Every five years, the Australian Bureau of Statistics (ABS) takes a snapshot of Australia through the Census. The last one was held in 2021, which means households across the country are being contacted again in 2026.

The information collected helps show how Australia's population and households are changing and is used to plan services and infrastructure, from schools and hospitals to transport and housing.

This guide answers the key questions about the 2026 Census, including when to complete it, who needs to be counted, what happens if you're late, fines, privacy and what to do if you're away from home.

The ABS also has a handy guide to doing the Census. It's a useful overview, but we've gone a little further below by answering some of the practical questions it doesn't cover.

If you're unsure how the Census rules apply to your situation, call the Census Contact Centre on 1800 181 227 for advice.

When is the 2026 Census in Australia?

Census night for the 2026 Census is Tuesday August 11 2026. Think of it as the snapshot date: questions about where you're staying and who's in your household should reflect Tuesday night.

You can complete the form earlier if you've received your instructions and already know where you'll be on August 11. If your plans are still up in the air, it's better to wait until you know where you'll be staying.

Can you complete the Census after Census night?

Census night isn't a hard submission deadline. If you haven't completed the form on Tuesday August 11, you can still submit it on Wednesday, Thursday, Friday or later.

The ABS continues collecting forms and following up with households after Census night, so if you're late, complete it as soon as you can.

Is the 2026 Census compulsory in Australia?

Yes. Completing and submitting the 2026 Census is compulsory under the Census and Statistics Act 1905. But being late doesn't automatically mean you'll be fined.

The ABS first follows up with people who haven't completed their form. If someone still doesn't respond, they may be sent a written direction formally requiring them to provide the information. Ignoring that direction can lead to prosecution and a fine.

How much is the fine for not completing the Census?

If you've received a written direction and still don't comply, you can face a fine of up to AU$364 for each day you don't comply.

You can find more detail on the ABS's penalty provisions page.

Who needs to complete the 2026 Census?

The Census is designed to count everyone who's in Australia on Census night. That includes babies and children, international students, visitors and visa holders.

That doesn't mean everyone fills out their own form. In most homes, one person completes the household form for everyone staying there that night.

What if you don't have a fixed address?

You should still be included in the Census. If you're experiencing homelessness, the ABS provides Census field staff and other support to help you take part. If you're staying with someone on Census night, you can be included on their household form.

If you're travelling around Australia without a fixed or return address, such as camping, or in a caravan or boat, you still need to be included in the Census.

You'll generally be counted where you're staying on Census night. A caravan, tent or boat can be counted as your dwelling for the night, including at a caravan park, camping ground or marina.

If you don't have a usual address, the 2026 Census form lets you record that. For the usual-address question, the ABS says to select Elsewhere in Australia, choose Enter address manually and enter None in the suburb/locality box.

How do you complete the 2026 Census?

For most households, the process starts with instructions to complete the Census online. These include a Census number and temporary password you can use to access the form.

The ABS says the average household takes about 30 minutes to complete it.

What if you haven't received or have lost your Census letter?

You can still complete the Census. If you haven't received your instructions or no longer have them, you can get a Census number online or request a paper form.

Can you use a paper Census form?

Yes. Paper forms are available to anyone who wants one. Some households will receive a paper form automatically, while others can request one if they'd rather not complete the Census online.

If you need help completing the Census, the Census website has support options, including language and accessibility support. You can also contact the Census Contact Centre on 1800 181 227.

What if there are more than six people in your household?

A paper Census form can include up to six people. If more than six people are staying at your address on Census night, you can complete the Census online or request an extra paper form.

Can you complete your Census answers privately?

Yes. If you don't want other people in your household to see your answers, you can request a separate Census form and complete your information privately. The ABS explains how this works in its 2026 Census collection notice.

Can you change your Census answers after submitting the form?

Once a Census form has been submitted, the ABS says it can't be reopened or edited.

We couldn't find specific online guidance for what happens if you complete the Census early and your plans later change, so we called the ABS media line to clarify. The advice was to contact the Census Contact Centre on 1800 181 227. They'll talk you through the options if your plans have changed or you've realised you made a significant mistake.

Separately, the ABS has processes for dealing with multiple responses. If more than one Census form is submitted for the same person, it keeps the one containing the most complete information about that person.

What if you're away from home on Census night?

If you're away from your usual home on Census night, you'll generally be counted where you actually spend the night. Your usual address can record you as temporarily away.

That can include staying with friends or family or spending the night in a hotel, hospital or aged care home. The ABS also has arrangements for people travelling overnight by plane, train or bus.

If you're temporarily overseas on Census night, you won't be counted in the Census for that night.

What if you're working on Census night?

If you're working overnight but will return to your usual home on Wednesday August 12 and won't be included on a Census form elsewhere, the ABS says you can be included on the form for your usual home.

How are FIFO workers counted in the Census?

Fly-in, fly-out (FIFO) workers who are staying at a worksite or camp on Census night should generally be counted there. For example, if you normally live in Perth but you're at a mining camp in the Pilbara, you'd be counted at the camp and listed as temporarily away from your usual address in Perth.

Why might a Census field officer visit your home?

Census field officers may visit homes to deliver letters or forms, help people complete the Census or follow up after Census night where a completed form hasn't been received.

So a knock at the door doesn't automatically mean there's a problem. In some areas, field visits are simply part of how the Census is carried out.

How can you check if a Census worker is genuine?

If someone claiming to work for the Census contacts you and you're unsure whether they're genuine, verify them through official ABS channels rather than relying on the contact details they give you. Use the official Census website or call the Census Contact Centre on 1800 181 227.

How can you tell if a Census text, email or website is genuine?

The same caution applies to texts, emails and websites. The ABS says genuine Census text messages use the sender ID CENSUS, while texts about other ABS surveys use ABSGov. The official Census website uses a .gov.au address.

If a message doesn't look right, don't use the link it contains. Go directly to the official Census website. The ABS also has online safety guidance.

What questions are on the 2026 Census?

The 2026 Census contains 66 questions covering areas such as your household, age, cultural background, education, work and income, housing, religion and people who are away on Census night.

You can see the full breakdown in the ABS's 2026 Census topics and data release plan.

What's changed in the 2026 Census?

One of the biggest changes in 2026 is the addition of questions about gender and sexual orientation. Everyone is asked about their sex as recorded at birth, while people aged 16 and over are also asked about their gender and how they describe their sexual orientation.

The ABS says the change is intended to build a clearer picture of Australia's LGBTQ+ population and help inform health and community services. If you'd prefer not to answer, both the gender and sexual orientation questions include a prefer not to answer option.

People can also report up to four ancestries in the 2026 Census.

Do you have to answer every Census question?

No. The question about religion is explicitly marked as optional on the 2026 Census form.

The gender and sexual orientation questions for people aged 16 and over also include a "prefer not to answer" option. So, while taking part in the Census itself is compulsory, that doesn't mean every question requires you to provide a particular answer.

Is your 2026 Census information private?

Yes. Census information is protected by law, and the ABS says it doesn't share identifiable information from your Census form with other government departments or direct marketing companies.

That means your individual Census answers aren't handed to the Australian Taxation Office (ATO) to check your tax return or to Centrelink to check your eligibility for a payment.

You can read more in the ABS's 2026 Census Privacy Statement.

Is the Census anonymous?

No. The Census isn't anonymous because it asks for names and addresses. However, the ABS separates names and addresses from other Census information after processing.

When will the 2026 Census results be released?

The first 2026 Census results are due to be released in June 2027, with further data releases continuing into 2028.

Census data will be free to access through ABS tools including QuickStats, TableBuilder and ABS Data Maps.

• Vodafone uncovers a “nation full of drive” with 85% of Brits building towards a major ambition
• Over half of those surveyed believe it is easier to find opportunities in the digital era
• Vodafone launched the UK’s first Opportunity Hub in London, intended to provide advice and connections to ambitious visitors

Vodafone has released the results of research into how being connected has opened the door to people being able to pursue their goals.

With 40% of Brits feeling “more ambitious than ever” and a majority aiming to achieve major targets in personal and professional development, the research is part of VodafoneThree’s investment into building the UK’s best network.

Interestingly, the findings also highlight that key career events take place in our 20s and 30s, and accompany the successful Opportunity Hub event targeted at this age group.

Connecting unlocks opportunity

Vodafone’s survey found over half (55%) of Brits feel it is easier to find opportunities in the digital era, with 20% wishing they had started working toward their ambitions sooner. 42% believe greater connectivity has had a positive effect on reaching their goals, with 49% now “better placed” to achieve than they were 5 or 10 years ago.

The survey also uncovered the blocks to personal and professional ambitions. Unsurprisingly, financial constraints are the main barrier, with 22% admitting this. But lack of confidence at 13%, fear of failure (12%) and limited access to the tools needed for self-improvement (9%) are also mentioned as challenges.

However, life’s disruptions can be seen as launch points, with bereavement (27%) and redundancy (25%) catalysts for chasing ambitions, alongside parenthood (25%).

“Britain isn't short of ambition," a VodafoneThree spokesperson said, "the challenge is turning ambition into action. Our research shows that connectivity is increasingly part of that equation, with nearly two-thirds of would-be founders saying unreliable connectivity has held them back from starting a business locally.”

KitchenAid KF2 two-minute review

The KitchenAid KF2 Fully Automatic Espresso Machine is the entry-level model in the company’s six-strong range of automatic coffee machines, and was launched in mid-2026 alongside the KF3 and KF4.

It adopts the exact same slim build as its two siblings, making any of them ideal for smaller kitchens. It also benefits from the exact same Intelligrind coffee bean grinder, which automatically determines the best grind size depending on the beans you’ve put in. The brewing unit is the same also, and in my taste tests, I couldn’t notice any obvious difference between the three. That means you’re going to get a wonderful, tasty espresso with each use.

What sets the KF2 apart — and where KitchenAid has made cuts to bring the cost down — is its feature set regarding drink presets and general usability. It’s also the only model in KitchenAid’s entire fully automatic coffee machine range to rock a steam wand for foaming milk — the others all use a hose-based system. While this may seem like a drop down in specification, I personally found the steam wand to perform the best when making foamy lattes and cappuccinos. What’s more, a steam wand is inherently going to perform better with non-dairy milk as it allows for greater control over temperature.

As with the KF3 and KF4, the KitchenAid KF2 is capable of making iced-coffee drinks, although as I found in my KitchenAid KF3 review, the results are a little underwhelming. This is because the brewed espresso is hot enough to quickly melt the ice in your cup or glass, despite KitchenAid claiming it’s brewed at a lower temperature.

Where the KitchenAid KF2 loses points slightly is the fact it doesn’t offer user profiles. I’ve been used to having profiles for some time at home, as I currently use the KitchenAid KF4 and have previously used the Philips LatteGo 4400, which offer four profiles and two profiles, respectively. Having them makes brewing a coffee incredibly quick and simple: just select a profile, select a drink and that’s it.

The fact the KF2 doesn’t have profiles means that brewing a variety of coffees will take longer, as you’ll have to select the drink and adjust any settings each time. It therefore might not be the best option for larger households, but for households of one or two people it’s still worthy of consideration.

The other slight niggle I have, which affects the KF3 and KF4 also, is that the unit can go through water relatively quickly due to it running a rinse cycle on start up and shutdown. The shutdown cycle I can understand, but I find the start-up cycle to be a little redundant.

All things considered, the KitchenAid KF2 Fully Automatic Espresso Machine is another fantastic option from the brand. It brews a well-extracted espresso each and every time and the steam wand works wonders for milk-based drinks.

KitchenAid KF2 review: Price & availability

• Launched in UK and US April 2026, Australia in July 2026
• List price $799.99 / £699 / AU$1,099

The KitchenAid KF2 Fully Automatic Espresso Machine launched in the US and UK in April 2026, before making its way to Australia in July 2026. It has a list price of $799.99 / £699 / AU$1,099, making it the company’s most affordable fully automatic machine.

