 "typical price", but that figure comes from a third-party seller and may not reflect what this laptop usually costs. A better yardstick is the spec sheet: 32GB and 1TB at this price is strong value for a 16-inch touchscreen. Amazon Visa cardholders can also get $50 off on approval.

Should you buy it?

✅ Buy the HP OmniBook 16 AI laptop if... you want to buy a well-equipped 16-inch laptop now, before analysts' predicted price hikes arrive, and you value 32GB of RAM and 1TB of storage.

❌ Skip the HP OmniBook 16 AI laptop if... you need a dedicated graphics card for gaming or video editing, want Wi-Fi 7, or prefer buying from HP or Amazon directly rather than a third-party seller, and we haven't tested this model ourselves.

• Check out other laptop deals at Amazon.

• The Tor Project is maintaining its technical collaboration with Mullvad despite intense community pressure to sever ties.
• The backlash stems from a recent political donation made by a Mullvad co-founder to a controversial Swedish populist party.
• Tor is pausing all "proactive co-branding" and joint promotional activities to ensure its human rights mission isn't compromised.

The Tor Project has confirmed it will not terminate its relationship with Mullvad following intense community pushback over a political donation made by one of the VPN's co-founders. Instead, the privacy-focused nonprofit is setting "clear boundaries" on its ongoing collaboration.

The controversy began earlier this year when it was revealed that a Mullvad co-founder had made a financial donation to a controversial Swedish populist party.

The move sparked immediate debate within the cybersecurity community, leading some users to demand that Tor, an organization deeply rooted in human rights and anti-censorship, sever ties with what many consider to be one of the best VPN providers on the market.

However, in a statement published on October 9, the Tor Project explained that while it strongly opposes rhetoric that threatens human rights, it must balance its principles with pragmatism.

Ending the partnership, the organization claims, would do more harm than good to the communities that rely on its privacy tools.

"We are continuing our technical collaboration with Mullvad to sustain work that Tor users and contributors depend on," the statement reads. The Tor Project highlighted that this joint effort has significantly improved the Tor Browser's codebase and created a safer environment to test new features.

Drawing the line on co-branding

While the technical work behind the scenes will continue, including ongoing development of the privacy-centric Mullvad Browser, the public-facing relationship between the two entities is being actively scaled back.

To distance itself from any implied political alignment, the Tor Project announced it has "paused proactive co-branding with Mullvad."

This means any joint promotional activities will be halted, and the language used on Tor’s owned channels has been updated to describe the relationship more accurately and "avoid notions of implied endorsement" of Mullvad’s leadership or values.

"We recognize that the potential benefits to Mullvad of such a collaboration are difficult for some in our community to accept," Tor stated, acknowledging the difficult trade-offs behind the decision.

"Our responsibility is to ensure that this collaboration does not compromise our mission or the Tor Project's independence."

Why the partnership matters

(Image credit: Tor)

The collaboration between Tor and Mullvad has been highly fruitful for internet privacy advocates.

By working together, the two organizations successfully built an open-source browser that delivers the anti-tracking and anti-fingerprinting perks of the Tor network without requiring users to actively route their traffic through Tor's slower onion nodes.

Maintaining this complex codebase requires vast resources, something that the Tor Project notes is severely lacking in the broader fight for digital privacy.

"The work to promote and preserve the human right to privacy, free speech, and free access to information online is severely under-resourced," Tor explained, adding that advancing these goals often requires making difficult compromises.

While the Tor Project admits that this decision will not satisfy everyone and carries a "cost to trust" for some of its community members, it ultimately believes the technical benefits of keeping the internet safe for vulnerable users outweigh the political fallout.

Memory prices are still painful, so any discount is welcome, and this V-Color 32GB DDR5-6000 kit is $68 off at Newegg as part of its big computing sale.

• Best for: AMD builders who need a 32GB DDR5 kit and will take any discount they can get on pricey RAM.

• The deal: The V-Color Manta XSky 32GB (2x16GB) DDR5-6000 CL30 RAM kit is $460 (was $528) at Newegg with code GTWF84.

• Why it matters: RAM remains one of the costliest parts of a PC build, and this is the lowest price this memory kit has been in 30 days, once you add Newegg's limited-time discount code.

XSky 32GB (2x16GB) DDRS-6000 CL30: was $527.99 now $459.99
The Manta XSky kit is two 16GB DDR5 modules running at 6000MHz (PC5-48000) with CL30 timings (30-38-38-96) at 1.35V. It uses an AMD EXPO profile, falls back to DDR5-4800 base speed, and has RGB lighting. Newegg sells and ships it with free shipping and a lifetime limited warranty. For the full discount use code GTWF84View Deal

Why we recommend it

In our experience, DDR5-6000 CL30 is the usual sweet spot for AMD Ryzen systems, and the EXPO profile should let you set it with one click in your BIOS. The 32GB capacity suits gaming and general use, and Newegg's page marks the price as its lowest in 30 days.

Price Context & Historical Value

At $479.99, this is $48 (9%) off Newegg's $527.99 price, and Newegg shows a "Lowest Price in 30 Days" badge. That is still a lot for 32GB of RAM, which is why we'll take any discount we can. The page also lists an extra $20-off promo code, which may not last.

Should you buy it?

✅ Buy the V-Color Manta XSky 32GB (2x16GB) DDR5-6000 CL30 RAM kit if... you're building or upgrading an AMD system and need 32GB of DDR5-6000, and you'd rather not wait while RAM prices stay high.

❌ Skip the V-Color Manta XSky 32GB (2x16GB) DDR5-6000 CL30 RAM kit if... you have an Intel system and want an XMP profile, since this kit is AMD EXPO-focused, you don't want RGB, or you can wait for lower prices, and we haven't tested this kit ourselves.