The step-up KF3 adds a small color display to improve navigation and usability, four customizable profiles and a larger menu of drink presets, with a list price $999.99 / £749 / AU$1,349.

The top-tier model, the KF4 ($1,299.99 / £899 / AU$1,599) offers much of the same functionality as the KF2 with regard to grinder and brewing unit. It gains even more recipes, a large color touchscreen and KitchenAid’s hose-based milk-foaming system, complete with milk carafe.

I’m of the opinion that the KF2’s steam wand is the best option for expertly-frothed milk and for using with non-dairy milk varieties, instantly giving the machine a leg up over its siblings. And, since the actual coffee-brewing process is the same across all three machines, the KF2 represents excellent value.

That being said, the Ninja Luxe Cafe Premier offers more in the way of features, including a dedicated cold brew setting and cold-extraction for improved iced-coffee results, along with intelligent grind and dosage settings. Its list price is less than the KF2’s at $599.99 / £549.99 / AU$799.99, although it too doesn’t offer user profiles either and requires more hands-on input.

• Value score: 4 / 5

KitchenAid KF2 review: Specifications

KitchenAid KF2 review: Design

• Compact size will suit smaller space
• Steam wand suits plant-based milks
• No dishwasher-safe components

Of the three new compact automatic models, the KitchenAid KF2 sticks out the most for a couple of reasons. Firstly, it’s only available in a matte black finish (the KF3 and KF4 also come in Porcelain White, which I’m personally a fan of), and secondly there’s no color display.

The matte black finish is still sleek and smart-looking, it must be said, although there is a shiny, fingerprint-loving panel on the front at the top. You’ll want to keep a cloth nearby to keep it looking its best.

I don’t consider the lack of a color display to be a huge negative against the KF2, as the backlit panel is bright and clear when selecting a drink and adjusting settings. The color screens on the rest of KitchenAid’s espresso machine range do offer useful assistance with features like images of the drink you want to make, but it’s not something I missed when testing the KF2.

Physically the KitchenAid KF2 is the same as the other two compact models released in 2026, measuring 14.2 x 7.7 x 18.5 inches / 361 x 196 x 470mm (H x W x D). This compact form makes it easy to place the unit on a kitchen counter, and should be especially ideal for those with smaller kitchens. A couple of built-in wheels on the bottom at rear of the machine aid with manoeuvring into position.

Interacting with the KF2 is done via five buttons on the front panel. These select the drink type, dosage and strength — each of which can be pressed multiple times to cycle through options — one to activate cleaning cycles and a final button to initiate and cancel brewing. Each button press reveals a different backlit symbol on the front panel which are relatively self-explanatory.

Drink type cycles between espresso, coffee and over ice, dose lets you choose between one or two servings (either two cups under the dispenser or one for a double serving), and strength is a scale of one to three. This refers to how much coffee is ground into the brewing unit, from 7g (setting one) to 14g (setting three).

The bean hopper recessed into the top of the KF2 is large enough to hold 250g worth of coffee beans, which is accessed via a removable lid. While this does mean there’s no visual indication as to how many beans are left, it does at least mean the beans are kept airtight and away from light, helping to retain their freshness. There’s also a small chute for pre-ground coffee, and a scoop is included in the box.

The 1.8L water tank slides in and out the rear of the machine and has an integrated handle to make the process simple, although I personally prefer a side-mounted tank like the one found on the Philips LatteGo 4400 as I find it makes refilling that little bit easier. However, the KF2’s tank does at least feature a hinged lid, allowing you to fill it up from a jug if you wish.

KitchenAid supplies a water filter with the KF2 that it claims lasts up to two months when used with standard tap water. If you use filtered water, its life can extend to up to six months. The filter can seem to be a little fiddly to install due to the water tank being so tall. Thankfully the included coffee scoop doubles up as a tool that clips onto the filter, giving you the extra length needed to screw it into place.

The drip tray is deep and extends virtually the full depth of the KF2, and has a large spout integrated into its rear for easy emptying. However, as I found with the KF3 and KF4, when you pour water out it can occasionally spill onto the front panel section as it can’t be removed. It’s not a huge dealbreaker, just prepare for a little extra clean up. Used coffee grounds are deposited into a removable bin in the center that can be tipped out when full. An icon will light up on the front panel to let you know when it’s time.

One of the standout features of the KitchenAid KF2, in my opinion, is its Pannarello steam wand used for milk foaming. I was surprised to find one used, not only because KitchenAid uses a proprietary hose-based system in the KF3 and KF4, but also because other brands such as De’Longhi and Philips have also turned their backs on steam wands (in general) in the pursuit of an entirely automatic coffee-making process.

Personally, I appreciate KitchenAid's choice to include a steam wand here, as it enables you to create genuine milk foam, even when using non-dairy alternatives. In my experience, the hose system on the KitchenAid KF3 and KF4, the LatteGo system on the Philips 4400, and the LatteCrema system on the De'Longhi Magnifica Evo Next all fell short of producing a satisfactory level of milk foam for my taste.

The use of the steam wand does mean there’s some hands-on input required, as you need to hold the jug (which isn’t included in the box) under the steam outlet until the foam reaches your desired level. There’s also no built-in thermometer, so unless you use a third-party thermometer, you’ll have to judge when the milk is ready based on touch.

Whether you are crafting a flat white or a cappuccino, the wand features a small slider for 'more foam' or 'less foam' to ensure you get the perfect texture. By opening and closing the integrated air intake, this slider regulates the airflow: increasing the intake pumps more air into the milk, which creates a more voluminous and aerated foam. Keeping it closed is ideal for texturizing milk for lattes that don’t require as high a level of foam.

The KF2 features thorough cleaning cycles, including the ability to completely evaporate the internal system of water. There are also programs for descaling and cleaning the brewing unit, which require you to press a variation of buttons to initiate. A complete set of useful instructions is included in the box. Do note that none of the removable parts are dishwasher safe.

• Design score: 4.5 / 5

KitchenAid KF2 review: Performance

• Consistently well-extracted espresso
• Steam wand delivers silky, well-foamed milk
• Lower brewing temperature for iced drinks still too hot

Having already reviewed the KF3, and knowing it uses the exact same brewing unit, I was expecting the same excellent results from the KitchenAid KF2. Fortunately my expectations were realised, as the compact KF2 produced a great-tasting espresso with each use.

Much of its success can be attributed to the IntelliGrind system that automatically determines the ideal grind size depending on your chosen beans. You can’t see the grinder at work, nor does the unit tell you what grind setting it’s chosen, you just have to put your trust in it to deliver optimal results. Your trust is rewarded.

However, if you're not completely satisfied with the crema or taste of your espresso, then the KF2 does offer the ability to manually adjust the grind to a finer or coarser level by way of a lever behind a compartment on the right-hand side of the machine. A finer setting is recommended for light-roast beans, while dark-roast beans benefit from a coarser setting. Since I was using medium-roast beans, I left the slider at the default setting in the middle.

Before you consider adjusting the grind level, you’ll first want to experiment with the strength of either the espresso or coffee drink presets. As mentioned earlier, there are only three coffee-based drink recipe presets: espresso, coffee and over ice. If you move up to the KF3 or KF4 you’ll gain recipes including latte macchiato, flat white and caffe latte, to name a few.

You can also adjust the volume of water used for each drink type, and the temperature it’s brewed at, by pressing and holding a specific set of buttons. You’ll find the full details in the supplied user manual. I was perfectly satisfied with the default settings, but it’s good to know KitchenAid offers the ability to personalize each drink.

Do note that if you do adjust any default settings, they will become the new norm until you change them again. Since there are no user profiles, you may want to consult with other people in your household before you make any tweaks.

If you want milk with your coffee, then the Pannarello steam wand is capable of producing thick, velvety foam that’s perfect for lattes, macchiatos and cappuccinos. It’s semi-automatic in use, as all you need to do is hold a milk pitcher under the wand and press the start button on the main unit (once you’ve cycled the ‘drink type’ menu through to ‘steam’). A word of warning, make sure you already have the steam wand submerged in milk. I pressed start with the pitcher still on the drip tray and ended up with milk all over the counter and myself.

A small slider on the side of the steam wand housing can be used to adjust the level of foam you achieve, either ‘less’ or ‘more’. In my tests this worked a charm, with the ‘more’ setting aerating the milk to such a high level that it nearly overflowed from the pitcher.

I much prefer the use of a steam wand compared to the hose systems used on the KF3 and KF4, as it’s capable of producing a much better level of froth. It’s also a hell of a lot easier to keep clean, as it doesn’t require constant rinse cycles to ensure the hose is free of residue. I actually wish the steam wand was featured on the company’s other machines.

Do note however that there’s no built-in thermometer to automatically detect when milk reaches your desired temperature. You’ll either have to invest in a dedicated one, or go off touch to determine when to shut the steam off. Also note that a milk pitcher isn’t included in the box. You can buy one directly from KitchenAid for $19.99 / AU$29. I wasn’t able to find a listing on the company’s UK store, but in all honesty, the pitcher available is relatively basic, so you could always get one from Amazon for the same or less money.

Where the KitchenAid KF2 falls down slightly — in a similar fashion to the KF3 and KF4 — is with its iced drinks function. KitchenAid has designed the KF2 to brew an espresso shot at a lower temperature compared to a regular shot when you select the ‘over ice’ drink type.

For starters, iced coffee fanatics will likely prefer a genuinely cold shot of coffee as opposed to having to pour a cooler shot of espresso over ice. Admittedly KItchenAid isn’t the only company to employ this function, as the Philips LatteGo machines also ask you to put some ice into a cup or glass before brewing an espresso.

However, I can’t agree with KitchenAid’s claim that the iced coffee espresso shot brews at a lower temperature, as I used a thermometer to check the temperature of a regular shot and iced shot, and both returned a measurements between 45ºC/113ºF and 50ºC/122ºF, which was hot enough to quickly melt the ice I had in my glass. It’s not necessarily a huge dealbreaker, but the fact remains there are other machines such as the De’Longhi Dedica Duo and the Jura J10 that are able to pour proper cold-extracted espresso for longer-lasting iced drinks.

It is, however, pleasing to find the KF2, like other KitchenAid espresso machines, is QuietMark-certified, signifying it should be whisper-quiet when grinding and brewing. And in reality, that’s certainly the case. I was able to record an average of 70dB while the machine was grinding and an average of 60dB when brewing. That, perhaps expectantly, puts it on par with the KF3 and KF4. That being said, we found the flagship KF8 to be quieter still with an average of 44dB during the entire coffee-making process.

• Performance score: 4.5 / 5

Should you buy the KitchenAid KF2?

Buy it if...

Don't buy it if...

KitchenAid KF2 review: Also consider

How I tested the KitchenAid KF2

I used the KitchenAid KF2 over a one-week period. I was confident knowing how to use the machine, having previously reviewed the KF3 and from using the KF4 on a daily basis at home.

I prepared all of the available drink recipes — espresso, coffee and over ice — and used a thermometer to test the temperature of the ‘over ice’ espresso shot to determine if it was actually brewed at a lower temperature, as KitchenAid claims.

I also used the integrated steam wand in both its ‘less foam’ and ‘more foam’ settings to determine if there was indeed a difference in results. I also tested it with dairy milk and a barista-style oat milk to find out if it worked well with non-dairy milk varieties.

Read more about how we test

First reviewed August 2026

The Big Bang Theory's Bernadette as a demon is something I didn't think I'd ever see in Stuart Fails to Save the Universe — but, now that I have, it somehow makes complete sense.

Last week's episode saw us meet her in a wizarding world but, in this week's entry we're swapping the magical action for something a little more One Flew Over the Cuckoo's Nest-coded.

But who, if anyone, really is crazy? And when does Stuart Fails to Save the Universe episode 4 arrive on HBO Max?

What time can I watch Stuart Fails to Save the Universe episode 4 on HBO Max?

For US viewers, Stuart Fails to Save the Universe episode 4 will drop on Thursday, August 13 at 6pm PT/ 9pm ET.

Internationally, you're looking out for these timings:

• US – Thursday, August 13 at 6pm PT / 9pm ET
• Canada – Thursday, August 13 at 6pm PT / 9pm ET
• UK – Friday, August 14 at 2am BST
• India – Friday, August 14 at 6:30am IST
• Singapore – Friday, August 14 at 9am SGT
• Australia – Friday, August 14 at 11am AEDT
• New Zealand – Friday, August 14 at 12pm NZDT

When do new episodes of Stuart Fails to Save the Universe come out?

New episodes of Stuart Fails to Save the Universe will make landfall every Thursday in the US and on Fridays everywhere else. Here are the all-important dates you need to know about:

• Episode 1: out now
• Episode 2: out now
• Episode 3: out now
• Episode 4: August 13
• Episode 5: August 20
• Episode 6: August 27
• Episode 7: September 3
• Episode 8: September 10
• Episode 9: September 17
• Episode 10: September 24

• Amazon revealed plans to build Gilroy, California data center were first submitted in 2020
• Gilroy citizens have accused the city council of inadequate communications concerning the development
• The 438,500-square-foot data center occupies a 56-acre chunk of farmland 30 miles south of San Jose

Citizens of Gilroy, California, have expressed dismay at the building of an Amazon data center, which they say they were not informed of - and when one resident attempted to lodge a complaint, she was told that comments on the plans were closed in 2024 – long before most citizens were even aware of the planned installation.

Occupying a 56-acre area of farmland between a Walmart and a local shopping center, the Amazon project appears to have been quietly pushed through using obfuscation and clever use of 45-year-old planning laws.

Amazon has apparently made provision to divert wastewater to the data center, and donated money for local spending. Meanwhile, the city mayor has responded to the concerns by likening the water-and power-hungry Amazon data center to a food distribution center that was also recently approved.

Opposition and sponsorships

Utilizing the old city zoning rule – originally intended for industrial development along a freeway – has enabled the data center to proceed into full development without planning scrutiny from residents. However, the apparent collaboration between Amazon and the Gilroy city council and mayor has not gone unnoticed, with opposition organized against the project.

One of those organizers, 21-year-old college student Landon Sepulveda, indicated to the Wall Street Journal that there was a trust issue between the population and the city administrators. “These processes were not enough for the citizens of Gilroy to do anything about it. We’ve elected them to tell us, and we shouldn’t have to watch over their shoulders.”

Another citizen, elementary school teacher Coleen Crew, told a recent council meeting “Many if not most residents first learned about this project after construction had already begun.”

Amazon and the Gilroy council have been at pains to highlight the advantages of the deal, citing jobs, tax revenue, and that solution to concerns over water use. Amazon has also donated $1 million for a new city fire truck, and donated money to non-profits and the city’s annal Gilroy Garlic Festival.

Mayor Bozzo and AWS

Responding to public dismay over the development, Mayor Greg Bozzo has highlighted an “opportunity to change the way that we do business in Gilroy. We’re adapting to the times.”

Meanwhile, Amazon Web Services’ VP of economic development Roger Wehner claims the project “followed standard approval processes.” He added: “We’ve since met directly with residents and hosted an open house to hear feedback and answer questions.”

The challenges and advantages of data centers seem to be the modern day equivalent of opposition to nuclear plants from 50 years ago. While the data center’s continued construction seems inevitable, and Amazon has made some generous provisions, it is disappointing that the city’s residents were not informed of the development – and that might be the biggest problem here.

Adding more fuel to the fire over which city is actually Australia’s best place to live, those residing in the all-seasons-in-a-day city of Melbourne have a new statistic on their side — and it all relates to fixed-line broadband speed.

Ookla, the company behind the hugely popular Speedtest tool that helps anyone diagnose their home internet speeds, has published its latest Global Index rankings, revealing median fixed-broadband download speeds by country and city.

As a country, Australia is slowly but surely moving up the table. There was a time we ranked as low as 86th, but we now sit at 50th, with a median fixed-broadband download speed of 162.34Mbps.

That improvement follows NBN Co’s September 2025 speed-tier shake-up, which introduced a 2Gbps wholesale option for eligible FTTP and HFC premises, alongside faster speeds on several existing higher-end tiers. Of course, not every Australian can access 2Gbps just yet — and, as we’ll see, the plan you choose is only one part of the equation.

For context, Singapore takes the crown with a median download speed of 447.75Mbps. One can only dream of a day when that becomes a reality Down Under.

Why the huge gap?

There are several reasons why Australia trails the pack so dramatically — although with 141 other countries clocking slower speeds, things could certainly be worse — and, in my mind, one of the most obvious is that many Australians are still connected to comparatively slow NBN plans.

The Australian Competition & Consumer Commission publishes a quarterly Wholesale Market Indicators Report that provides an overview of active NBN connections and their download data rates.

The most recent report, covering the March quarter of 2026 and published in July, revealed there were just over 2.665 million active 50Mbps NBN connections, out of a total of 8.839 million residential broadband services. That works out to around 30.2%.

A 50Mbps plan naturally puts a ceiling on the speed a customer can record in a Speedtest. If enough users on lower-speed tiers are represented in Ookla’s sample, that will bring the national median down.

One relatively straightforward way for Australia’s international standing to improve, then, is for more eligible households to take advantage of NBN Co’s free fibre upgrade. The program is open to millions of eligible FTTN and FTTC premises, allowing them to upgrade to FTTP if they take out an eligible high-speed service — such as one of the best NBN 500 plans.

A fixed-line civil war

Move the slider over to ‘City’ and the results become much more varied, with cities from North and South America, Asia and Europe all jostling for position. Valparaíso in Chile takes the crown with a huge 425.84Mbps, followed by Abu Dhabi with 399.92Mbps, for example.

Australian cities evidently have some work to do. Scroll down to position 64 and you hit Melbourne with a median download speed of 177.69Mbps — faster than Australia’s national median, it must be said — while Sydney trails behind in position 76 with a meagre 149.91Mbps.

A difference of 27.78Mbps might not sound like much, but Melbourne’s median result is around 18.5% faster than Sydney’s. When intercity tensions are already high, that’s a huge bragging right for the southerners.

Wait, we can (try to) explain

Nobody ever wants to be seen as the weaker of two, and while the reported figures are great news for people in Victoria, there is potentially a plethora of factors at play that pave the way for a Melbourne victory.

We’re not here to make excuses, of course, but there are plenty of unknowns.

Firstly, there could simply be a larger proportion of Melburnians connected via superior fibre-to-the-premises infrastructure. FTTP allows households to access the fastest NBN plans, including the newer multi-gigabit options where available.

While it’s not totally conclusive, there is some evidence to support this idea. That evidence is publicly available technology-mix figures, although they’re reported at a state level, rather than for Greater Melbourne and Greater Sydney specifically. Data compiled from NBN rollout information indicates Victoria has a higher share of premises using FTTP than NSW — roughly 66% versus 60% — while it also indicates there are more active connections on the inferior fibre to the node (FTTN) connection type in NSW too.

This doesn’t completely settle the argument. The figures not only include regional areas, but they also show the technology available to a property, as opposed to an active plan supplying an address.

All fixed-line connections are welcome

Furthermore, on the subject of NBN, Ookla’s Speedtest results don’t only capture NBN connections, but fixed-broadband tests from other fixed networks too, including private fibre operators. Both Melbourne and Sydney are home to several such networks, including Vocus, GigaComm and OptiComm, while Melbourne also has DGtek — a private FTTP network concentrated in the city’s CBD and inner south-east.

Exact figures for active connections on those networks are tricky to come by. But if Melbourne has a higher concentration of people connected to these kinds of high-speed private fibre services, they could be recording considerably quicker test results and helping push its city median upwards.

Not telling the complete story?

Another excuse potential explainer is the way Ookla collects its data. Whereas the ACCC’s Measuring Broadband Report — which published its final instalment in June 2026 — monitors a selected group of Australian homes across nine popular ISPs, Ookla receives data when people choose to run a Speedtest.

This doesn’t mean someone repeatedly checking their connection can single-handedly skew the data, as Ookla averages each unique user’s results into a single sample for the relevant time period and location. Still, the results are dependent on who chooses to test. Melbourne’s sample could contain a different mix of internet plans, connection technologies, devices and Wi-Fi setups than Sydney’s.

When did you last check your Wi-Fi router?

Home equipment can make a considerable difference too. A household with a Wi-Fi 5 router tucked away in a cupboard, a busy 2.4GHz network and a phone testing from the other end of the house could record a much slower result than the same connection tested over Ethernet.

On the flip side, a newer Wi-Fi 6E or Wi-Fi 7 router, a properly positioned mesh system and a compatible device can give a household a far better chance of getting close to the speeds it actually pays for. That doesn’t make the underlying internet plan any faster, of course — it simply means the home network is less likely to be the bottleneck.

If your own Speedtest result seems suspiciously low, it may be worth checking out our guide to the best mesh Wi-Fi systems to find out if your home could benefit from an upgrade.

And, trying to claw a point back for Sydney, it can be argued that people often run a speed test when their internet connection starts lagging. While Melbourne’s median tested speed is higher, that doesn’t necessarily mean every connection in the city is consistently faster or more reliable.

• Virginia now makes data centers pay for their own power lines
• Virginia hosts 570 data centers, more than any other US state
• Spanberger says the policy could save residents hundreds of millions

The US state of Virginia is requiring data centers to cover electricity infrastructure costs created specifically by their rapidly growing power demands.

The move comes as electricity prices across the state have risen sharply alongside the expansion of artificial intelligence facilities.

State officials say the new requirement could prevent ordinary electricity customers from carrying costs associated with large data center developments.

Data centers face higher infrastructure costs

Virginia has become the largest data center hub in the United States, with at least 570 facilities operating across the state.

Their growing electricity consumption has required major investments in transmission infrastructure, increasing pressure on the regional power system.

The State Corporation Commission, Virginia’s public utility regulator, has ordered data centers to pay for transmission infrastructure used exclusively by their facilities.

The requirement means utilities can no longer spread those dedicated infrastructure costs across their wider customer base.

Electricity prices in the region have already increased substantially, with PJM Interconnection raising prices by 76% during recent capacity market changes.

Monitoring Analytics, an independent market watchdog, has attributed much of that increase to growing demand from data centers.

The issue has become politically important as artificial intelligence companies continue expanding computing capacity across Virginia.

Some local authorities have even asked employees and institutions, including schools, to reduce electricity consumption during periods of high demand.

"We are taking real steps to address rising energy costs for Virginians," said the Governor of Virginia, Abigail Spanberger.

"I will continue to work with the General Assembly and take action to make sure data centers pay their fair share, adhere to strict environmental standards, and listen to the concerns of local communities.”

Washington’s promise reaches the states

The Virginia decision follows President Donald Trump’s effort to make major AI companies finance infrastructure required by their expanding operations.

Trump met leading AI companies at the White House in March and urged them to “pay their own way” for necessary infrastructure.

The administration expanded that approach in July by calling on governors, utilities, and data center developers to follow similar principles.

Critics have questioned whether voluntary commitments can significantly change electricity costs without enforceable state or federal requirements.

Virginia is currently among the few states taking concrete action against rising electricity costs linked to large electricity users.

Oregon has also introduced higher electricity rates for data centers consuming at least 20 MW, under provisions contained within its POWER Act.

Portland General Electric subsequently increased rates for those large consumers by 30%, while residential electricity costs fell by approximately 1.3%.

The Virginia approach differs by focusing specifically on infrastructure dedicated to individual data center projects.

The policy could become increasingly important as AI companies expand AI tools requiring increasingly powerful computing facilities and greater electricity supplies.

If other states adopt similar requirements, data centers could face more direct responsibility for infrastructure costs they generate.

For Virginia households, however, the immediate question is whether the policy actually prevents future electricity increases rather than simply changing who pays existing infrastructure bills.