• See all desktop memory at Newegg.

• Attackers abused Microsoft Power BI dashboards to lure victims into downloading ScreenConnect software
• Public dashboards mimicked blurred documents and redirected users to attacker-controlled sites
• Campaign established persistent remote access, potentially enabling ransomware, espionage, or data theft

Microsoft's Power BI has seemingly become the latest in an ever-growing line of legitimate software being abused in phishing attacks.

Security researchers Huntress said they uncovered a new campaign that leverages publicly accessible Power BI dashboards to trick people into downloading remote monitoring and management software (RMM) called ScreenConnect. The end game of the campaign is not known at this time.

Power BI

Power BI is Microsoft’s business intelligence and data visualization platform, allowing users to create charts, reports, and various dashboards, out of their data and other raw information. The tool itself can collect data from sources such as spreadsheets, databases, cloud services, and more, and users can publish reports and dashboards online, or share them with colleagues privately.

In this campaign, which started on or around September 10 2026, the threat actors created a public dashboard, likely using a throwaway Microsoft account (or a compromised one). Huntress does not confirm how that account was obtained, but says that the dashboard was designed to look like a blurred document with a button labeled “Download Reference”.

The blurred document trick is rather common in phishing emails. Other services such as DocuSign, Dropbox, and even Office 365, were being abused in the same way. To “unblur” (or unlock) the document, victims are told to click a link and follow along.

Sometimes they were asked to download a “premium” version of a document reader, or an “updated” one, and sometimes they were asked to log in using their Office or Google credentials. This attack is no different - clicking on the “Download Reference” button redirects the victim to an attacker-controlled website outside Microsoft’s infrastructure, where an automatic download of ScreenConnect is launched.

On its own, ScreenConnect is not malicious. It is a legitimate remote access and support tool developed by ConnectWise, and designed to help IT staff manage devices remotely. Through the product, users can connect other devices on the internet, troubleshoot problems, transfer files, and install software. Once installed and connected to an attacker-controlled endpoint, ScreenConnect can provide persistent remote access.

Establishing persistence

The researchers didn’t discuss how the emails were delivered, but since they contained a link to a legitimate Microsoft service, they most likely evaded at least some email filters and landed in people’s inboxes. After all, this is probably why Power BI was used in the first place (aside from the fact that it’s not commonly abused this way).

But what’s particularly interesting about this attack is that after deploying one ScreenConnect instance, the attackers deployed a second one and configured a scheduled task to repeatedly execute a PowerShell script. In at least one incident, they also deleted the first instance.

Why they did it remains a mystery. An educated guess would be that they were trying to establish reliable, persistent remote access, while at the same time making sure their activities were harder to detect and disrupt. If defenders were to discover and remove one malicious ScreenConnect installation, a second one might remain available.

This is all in the domain of speculation, unfortunately, since Huntress stopped the observed attack before it progressed beyond establishing remote access. Therefore, we don’t know if the goal was to steal data, deploy ransomware, or engage in financial fraud. What we do know is that with two ScreenConnects, they were able to establish a durable foothold inside corporate networks, which they could subsequently leverage for espionage, data theft, ransomware, or other financially motivated attacks.

As a general rule of thumb, the best way to defend against these types of attacks is to be highly skeptical of incoming emails, especially those with links to “locked” documents, and even more so those that end up with a ScreenConnect deployment.

Got your eye on the new iPhone Duo? Preorders for Apple's first-ever foldable flagship are set to go live this Friday and it's looking like it's going to be a popular launch. So much so, in my opinion, that it could potentially sell out.

If you are interested, however, note that there's a little trick you can use to get ahead of the crowds to make preordering easier.

Apple is opening its 'Get Ready' feature today, which allows potential customers to input details ahead of preordering. That includes shipping details, storage capacity, color, and carrier preferences. That's everything that could slow you down if you were to input them once preorders go live, essentially.

You can use the Get Ready feature on both the Apple Store page for the iPhone Duo and the Apple Store‌ app, so it's definitely worth doing if you're looking for a preorder. With this feature, all you have to do is turn up on time and click order, which could be handy if the device is flying off the shelves.

It's quite tough to predict whether the iPhone Duo will sell out or not, but I would err on the side of caution here if you're interested. Excitement is high since it's Apple's first foldable device and these types of devices tend to be manufactured in smaller batches. For example, the Galaxy TriFold sold out earlier this year.

What are the opening Apple iPhone Duo deals?

(Image credit: Apple)

Apple has already listed a few of the various preorder deals that will be going live this Friday, so you can also check those out to see if they fit the bill.

First up, there's Apple's trade-in rebate of up to $885 for unlocked devices. That matches the maximum rebate we saw on the iPhone 18 Pro and Pro Max, so it's a pretty good deal — albeit one that we were expecting.

Two other options listed on the iPhone Duo page are AT&T and T-Mobile. In the case of the former, it's a competitive trade-in rebate of up to $1,200 off with a new line. In the case of the latter, a rebate of up to $1,000, which is a little less than what I was expecting, to be honest.

One deal that's conspicuously missing from Apple's iPhone Duo page is Verizon's opening promotion. You'll have to navigate to Verizon's iPhone Duo page to see that, which is a great all-in-one bundle option. In short, you can combine a Duo with the carrier's 'Simplicity Pro' unlimited plan for $72.22/mo without having to trade in. The carrier is offering a whopping $33/mo discount on device payments, so this could be an extremely competitive opening promotion.

Note: we'll have more recommendations, tips, and news on the upcoming iPhone Duo in the next week. Stay tuned for more preorder content, including recommendations on carrier promotions once they're live.