Spanberger said the state expects the measure to save Virginians “hundreds of millions of dollars” while requiring data centers to pay their fair share.

Via Tom's Hardware

Quordle was one of the original Wordle alternatives and is still going strong now more than 1,500 games later. It offers a genuine challenge, though, so read on if you need some Quordle hints today — or scroll down further for the answers.

Enjoy playing word games? You can also check out my NYT Connections today and NYT Strands today pages for hints and answers for those puzzles, while Marc's Wordle today column covers the original viral word game.

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

Quordle today (game #1660) — hint #1 — Vowels

* Note that by vowel we mean the five standard vowels (A, E, I, O, U), not Y (which is sometimes counted as a vowel too).

Quordle today (game #1660) — hint #2 — repeated letters

Quordle today (game #1660) — hint #3 — uncommon letters

Quordle today (game #1660) — hint #4 — starting letters (1)

If you just want to know the answers at this stage, simply scroll down. If you're not ready yet then here's one more clue to make things a lot easier:

Quordle today (game #1660) — hint #5 — starting letters (2)

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

Quordle today (game #1660) — the answers

The answers to today's Quordle, game #1660, are…

• ONSET
• BRAWL
• TEPID
• SLANG

Sometimes I like to look at the four Quordle words as a creative writing task and imagine a story, or the opening of a story based on them.

Most days, it’s impossible but today is a good one — an urban tale of young men who are quick with their fists but whose lack of streetwise vernacular is embarrassingly poor.

Daily Sequence today (game #1660) — the answers

The answers to today's Quordle Daily Sequence, game #1660, are…

• DONUT
• FLESH
• PLANK
• BIDDY

Quordle answers: The past 20

• Quordle #1659, Monday, 10 August: GAVEL, CRUSH, SPEED, ETHOS
• Quordle #1658, Sunday, 9 August: KAPPA, WIDEN, FOLIO, STEEP
• Quordle #1657, Saturday, 8 August: FUROR, BUGLE, REUSE, SCOOP
• Quordle #1656, Friday, 7 August: AMISS, SHIRK, SALSA, RABBI
• Quordle #1655, Thursday, 6 August: AMUSE, BRAND, CEDAR, BLURB
• Quordle #1654, Wednesday, 5 August: MOUND, WRONG, STOOD, WEIRD
• Quordle #1653, Tuesday, 4 August: GOOEY, TARDY, HUMAN, CHALK
• Quordle #1652, Monday, 3 August: FUNGI, MUDDY, DEBUT, MANIA
• Quordle #1651, Sunday, 2 August: BOOTH, EASEL, REACH, MAKER
• Quordle #1650, Saturday, 1 August: OMEGA, BIRCH, SMOCK, EERIE
• Quordle #1649, Friday, 31 July: MANIA, SINGE, STOOL, LOFTY
• Quordle #1648, Thursday, 30 July: MODAL, IDLER, CHUMP, LUMPY
• Quordle #1647, Wednesday, 29 July: GAWKY, ALLOY, AUDIT, LATCH
• Quordle #1646, Tuesday, 28 July: SMITE, ROACH, RUDDY, GLOBE
• Quordle #1645, Monday, 27 July: TAFFY, FIGHT, GUILD, WAGER
• Quordle #1644, Sunday, 26 July: STOMP, LIMIT, FUNKY, STEAD
• Quordle #1643, Saturday, 25 July: SALON, SHEEN, BURST, GOURD
• Quordle #1642, Friday, 24 July: DOLLY, EDIFY, LAGER, PRANK
• Quordle #1641, Thursday, 23 July: LADEN, EVICT, WOVEN, SHIRE
• Quordle #1640, Wednesday, 22 July: COYLY, SINGE, AWASH, AWOKE

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 #891) - hint #1 - today's theme

NYT Strands today (game #891) - hint #2 - clue words

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

• CHEEP
• HINTS
• CHIVE
• COME
• TRACE
• MANOR

NYT Strands today (game #891) - hint #3 - spangram letters

NYT Strands today (game #891) - hint #4 - spangram position

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

NYT Strands today (game #891) - the answers

The answers to today's Strands, game #891, are…

• TOMATO
• ZUCCHINI
• PEPPER
• MELON
• CORN
• CARROT
• SPANGRAM: SUMMERHARVEST

• My rating: Easy
• My score: Perfect

One for the veggie lovers among us.

ZUCCHINI took me a while to put together, in part because I come from a land where we call zucchini courgettes, but mainly because I can’t spell and missed the double C.

This misstep aside I sped through today’s game, finding it harder to find a decent crop of non-game words.

Yesterday's NYT Strands answers (Monday, August 10, game #890)

• DATA
• FACTS
• FIGURES
• STATISTICS
• SPECIFICS
• SPANGRAM: PROOFPOSITIVE

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 #1157) - today's words

Today's NYT Connections words are…

• TOAST
• AYE
• BRAVO
• SOLD
• BET
• DONE
• HISTORY
• KNOWS
• HARE
• GOING
• THROUGH
• TWICE
• ONCE
• SUNK
• TUNG
• USA

NYT Connections today (game #1157) - hint #1 - group hints

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

• YELLOW: Well and truly over
• GREEN: On your TV
• BLUE: Said after the final bid
• PURPLE: Sounds like it’s on your face

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 #1157) - hint #2 - group answers

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

• YELLOW: DOOMED
• GREEN: BASIC CABLE CHANNELS
• BLUE: WORDS IN AN AUCTIONEER'S CHANT
• PURPLE: HOMOPHONES OF FACIAL FEATURES

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

NYT Connections today (game #1157) - the answers

The answers to today's Connections, game #1157, are…

• YELLOW: DOOMED: DONE, SUNK, THROUGH, TOAST
• GREEN: BASIC CABLE CHANNELS: BET, BRAVO, HISTORY, USA
• BLUE: WORDS IN AN AUCTIONEER'S CHANT: GOING, ONCE, SOLD, TWICE
• PURPLE: HOMOPHONES OF FACIAL FEATURES: AYE, HARE, KNOWS, TUNG

• My rating: Easy
• My score: Perfect

It was very pleasant to see some tiles made up of single words after a run of two-worded tiles, but it didn’t make pouting the four groups together any easier.

That is until I had a blinding moment of revelation and remembered that I had watched the English Premier league on USA Network and BET, BRAVO and HISTORY were also TV stations of note, not that I know them as BASIC CABLE CHANNELS.

From here it all came together like a charm and I wondered why I was ever confused. That, though, is part of the magic that is Connections: bafflement and enlightenment within seconds.

Yesterday's NYT Connections answers (Monday, August 10, 2026, game #1156)

• YELLOW: HAPHAZARDLY: ANY OLD HOW, AT RANDOM, HELTER-SKELTER, WILLY-NILLY
• GREEN: WAYS TO RECEIVE EVENT TICKETS: MAIL DELIVERY, MOBILE TICKET, PRINT-AT-HOME, WILL CALL
• BLUE: SECTIONS ON AN INVOICE: BALANCE DUE, BILL TO, ORDER NUMBER, UNIT PRICE
• PURPLE: FAMOUS GOATEES: BILLY GOAT, COLONEL SANDERS, DOCTOR STRANGE, THE DEVIL

• Firestorm Labs printed 12 drones aboard USS Essex during a Pacific voyage
• More than 1,000 drone components were manufactured directly aboard the warship
• The Navy tested drone production while waves reached 12 feet

Firestorm Labs has produced 3D-printed drones aboard a US Navy ship at sea, testing manufacturing without conventional shore-based logistics support systems.

The company manufactured more than 1,000 components and assembled 12 Squall first-person-view drones aboard USS Essex during its voyage toward Hawaii waters.

The drones were subsequently flown as opposing aircraft during a counter-drone exercise conducted during the Rim of the Pacific, or RIMPAC, 2026.

Manufacturing drones at sea

Firestorm Labs used its xCell containerized microfactory to manufacture drone components and additional parts while the amphibious assault ship remained underway.

The system functions as a compact production facility designed for Navy vessels, allowing crews to manufacture unmanned aircraft and selected replacement components.

During the trial, service members assembled drones while waves reached 12 feet, providing a demanding environment for onboard manufacturing operations.

Personnel also printed mechanical test components used to assess the effectiveness and performance of the equipment during production aboard the vessel.

The same equipment was used to fabricate military repair parts required by USS Essex personnel during the ship’s Pacific deployment.

“Every part xCell printed on deck is one that doesn’t need to be flown or shipped across contested waters — cutting the fuel, aircraft hours, and personnel it takes to keep a ship operational,” Firestorm Labs said via LinkedIn.

“Repairs that once meant days or weeks waiting on a resupply run can now happen in hours, on station.”

Under the company’s description, repairs requiring conventional supply missions could instead be completed within hours when suitable production capability remains available.

The trial also gives Navy personnel experience operating a manufacturing system while managing shipboard conditions, equipment requirements, and production tasks simultaneously.

A wider move toward containerized military capabilities

The experiment comes as the Navy develops modular capabilities transported inside containers and deployed where conventional infrastructure remains limited during operations.

Chief of Naval Operations Adm. Daryl Caudle announced a containerized capability campaign plan in March during the McAleese Defense Programs conference.

His plan calls for transportable containers carrying capabilities such as drones and weapons to be deployed across different operational regions.

The Pentagon has separately pursued containerized munitions, with framework agreements announced in May involving Anduril, CoAspire, Leidos, and Zone 5.

Those agreements cover more than 10,000 containerized missiles scheduled for acquisition over three years beginning in 2027 under the program.

The Navy’s onboard drone production test combines this containerized approach with additive manufacturing conducted directly from an operating warship.

For crews operating far from established bases, such production could provide another method for obtaining selected drones and replacement components when needed.

However, the trial does not establish how reliably the system can sustain production during prolonged operations or under more demanding combat conditions.

Its practical value will depend on whether the equipment can consistently produce dependable aircraft and parts at useful rates during extended deployments.

Via Defense News

The programmer Richard Stallman is not a household name, but his work has been instrumental in building the software industry, as has his long-term campaign for free software. He's also been a huge advocate for privacy in the digital age, and has railed against the rise of cell phones for that reason.

Who needs phones anyway?

Stallman first disclosed his views on cell phones in an interview with Network World, during a time in which smartphones were exploding in popularity.

During this interview, Stallman indicated his long-held belief that the portable phones that many millions use would be the perfect tool that authoritarian forces could exploit and use to track the movements of populations.

He also advocated for free software, which you would expect from the founder of the Free Software Foundation (FSF), which he established in 1985. This was backed by the creation of the GNU project – a free software, mass collaboration movement to give users freedom of choice to use and develop software for their devices.

The legacy of free software

Despite his reluctance to ever use a cell phone, one of Stallman's achievements – which he himself acknowledged in the interview – was the third-party version of the Android mobile OS, from which all proprietary software was stripped out.

He pointed to new systems like Replicant, an alternative version of Android, that can run on certain devices without additional proprietary software. The catch is that this only works with older and outdated handsets, like the Samsung Galaxy S3 or the Galaxy Note 2.

What is AI good for? It's a question I get asked and ask myself almost every day. Unlike a wrench, hammer, or screwdriver, which are each good for one or two tasks, AI is amorphous and intimidating in its breadth. It seems capable of almost anything, especially as it changes and grows more intelligent. People pump it up as a do-it-all wonder, and yet, the question remains: what would I do with it?

Often, I find the answer to that question in the moment. For instance, as a partially color-blind person, I struggle to match clothing. Now, I often ask Gemini. I'll take my iPhone 17 Pro Max, open Gemini, turn on Gemini Live, pick out a shirt or pants (or both), and ask if they go together or which option is best. I always get an answer and, honestly, I usually follow the uncomplicated AI-generated advice.

Success in one area of your life with technology usually leads you to try it in another, especially with the ever-fungible AI.

In my house, I'm known as the risk taker, at least when it comes to food. "Best by" is a suggestion. "Use before,' is friendly advice. My family often blanches at my aged food consumption habits. Yes, I'll eat that five-day-old steak. Those English Muffins expired two weeks ago? Slide them over.

Enter the risk-taker

I know, it's not great, and when it was lunchtime this weekend I slid the week-plus-old store-bought-and-made chicken salad from the fridge and considered making a sandwich.

Staring at the "made on" date as if my gaze might rearrange the figures, I realized that I might be taking a risk. So I opened the container and sniffed the still spry-looking salad. It's at moments like this that I wonder, "What am I doing? Does my nose really know the difference between safe and stale or, more importantly, safe and decidedly turned?" I can tell you with some confidence: it does not.

Still, having grown up in a house where money was tight and you rarely threw anything away, I was hesitant to dump this half-a-pound of chicken salad (slightly turned or not).

I stood there in my kitchen for a minute, weighing my options and thinking that slightly sour chicken salad might not, with the right toast, be that bad.

Then I looked at my phone, lying face down on the counter, but surely judging me.

I would ask Gemini.

Putting Gemini on the menu

After launching the app, I turned on Gemini Live and pointed the camera at the open container. "Hey, this chicken salad was made on 8/1. Do you think it's still good?" I asked hopefully.

Gemini thought for a moment and then responded, but not in the way I expected.

Before giving me the answer, Gemini launched into what sounded a bit like one of those disclaimers for an Ozempic TV commercial:

"This information is not intended to be a substitute for professional medical advice, diagnosis, or treatment. Always seek the advice of your physician or other qualified health provider with any questions you may have regarding a medical condition."

Clearly, I struck a nerve. I only asked about the chicken salad, not if I should have a kidney removed.

I get it. If I'd asked this question a couple of years ago, Gemini might've launched right into an analysis, but concerns about AI accuracy and understanding of human needs and context have risen since the emergence of generative AI. Sure, it's way smarter and less likely to give you false or hallucinated information, but it's also, well, not a doctor or even a dietician.

A recent Gallup poll found that more than 25% of Americans are using AI for health info or advice, and of them, more than half query AI for health-related advice and analysis before visiting a doctor. The problem is that AI is not a medical professional, and while it can dig through the billions of health- and medical-related bits stored online, it doesn't have a medical license. It can't understand or treat your various maladies.

In my case, though, Gemini drew a long thin line from my seemingly innocuous question about chicken salad freshness to a potentially incorrect answer, my possible gastric emergency, and a potential lawsuit against Google and Gemini brought on my behalf by my surviving family members.

Look, food poisoning isn't usually fatal, at least according to Healthline, but it is yucky and uncomfortable. I think Gemini was just trying to avoid me taking its word as law on this topic and a potentially bad outcome for both of us.

I'm still hungry

But what about eating this rapidly warming chicken salad?

After the voluminous disclaimer, Gemini got to the heart of the matter:

"According to food safety guidelines, chicken salad should be stored in the refrigerator for only three to four days. Since this chicken salad was made on August 1st and today is August 9th, it is no longer safe to eat and should be discarded."

The answer, based less on the looks of this chicken salad and more on general advice that I often ignore, ultimately had the desired result.

You and my family will be pleased to know I took Gemini's advice, threw out the chicken salad, and made myself a peanut butter and jelly sandwich.

Naturally, I asked Gemini's opinion:

"Yes, absolutely! A PB&J is a much better and safer option right now.

Peanut butter and jelly have a significantly longer shelf life and aren't prone to the rapid bacterial growth that affects meat-based, mayonnaise-heavy dishes like chicken salad. It's a classic, risk-free, and delicious backup plan. Enjoy your lunch!"

Fine. Thanks, Gemini, and bon-appétit.

• Washington wants new rules for civilian nuclear reactors operating at sea
• IAEA will launch ATLAS as America pushes nuclear power offshore
• Small modular reactors could power ships without frequent nuclear refueling

The United States government is backing an international effort to clear regulatory barriers around civilian nuclear reactors operating on ships and floating platforms.

The initiative could eventually enable floating power plants serving coastal communities and industries requiring substantial electricity or desalination capacity.

The International Atomic Energy Agency (IAEA) will launch its Atomic Technologies Licensed for Applications at Sea (ATLAS) initiative in Washington on August 26 and 27, aiming to clear the path for reactors at sea.

US pushes for nuclear power at sea

"The global maritime sector is at a critical turning point, facing urgent pressure to sustain long-distance, high-speed operations while ensuring reliability and energy security," said Rafael Mariano Grossi, Director General of the IAEA.

"Small modular reactors offer a safe and viable option for maritime transport and offshore energy systems, delivering high energy density and long operating cycles that eliminate the need for frequent refueling."

Ahead of the official ATLAS launch, the United States will hold an Industry Day event on August 25 to showcase relevant technology.

American nuclear and maritime companies are expected to attend, each hoping to secure a share of any emerging global market.

US Secretary of Energy Chris Wright tied the event to the Department of Energy's wider push for more atomic power.

"DoE remains focused on unleashing American energy dominance, accelerating innovation, and advancing sources of energy that are affordable, reliable, and secure for the American people," Wright said.

Floating reactors carry old and new risks

The Department of Energy is counting on small modular reactors, whose high energy density and long operating cycles could suit ships and offshore installations.

Nuclear-powered vessels are not new to America, since every active US Navy aircraft carrier already runs on atomic propulsion today.

The Navy once operated nuclear-powered cruisers as well, including the Virginia class, before retiring them for being too costly to run.

Civilian nuclear ships also have precedent, since the NS Savannah became the first nuclear-powered merchant vessel when it launched in 1959.

That vessel proved too expensive to operate, a challenge small modular reactors (SMR) advocates now hope newer reactor designs can finally overcome.

Russia currently operates the world's only floating nuclear power plant, the Akademik Lomonosov, which first launched back in August 2019.

Its two KLT-40S reactors can generate a combined 70 MW of electricity, enough to serve remote coastal communities and light industry.

Placing a reactor on water changes the associated risks rather than eliminating them, since remote platforms may lack backup power.

The surrounding sea can serve as a vast heat sink for cooling, though emergency access could prove more difficult offshore.

Any accident could also risk radioactive contamination of nearby marine environments, endangering the very coastal community the reactor intended to serve.

The push toward floating and shipborne reactors suggests nuclear power is being reconsidered as a serious option for maritime and offshore energy needs.

Whether SMR technology can finally make such projects economical, after decades of costly false starts, remains to be seen.

Via The Register

• Brand called Saccater lists 'aéra' IEMs on Kickstarter...
• ... and very quickly jumps past funding targets
• Designed to protect your hearing (and they look good too)

About one third of adults in the UK, and a similar number in the US, have some kind of hearing loss (according to RNID and Hearing Health Foundation respectively).

It stands to reason that those little music-playing speakers we place into our ear canals, really rather close to our ear drums (with their ever increasing driver sizes and/or driver counts nestled into the headshells) don't exactly help. But one company is offering something of an alternative.

On Kickstarter, a brand called Saccater has unveiled some new in-ear monitors called aéra. These are audiophile-friendly buds with 9.6mm planar magnetic drivers, silver-plated copper wire and dual-sided concentric coils, but with some added extra considerations to protect your hearing.

That fits with Saccater's other Kickstarter projects; a series of earplugs designed for various everyday tasks. The aéra have already been fully funded on Kickstarter, with HK$210,000 (about $26,000, £19,000, AU$38,000) already pledged out of the modest HK$30,000 (roughly $4,000, £3,000, AU$6,000) goal. Each pair of buds cost HK$860 (about $110, £74, AU$150), and are due to ship in October 2026.

That means the Saccater aéra have hit their goal nearly seven times over, proving there's a big appetite for hearing-friendly earbuds.

Hear, hear(ing health)

You're probably wondering, what does 'hearing-friendly' actually mean in IEMs such as these — because they look just like the IEMs I know and love?

One of the key considerations is weight and design; according to Saccater, you perceive sound worse when your ear is uncomfortable, often prompting people turn turn up the volume higher than they need to. So the aéra use a "zero-pressure fit" and are incredibly light, at just 4g per bud.

Whether or not they actually protect your hearing health, then, they'll be pretty comfortable to wear.

The earbuds also come with a hearing manual to guide you through best practice to avoid damaging your ears, as well as the brand's other earplugs in the box, so you don't use the aéra just to block out noise. This latter is, according to Saccater, another way people unintentionally damage their hearing.

Despite the focus on hearing health, the aéra seem suitable for audiophiles too. They support 24 bit / 96kHz playback, a 20Hz-20kHz frequency response, and come with a range of ear tips to choose between.

They also look pretty simple and beautiful, ditching the shiny or textured designs of most modern IEMs for something understated.

Now, all of these specs and pledges are from Saccater itself, so it's hard to know just how effective the aéra will be. But the early success of the buds goes to show that hearing health is something we're all waking up to.

• SpaceX and Tesla commit $16.8 billion to build a gigantic Texas chip factory
• TeraFab could cover more than 100 million square feet when fully built
• Elon Musk claims TeraFab will reach 50 times the size of the Pentagon

SpaceX and Tesla have revealed more details in their plan to invest $16.8 billion in a massive semiconductor factory planned for Grimes County, Texas.

The facility, known as TeraFab, is expected to manufacture chips for several computing-intensive applications associated with both companies.

At full construction, SpaceX says the factory could exceed 100 million square feet, or approximately 9.3 million square metres.

A semiconductor factory on an unprecedented scale

SpaceX CEO Elon Musk described the planned facility as potentially the largest and most valuable building on Earth after completion.

“TeraFab will be 50 times the size of the Pentagon when complete,” Musk said on X.

Since the Pentagon is 6.6 million square feet and SpaceX has only confirmed TeraFab will exceed 100 million square feet, Musk's "50 times" figure would require a final size closer to 330 million square feet — a number the company has not disclosed or confirmed.

Even so, the scale remains extraordinary: the Pentagon, Apple Park, Mall of America, and Giga Texas together total 25.04 million square feet, meaning TeraFab's confirmed 100 million square feet alone would already be roughly four times larger than all four combined.

The project originally emerged from Musk’s concern that existing semiconductor suppliers could not satisfy Tesla’s future computing requirements.

Tesla first discussed the concept publicly in November 2025, describing it as a much larger version of its Gigafactory model.

Plans announced in March 2026 called for the semiconductor plant to operate beside Tesla’s Austin headquarters.

SpaceX and Tesla subsequently established the project as a joint venture, with Intel later joining the development effort.

Musk has said the facility would use Intel’s 14A manufacturing process, although production plans may evolve during development.

The companies have linked the factory to chips required for space-based computing, advanced vehicles, and humanoid robotic systems.

Costs, employment and environmental commitments

The initial project estimate was $20 billion, but later filings indicated substantially larger potential spending during subsequent development phases.

A May property tax abatement application revealed that an initial phase could cost $55 billion, with eventual spending potentially reaching $119 billion.

The latest $16.8 billion commitment, therefore, does not represent the maximum investment that may eventually be required.

SpaceX expects the facility to employ at least 3,000 workers, including people from Grimes County and neighboring Brazos County.

Local authorities approved a 100% property tax abatement in June 2026 after county commissioners voted 4-1 on three related proposals.

The companies also say industrial operations will use water from Gibbons Creek Reservoir instead of relying on nearby groundwater supplies.

SpaceX and Tesla have also committed to wastewater treatment, water reuse, conservation measures, and compliance with environmental regulations.

Their plans include regulated handling of hazardous materials, chemicals, and industrial byproducts generated during semiconductor manufacturing.

“The combined SpaceX and Tesla demand for chips is expected to be in excess of 1 terawatt (TW) of compute,” said SpaceX in its website post.

The proposed scale makes TeraFab an unusually large industrial project, while its final cost and construction scope remain subject to future expansion.

Via DataCenterDynamics

• TSMC reportedly holding roughly $1 billion of finished A20 Pro wafers that cannot be packaged because the DRAM they must be integrated with has not yet arrived
• The bottleneck exists because Apple moved the A20 Pro to TSMC's WMCM packaging, which bonds processor and memory at the wafer level, removing the option of stockpiling finished dies and adding memory later
• The launch is not expected to slip but could mean that post-launch inventory could be dicey, with thin stock and extended delivery estimates the likely outcome for Apple's upcoming devices

Around a billion dollars of finished Apple silicon is reportedly sitting at TSMC with nowhere to go, reports have claimed.

The wafers are processed, the yields are fine, and the chips cannot move to the next stage of production because the memory they need to be packaged with has not arrived.

The reason they cant be stockpiled for later use is a design choice Apple made with TSMC to use its Wafer-Level Multi-Chip Module (WMCM) process instead, an extension of the CoWoS packaging family adapted for mobile parts.

Trouble in smartphone chip paradise?

The reported delay was revealed by Tim Culpan, a Taipei-based semiconductor analyst, who says production of the A20 Pro on TSMC's N2 node is progressing well, with good volume and yields, but wafers are piling up waiting for DRAM before they can be packaged and sent onwards.

These chips will almost certainly power the next generation of Apple devices (probably the iPhone 18 Pro) but there might be a slight twist in the making due to design decisions made by Apple.

Memory shortages delay products all the time, but this one is worth attention because of what Apple changed about how it builds its processors.

For roughly a decade, TSMC packaged Apple's iPhone processors using Integrated Fan-Out, or InFO. Under that approach, the logic die containing the CPU, GPU, and Neural Engine is first fabricated and diced from the wafer, and memory is attached afterward, stacked on top of the finished die in a later assembly step.

The sequencing matters: a memory shortage under InFO delays final assembly but does not stop anything upstream. You keep fabricating logic dies, stockpile them, and attach DRAM when supply arrives.

The A20 Pro, as mentioned previously, uses TSMC's WMCM process instead, an extension of the CoWoS packaging family adapted for mobile parts. WMCM integrates the processor and DRAM at the wafer level, side by side, before either can proceed.

That reverses the sequence. TSMC needs the memory in hand to complete the package, and it cannot be added at a later step, which means Apple cannot bank finished dies against a memory delivery that has not happened yet.

There is a certain amount of irony here: Apple reportedly moved to WMCM specifically to fit more DRAM at lower latency for on-device AI features, since placing memory on the wafer improves both, but an AI-related memory shortage has stalled its progress.

The memory in question is LPDDR5X, and it is scarce because AI infrastructure buildout has absorbed the manufacturing capacity that would otherwise serve smartphones. Apple's on-device AI ambitions are being throttled by the datacenter version of the same ambition.

Culpan reports that Apple and its assemblers remain confident they can meet initial demand and that launch-day orders will be fulfilled without delaying the launch, though we might see thinner supply as a result of the delay.

Apple's exposure runs deeper on the other side of launch day: main logic board assembly and final assembly are largely handled by Foxconn and BYD Electronics, and late deliveries of packaged processors compress their schedules, even if Apple ultimately secures the memory.

The likely outcome is thin retail inventory, a fast preorder sellout, and delivery estimates that stretch out for weeks rather than a phone that does not exist. Apple reportedly plans to ship around 200 million units of the upcoming iPhone across the lineup over its lifespan, against an iPhone 17 range that surpassed 245 million shipments in 2025, so there isn't much room for error here.

Apple's usual LPDDR5X route runs through Micron, with SK Hynix and Samsung alongside, and an attempt to widen it fell over: a reported approach to Chinese memory maker CXMT, apparently aimed at cheaper pricing, did not produce a deal. Renegotiations with existing suppliers reportedly have not landed either, which leaves Apple paying whatever the current market asks. Whether it secures enough supply to avoid the thin-inventory scenario remains to be seen.

Hisense UR8 review: Two-minute review

The Hisense UR8 is the first ‘affordable’ RGB TV on the market, and while it has a strong feature set and mostly good picture quality, the UR8 doesn't demonstrate the ‘OLED-killer’ status the RGB tech has been given. While it’s a good TV, it can’t count itself among the best TVs I've tested this year.

The UR8 has a good number of smart and AI features, as well as features for gaming. It also has four HDMI 2.1 ports, an anti-glare screen akin to Samsung’s own Glare Free screens, and comes equipped with Hisense’s Hi-View AI RGB processor.

Picture quality on the UR8 is good, but delivers some disappointments. It demonstrates excellent backlight control for an LED, backlit TV, as well as strong contrast, decent black levels, and crisp textures. Its motion is also good, even with out-of-the-box picture modes, though I did note some blurring at times.

However, as this is an RGB TV, its big selling point should be its colors, and while they are natural, and can appear punchy, they’re not as bold as I hoped for, especially given its ostensibly impressive test measurements (more on that below). While RGB is a step forward in some aspects for mini-LED, TVs using the technology still can’t compete with the best OLED TVs.

Sound quality is solid. The UR8 delivers some weighty bass, good accuracy and precise sound mapping, and delivers clear speech in movies and TV shows. The soundstage can sound narrow, however, and Dolby Atmos effects aren’t always clear. While the built-in sound is good for the most part, home theater fans will likely want to supplement it with one of the best soundbars.

Gaming is one of the UR8’s strengths. With four HDMI 2.1 ports, all kitted out for gaming, it supports up to 4K 180Hz and VRR. When actually playing games, the UR8 feels responsive and delivers a smooth experience. Getting a low input lag time does require adjustment of settings and the picture isn’t always perfect; but while it doesn’t beat the best gaming TVs, the UR8 is still good for gaming.

Value is a mixed bag with the UR8. The 65-inch model I tested launched for $1,999 / £2,299 / AU$2,495, and while prices are already dropping in the US, and it's great value in Australia, its price in the UK puts it in direct competition with some of the best OLED TVs, like the LG C6. While it does deliver good performance and features for the price in the US and Australia, the same isn’t true for the UK.

Hisense UR8 review: Prices & release dates

• Release dates: June 2026 (US) / July 2026 (UK)
• 55-inch: $1,299 / £1,699 / AU$1,895
• 65-inch: $1,999 / £2,299 / AU$2,495
• 75-inch: $2,499 / £2,999 / AU$3,195
• 85-inch: $3,499 / £3,499 / AU$4,695
• 100-inch: $5,999 / £7,999 / AU$N/A

The UR8 is the Hisense's entry-level RGB mini-LED TV, sitting below the flagship Hisense UR9, but above TVs in the brand's standard mini-LED range such as the Hisense U7S. It’s available in sizes 55-85-inch in most regions, and available in 100-inch in the US and UK (there's no listing for an Australian model yet). The 65-inch model I tested launched for $1,999 / £2,299 / AU$2,495.

Since its release, the UR8’s prices have already dropped in the US, with the 65-inch now available for $1,499. The 55-inch has dropped as low as $999, putting it in the same price bracket as some standard mini-LED models. The same can’t be said in the UK, where its prices are not only higher than standard mini-LED competition, but even pricier than the LG C6 OLED.

Hisense UR8 review: Specs

Hisense UR8 review: Benchmark results

Previous

1 / 4

Next

Brightness measurements

Taken using a colorimeter and test pattern generator. Click the test name for measurements and select the chosen test from the dropdown menu.

HDR Filmmaker ModeHDR StandardSDR Filmmaker ModeSDR Standard

Hisense UR8

HDR Filmmaker Mode (10% window)

2097

HDR Filmmaker Mode (Fullscreen)

771

TCL QM8L

HDR Filmmaker Mode (10% window)

3719

HDR Filmmaker Mode (Fullscreen)

759

Samsung QN80H

HDR Filmmaker Mode (10% window)

1149

HDR Filmmaker Mode (Fullscreen)

603

01,0002,0003,0004,000

nits

HDR Filmmaker Mode Data

Product	HDR Filmmaker Mode (10% window) (nits)	HDR Filmmaker Mode (Fullscreen) (nits)
Hisense UR8	2097	771
TCL QM8L	3719	759
Samsung QN80H	1149	603

Hisense UR8

HDR Standard Mode (10% window)

2394

HDR Standard Mode (Fullscreen)

767

TCL QM8L

HDR Standard Mode (10% window)

3771

HDR Standard Mode (Fullscreen)

795

Samsung QN80H

HDR Standard Mode (10% window)

1204

HDR Standard Mode (Fullscreen)

672

01,0002,0003,0004,000

nits

HDR Standard Data

Product	HDR Standard Mode (10% window) (nits)	HDR Standard Mode (Fullscreen) (nits)
Hisense UR8	2394	767
TCL QM8L	3771	795
Samsung QN80H	1204	672

Hisense UR8

SDR Filmmaker Mode (10% window)

314

SDR Filmmaker Mode (Fullscreen)

372

TCL QM8L

SDR Filmmaker Mode (10% window)

229

SDR Filmmaker Mode (Fullscreen)

0

Samsung QN80H

SDR Filmmaker Mode (10% window)

199

SDR Filmmaker Mode (Fullscreen)

231

0100200300400

nits

SDR Filmmaker Mode Data

Product	SDR Filmmaker Mode (10% window) (nits)	SDR Filmmaker Mode (Fullscreen) (nits)
Hisense UR8	314	372
TCL QM8L	229	N/A
Samsung QN80H	199	231

Hisense UR8

SDR Standard Mode (10% window)

1514

SDR Standard Mode (Fullscreen)

778

TCL QM8L

SDR Standard Mode (10% window)

581

SDR Standard Mode (Fullscreen)

0

Samsung QN80H

SDR Standard Mode (10% window)

588

SDR Standard Mode (Fullscreen)

340

05001,0001,5002,000

nits

SDR Standard Data

Product	SDR Standard Mode (10% window) (nits)	SDR Standard Mode (Fullscreen) (nits)
Hisense UR8	1514	778
TCL QM8L	581	N/A
Samsung QN80H	588	340

HDR color gamut

Closer to 100% is better. Taken using a colorimeter and test pattern generator.

Hisense UR8

UHDA-P3

145.2

BT.2020

95.1

TCL QM8L

UHDA-P3

97.9

BT.2020

90.2

Samsung QN80H

UHDA-P3

95.2

BT.2020

75

037.575112.5150

HDR Filmmaker Mode Data

Product	UHDA-P3 ()	BT.2020 ()
Hisense UR8	145.2	95.1
TCL QM8L	97.9	90.2
Samsung QN80H	95.2	75

Picture accuracy

A lower score is better. Taken using a colorimeter and test pattern generator.

SDR Delta-E

Hisense UR8

7.9

TCL QM8L

3.7

Samsung QN80H

1.8

012345678

SDR Delta-E Data

Product	Value
Hisense UR8	7.9
TCL QM8L	3.7
Samsung QN80H	1.8

Input lag

Lower is better. Taken using a Leo Bodnar 4K HDMI Video Signal Lag Tester.

Hisense UR8

4K 60Hz

11.7

1080p 120Hz

11.7

TCL QM8L

4K 60Hz

9.5

1080p 120Hz

0

Samsung QN80H

4K 60Hz

12.3

1080p 120Hz

5.8

03.757.511.2515

ms

HDR Filmmaker Mode Data

Product	4K 60Hz (ms)	1080p 120Hz (ms)
Hisense UR8	11.7	11.7
TCL QM8L	9.5	N/A
Samsung QN80H	12.3	5.8

Hisense UR8 review: Features

• RGB mini-LED panel
• Four HDMI 2.1 ports, up to 180Hz refresh rate
• Anti-glare screen

The Hisense UR8 is the first affordable RGB mini-LED TV to come onto the market, with the Hisense UR9 taking the flagship spot. The UR8 and UR9 are the first RGB mini-LEDs available in smaller sizes, after the tech's introduction in the Hisense 116UX last year. While the UR9 and UR8 are both significantly cheaper than the 116UX, the UR8 is the more affordable option of the two, although it is still pricier than some standard mini-LED.

RGB mini-LED has a similar setup to standard mini-LED, but instead of purely white LEDs within the TV’s backlight, there are red, blue and green LEDs, which are designed to produce bolder colors and higher brightness than a standard mini-LED in the same bracket. Hisense claims that RGB mini-LED can hit 100% of the BT.2020 color space (more on that below).

The UR8 comes equipped with an anti-glare screen, similar to Samsung’s Glare Free screens. It’s a matte screen designed to reduce distracting, mirror-like reflections, to make viewing in brighter conditions easier.

The UR8 is equipped with Hisense’s Hi-View AI Engine RGB processor, which is designed to help boost color performance, as well as offer AI picture and sound features.The UR8 supports both Dolby Vision and HDR10+ HDR formats, as well as the new Dolby Vision 2 format, though no content is currently available for this format at the time of writing. IMAX Enhanced is also supported.

The UR8 comes with a great list of gaming features, including four HDMI 2.1 ports which all support up to 4K 180Hz (up to 4K 120Hz for consoles), VRR including AMD FreeSync Premium Pro, Dolby Vision and HDR10+ gaming, and ALLM. There’s also a game menu, where settings for brightness boosting can be adjusted.

For sound, there’s a Devialet-tuned, 2.1.2 channel speaker system, totaling 70W of power. Dolby Atmos and DTS: X are both supported, and there are a plethora of audio settings and preset modes to dig into.

Which smart TV system the UR8 uses depends on your region. In the US, you get Google TV, while in the UK and Australia the UR8 uses Hisense’s own VIDAA smart TV. As I’m based in the UK, I’ll be discussing VIDAA. It supports an extensive number of apps, including big hitters like Netflix and Prime Video, as well as BBC iPlayer and ITVX in the UK. It also supports Freely in the UK, which can be used to watch live, broadcast TV over Wi-Fi, without the need for satellite or antenna connections.

• Features score: 4.5 / 5

Hisense UR8 review: Picture quality

• Great backlight control
• Effective anti-glare screen
• Colors don’t live up to measurements

In terms of brightness, the UR8 hits some interesting numbers. In Filmmaker Mode, it hit a measured peak HDR brightness of 2,097 nits and in Standard, it hit 2,394 nits. These are solid results, and while flagship OLEDs such as the Samsung S99H are now beating this (it hits 2,743 nits peak HDR), it does beat more standard mini-LEDs in its price bracket such as the Samsung QN80H, which I measured at 1,149 nits peak HDR (though this was on the 55-inch model).

Fullscreen HDR brightness is also solid, clocking in at 771 nits and 767 nits in Filmmaker and Standard modes respectively. Where the UR8 delivers more impressive brightness is in SDR Standard, where it hit a peak of 1,514 nits and fullscreen of 778.

The most surprising measurement, though, was the UR8’s HDR color gamut coverage. Not only did it hit 95.1% of the BT.2020 color space, the highest I’ve measured on a TV to date, but it registered a staggering 145.2% of the UHDA-P3 color space, which massively exceeds the high of 100% I registered on the Samsung S95F last year with the same color space.

Looking at these results, I expected colors to be spectacular — and this brings me to my main criticism of the UR8: the colors I saw did not reflect those measurements. When I watched Elemental, the UR8 still delivered punchy, vibrant colors such as Ember’s flame and the blue of Wade’s water, but they didn’t have the pop I was expecting. In La La Land, at the Spring party, Mia’s yellow dress is bright and natural looking, but I’d seen more engaging colors on other TVs.

I found when I compared the UR8 to the Samsung QN80H, a standard mini-LED in the same price range, the QN80H actually delivered bolder colors. Taking the same La La Land scene for comparison, Mia’s yellow dress, the green bushes, and the red outfits of some party-goers all looked much more vibrant on the QN80H. The UR8’s colors looked natural, but lacked that wow factor I expected from an RGB TV.

It was a shame the UR8’s colors weren’t as dazzling as I’d hoped, because in other areas, the UR8 shines, especially its backlight control. Standard mini-LED TVs can really suffer from blooming (where light from a bright object against a dark background can create a halo effect around said object), but this wasn’t the case with the UR8, which demonstrated little to no blooming through most tests. Watching the UR8 at more extreme angles did reveal some blooming, but it was better than what I’ve seen on some mini-LED TVs in the past.

Watching The Batman, the UR8 demonstrated solid contrast too, with mostly deep black tones and a good balance between light and dark tones, shown as Batman wanders around the crime scene at Mayor Mithcell’s home, or in the bright lights around the Batcave. I did notice blacks could take on a gray hue in some scenes, but for the most part, The Batman looked good.

I did find that the Dolby Vision Dark picture mode was too dim to watch in a bright room, despite the anti-reflection screen, but worked well in ambient and dimmer conditions. The brighter Dolby Vision IQ picture mode works best for brighter rooms. If you do use Dolby Vision IQ, I recommend dropping the brightness from 50 to 49 and turning off precision detail and super resolution for a more natural-looking image with better contrast.

It was a similar story with Alien: Romulus. The UR8 did a solid job displaying the scenes in space, with good contrast between the light from the moon and the black of space as the crew approached the Renaissance station. The black of space was occasionally gray, but still very good for a mini-LED, backlit TV. Again, backlight control proved to be the UR8’s strong suit.

In all HDR movies, the UR8 delivered crisp textures, with refined detail in people, clothing and objects. The UR8 did a good job showing the intricate textures of the various consoles onboard the mining ship in Alien: Romulus. Close-up shots showed sweat, hairs, and facial details like wrinkles with real clarity as well.

One thing I did find odd was that for non-Dolby Vision content, the Filmmaker, Movie, and IMAX Enhanced picture modes all looked identical. There seemed to be no real difference in brightness, color or contrast — something I would expect from a TV.

The UR8 did a good job with sports. Watching some World Cup highlights on BBC iPlayer, motion was smooth in the default Sports picture mode. While most of the motion styles worked, I found myself using either Clear or Standard. Color enhancer is on by default in this mode, and I personally found the colors too oversaturated, so I turned this off, and colors looked more natural. I did notice there was muddying of textures during very quick camera pans, but for the most part, the UR8 worked well for sports.

• Picture quality score: 4 / 5

Hisense UR8 review: Sound quality

• 2.1.2 channel speaker array
• Good bass and accuracy
• Soundstage can be narrow

The UR8 has a built-in 2.1.2 channel speaker array, tuned by Devialet, with a total power of 70W. There are a good number of preset sound modes and settings, but I mainly used Standard and Theater mode for my testing.

Watching the Penguin chase scene from The Batman, the UR8 delivered good bass levels, accurately conveying the loud rumble of the Batmobile’s engine. The UR8 demonstrated good accuracy too, mapping the sound of swerving tires, bullet sprays and the horns of passing traffic to the action on screen with solid precision.

While I typically use Theater mode on most TVs, I found that the UR8’s could get harsh at times, with a sudden brightness that I found distracting. Standard mode was a more than suitable alternative, and although it didn't have as wide a sound field as Theater mode, it was more controlled.

The UR8 delivers solid sound overall, but I did feel its soundstage was narrow, meaning not every element of The Batman scene was given room to breathe. It’s a decent sound system for a TV, but overall I’d still recommend a soundbar to go with it.

• Sound quality score: 4 / 5

Hisense UR8 review: Design

• Sturdy build
• Solid pedestal stand
• Slightly bulkier depth

The UR8 has a functional, safe design. The TV itself feels sturdy when installed on its central pedestal stand, which does feel solid even though it’s made of a plastic material. The TV is slighter deeper than some rivals, such as the Samsung QN80H, but again it feels well built.

As I mentioned above, the UR8 features an anti-glare screen, which does a good job of limiting reflections in brighter conditions. The 65-inch model I tested felt light for its size and it has a trim bezel, although the bottom bezel is thicker than those on the top and sides. Connections are easily accessible on the side of the rear of the TV.

The UR8 comes with Hisense’s own USB-C and solar rechargeable remote, which for me is one of the better controllers on the market. It does feel a bit light, but for me the button layout is logical, and it’s nice to see a rechargeable remote supplied with a cheaper TV.

• Design score: 4 / 5

Hisense UR8 review: Smart TV & menus

• Google TV US, VIDAA UK and Australia
• Good number of picture and audio settings
• Dated appearance

As I’m based in the UK, I’ll be discussing the VIDAA smart TV platform featured in the UR8 (also used in Australian models). If you’re based in the US, the UR8 uses Google TV as its smart TV platform instead.

VIDAA is a serviceable smart TV system. It supports a good number of major apps including Netflix, Disney Plus and Prime Video, as well apps from UK broadcasters, including BBC iPlayer and ITVX. The home screen offers a lot of content recommendations, but there's also a large banner space at the top of the screen for ads, which can be distracting.

The menus are neatly and logically laid out, with most settings easy to find. Response times when navigating these menus is good too. There are plenty of settings for adjusting both the picture and sound, although I didn’t use a lot of them.

VIDAA does, however, look a little dated, especially in the menus. It doesn’t have the sleekness of Samsung’s Tizen or Google TV, or the simplicity and color of Roku. Secondly, as someone who does change settings a lot, a pop-up eco warning appears every time and this can get frustrating; and there seems to be no way to turn it off.

• Smart TV score: 3.5 or 4

Hisense UR8 review: Gaming

• Near-full suite of features
• 11.7ms input lag (with the right settings)
• Up to 4K 180Hz refresh rate supported

The UR8 is actually very well equipped for gaming. It has four HDMI 2.1 ports that support 4K up to 180Hz (4K 120Hz for consoles is supported), VRR including FreeSync Premium Pro, both Dolby Vision and HDR10+ gaming, and ALLM.

To get the lowest input lag time, which I measured at 11.7ms in 4K 60Hz, you do have to mess around with some settings. Having put the UR8 into Game mode, I then had to turn on a refresh rate feature, which prioritized faster frame rates, and this is how I got the above result. Without adjusting those settings I measured the input lag time at 20ms, above the 15ms threshold we look for in testing.

Playing Battlefield V in 4K 120Hz, actual gameplay felt smooth, and the UR8 did a good job handling fast-moving action sequences like a shootout in a desert mission. The overall picture quality was good, with high brightness, bold color and solid contrast. I did notice that, during particularly quick camera pans, some textures could muddy, but this didn’t happen every time.

• Gaming score: 4.5 / 5

Hisense UR8 review: Value

• Great value in the US and Australia
• Overpriced in the UK
• Overall good performance for the price

Value is a tough attribute to grade for the Hisense UR8. As RGB mini-LED is still in its infancy (having only been introduced in 2025, and only come to smaller screens for the first time this year), I did anticipate that RGB TVs would be pricey, even more entry-level models like the UR8.

With the UR8, it all comes down to where you are. When it's on sale, you can pick the 65-inch model up for $1,499 in the US, and prices are only going to drop. This puts the UR8 in direct competition price-wise with standard mini-LED, giving people looking for a TV within that price range more choice. It's also good value in Australia, with the 65-inch launching for AU$2,495, which translates to $1,799 / £1,299. I expect prices in Australia to come down reasonably quickly too.

The same can’t be said for the UK, unfortunately. The 65-inch launched recently at £2,299, which is actually pricier than the LG C6 OLED — and while RGB is meant to be a direct rival to OLED, as I've mentioned, it’s not there yet. I have seen the UR8 drop to £1,999 at some retailers, but this is still OLED-level pricing.

Overall, the UR8 does deliver pretty good value for money. It’s got a good feature set, delivers decent built-in sound, and is good for gaming, and while the picture doesn’t quite deliver what I’d hoped for, it’s still solid compared to most mini-LED. It’s just a shame that the value proposition is region-dependent.

• Value score: 4 / 5

Should you buy the Hisense UR8?

Buy it if...

Don't buy it if...

Also consider

How I tested the Hisense UR8

• Tested over two weeks
• Tested in different lighting conditions in lab
• Measurements taken using Portrait Displays' Calman color calibration software

My first step in testing the Hisense UR8 was to establish the best picture modes for critical testing, and I landed on Filmmaker Mode and Cinema mode (which admittedly looked very similar).

I then used reference scenes to test the UR8's picture quality, focusing on brightness, color, detail, upscaling, contrast and backlight control. I used a couple of these same scenes to test the TV's built-in speakers.

I also used an Xbox Series X to test the TV's gaming features and performance, playing Battlefield V.

Apart from objective testing, I took measurements of the UR8 using specialized equipment. This included a Jeti Spctral 15VA Spectroradiometer and Klein K-10A colorimeter (with both meters profiled together), with a Murideo Six G 8K Metal test-pattern generator. I took measurements using Portrait Displays' Calman color calibration software.

The first measurements I took were the UR8's brightness, in both SDR and HDR. I used white window patterns in sizes ranging from 1-100%, with peak (10%) and fullscreen (100%) windows serving as the primary focus. I took measurements in the UR8's Standard and Filmmaker Mode pictures modes.

After this, I measured the color and grayscale accuracy (in SDR), focusing on the Delta-E values (the margin for error between the test signal sent and what's displayed on screen).

Next, I measured the UR8's HDR color gamut coverage, focusing on the BT.2020 and UHDA-P3 color spaces.

I also measured the UR8's HDR EOTF in 1,000, 4,000 and 10,000 nits. This shows the UR8 accuracy at both light and dark levels. I also tested the UR8's spectral power distribution, which demonstrates color performance.

Finally, I tested the UR8's input lag time using a Leo Bodnar 4K HDMi Input Lag Tester.

Read our in-depth overview of how we test TVs at TechRadar.

• First reviewed: July 2026
• Read TechRadar's reviews guarantee

House of the Dragon season 4 has plenty riding on it. With the high-fantasy show's next chapter set to be its final installment, it'll have to deliver a fittingly fiery conclusion to the Targaryen Civil War, and wrap numerous other subplots and character arcs before its final credits sequence rolls.

All of that, and the final season's 2028 launch, feel a long way off at this stage — but that doesn't mean we can't look forward to its eventual release. So, if you want to know more about its possible cast, expected storylines, and more besides, this guide to the Game of Thrones prequel's next chapter is geared towards you.

Hop on, dragonrider, and let's take flight together.

House of the Dragon season 4 release date: what we know

HBO hasn't revealed next season's official launch date yet. However, as part of a multi-show renewal announcement that confirmed House of the Dragon season 4 was in development, the US TV network's boss Casey Bloys said it would premiere sometime in 2028.

Of course, plans can change, but word is that the cameras will start rolling in early 2027. Should that be the case, I'll be more confident that House of the Dragon season 4 will make its HBO Max debut in 2028.

House of the Dragon season 4 trailer: is there one?

Unsurprisingly, no — and we won't see one for a long time yet. Once a trailer for House of the Dragon season 4 is unveiled, I'll update this section.

House of the Dragon season 4 likely cast

Full spoilers follow for House of the Dragon season 3, and Fire and Blood, aka the book the show is based on.

Here's who I expect to see in House of the Dragon season 4:

• Emma D'Arcy as Rhaenyra Targaryen
• Olivia Cooke as Alicent Hightower
• Matt Damon as Daemon Targaryen
• Steve Toussaint as Corlys Velaryon
• Tom Glynn-Carney as Aegon II Targaryen
• Ewan Mitchell as Aemond Targaryen
• Bethany Antonio as Baela Targaryen
• Phoebe Campbell as Rhaena Targaryen
• Abubakar Salim as Alyn of Hull
• Clinton Liberty as Addam of Hull
• Sonoya Mizuno as Mysaria
• Kieran Bew as Hugh Hammer
• Tom Bennett as Ulf the White
• Benjamin Evan Ainsworth as Daeron Targaryen
• Freddie Fox as Gwayne Hightower
• Jefferson Hall as Tyland Lannister
• Gayle Rankin as Alys Rivers
• Joplin Sibtain as Jon Roxton
• Kyrt Egyiawan as Orwyle
• Archie Barnes as Oscar Tully
• Annie Shapero as Alysanne Blackwood
• Dan Fogler as Torrhen Manderly
• Amanda Collin as Jeyne Arryn

Three big names were taken off the proverbial chess board in last season's finale, meaning they won't return for chapter 4.

James Norton's Ormund Hightower and Tommy Flanagan's Roderick Dustin were both gravely injured as part of their Tumbleton duel. The pair were then burned alive by Ulf and his dragon Silverwing in the final throes of said conflict.

Meanwhile, Phia Saban's Helaena committed suicide by throwing herself from a window in Maegar's Holdfast as a way out of being Rhaenyra's perennial prisoner. This trio, then, won't be back for another round.

There are bound to be a few returning faces, including Jamie Kenna's Ser Alfred Broome, while individuals like Ser Marston Waters and an unhinged, so-called prophet known simply as The Shepherd, who are yet to appear in House of the Dragon, might also show up.

In Fire and Blood, Broome and Waters play integral roles in the Fall of Dragonstone and/or other late-game events. Meanwhile, The Shepherd is responsible for instigating the Storming of the Dragonpit, a hugely significant event that was teased as part of Helaena's dream sequence in season 3 episode 7. I have a wild theory about the identity of The Shepherd, though, which you can read more about in my season 3 ending explained article in this guide's plot section.

That all said, House of the Dragon has taken some creative liberties with its source material, so there's no guarantee that this triumvirate will be part of proceedings. Hopefully, it won't be too long until HBO confirms season 4's full cast roster, and we can say with certainty that they, plus other returnees and newcomers, will take part.

House of the Dragon season 4 story speculation

Full spoilers follow for House of the Dragon season 3. Fire and Blood spoilers are also discussed.

Season 4's story synopsis hasn't been revealed yet, but we've got a good idea of what'll happen next

My House of the Dragon season 3 ending explained piece contains some clues about where things will go from here. Meanwhile, those who've read Fire and Blood will definitely know what's in store for Rhaenyra, Aegon II, and everyone else moving forward.

I'm about to dig into some aspects of the latter, so this is your one and only warning: potentially big spoilers immediately follow for House of the Dragon season 4. Skip to the next section if you don't want any major story beats ruined ahead of time.

Still here? You'll already know — or want to know — that there are many more deadly events yet to be played out in the Dance of the Dragons.

Indeed, the Fall of Dragonstone and Storming of the Dragonpit notwithstanding, we should also see live-action depictions of the Battle Above the Gods Eye, the Second Battle of Tumbleton, and the final showdown between Rhaenyra and Aegon II.

Other moments, such as the Moon of the Three Kings, Battle of the Kingsroad, the False Dawn, and Hour of the Wolf could also be adapted. However, House of the Dragon season 4 already has plenty of destructive and life-threatening ground to cover, so I'd be surprised if the latter pair are shown in any great capacity, if at all.

That said, the series could end on an unexpected note. Asked by Business Insider about how he planned on concluding things, Condal replied: "We've talked about this being kind of a World War II movie: There are plenty of World War II movies that exist where you tell the story within this greater war, but the curtain drops at some point in the story, and it's not necessarily with Japan's surrender or Hitler dying in the bunker — these big kind of endpoints where you could say World War II truly ended. You're telling the story within the story.

"So, in no way to comment at all on how we're planning on ending this, but I think that was the trick," he continued. "Because Fire & Blood is a decades-spanning history, the question is really more 'Where do you drop the curtain on this particular story?. We're telling the story of the Dance of the Dragons... at what point do you drop the curtain on this, [in a way] that feels satisfying and like you've told a complete story, but also that leaves certain things open-ended? [By] telling the audience, intentionally, that history marches on."

Before things come to an end, there are many subplots and character arcs to tie up. Those include — but aren't limited to — Rhaenyra potentially losing more allies in Baela, Addam, and Corlys; Aegon and Aemond teaming up to tackle Daeron and Rhaenyra; Alyn having to deal with taking the life of the High Septon under Rhaenyra's orders; and Hugh reckoning with Cat's death, and how that'll impact his loyalty to Rhaenyra.

Furthermore, there will be Alicent learning of Helaena's death and Aegon II's return, Mysaria's possible role in instigating a rebellion against Rhanyera, and Ulf being a dragon-wielding wildcard whose no longer loyal to Team Black or Team Green — and that's just dealing with some of the show's expansive ensemble.

Of course, the biggest question of all is: will House of the Dragon be able to give closure to everyone in the eight, 60-minute-long episodes that remain before its final curtain call? That remains to be seen. But, with other popular shows recently failing to stick the landing from a finale standpoint, Condal and company will hope House of the Dragon doesn't become the latest one to under-deliver for its fanbase.

Will there be a fifth season of House of the Dragon?

Right now, no. Speaking to Deadline in January, HBO boss Casey Bloys replied "yes, that is the idea" when asked if the show will end with season 4. In a separate chat with Deadline this June, showrunner Ryan Condal also said "I think I have said what I have to say about Westeros" — the implication being that, once House of the Dragon's fourth season has ended, it won't return for another outing.

There's always the prospect of HBO replacing Condal with someone else if it felt that a fifth season was necessary, but it would be bizarre to see another individual helm a single season of it. It would be even weirder if House of the Dragon's story continued after season 4, because the show was only intended to cover the Targaryen Civil War as it's outlined in Fire and Blood.

There's plenty more ground to cover with House Targaryen and Westeros more generally post-The Dance of the Dragons. Those stories are arguably best told in separate Thrones spin-offs, though, rather than being the foundations on which to build a fifth season. In short: don't expect to see House of the Dragon season 5 once the show reaches what's expected to be its natural end with its fourth and final installment.

• Cloudflare CFO Thomas Seifert predicts machine-generated traffic could outnumber humans by a 1,000 times in the next five years
• The answer to this, Seifert believes, is to ensure servers are efficient enough to handle demand
• Cloudflare previously predicted machines would outnumber humans in 2027, but this digital soothsaying was 12 months out

Machine-generated internet traffic is already outnumbering that by humans, and it shows no sign of declining, experts have warned.

Cloudflare Chief Financial Officer Thomas Seifert has predicted man-made activity will be dwarfed by bots and agentic applications, with such machine traffic could be 1,000 times more than human traffic within the next five years.

However, Seifert was at pains to mention he has form for getting these predictions wrong.

The “rounding error” online

However, right or wrong, human traffic is already outnumbered by bots online, which means measures have to be taken for the internet to remain useful to people, or else grind to a halt under the weight of automated traffic.

Cloudflare believes making efficiencies is the solution, but those optimizations may be something other companies need to focus on.

While we may feel that the internet exists for humans to communicate and shop or spend money on a service, the truth is that it is increasingly dominated by machine-generated traffic. Traditional bots have been replaced with agentic AI and other automated tools, and every request has an impact on servers (and the rest of the internet infrastructure).

Cloudflare’s role in providing cloud services that provide security, performance, and general infrastructure is key to overseeing the traffic, with many websites relying on its protective shield to improve load times and mitigate online threats.

Explaining the thinking behind the predicted increase, Seifert said on a recent Cloudflare earnings call: “To give you a sense of how this trend is playing out, and with the big caveat that I have called it wrong at every point along the way, if the current trends continue, we think in five years, non-human traffic will be as much as 1,000 times as much as human traffic.”

“In other words, humans will be a rounding error on the internet, not because human traffic goes down, but that’s just how fast we’re seeing non-human traffic grow.”

Efficiency questions

Just how can servers and infrastructure be more efficient? Will this efficiency focus on human or machine generated and agentic transactions?

Some solutions are intelligent traffic management, optimized protocols, hardware acceleration, and edge computing.

Discussions in Cloudflare’s earnings call pointed at where the greatest efficiencies may be made – in its competitors, where GPU utilization is “super low,” due to them “buying a server and then trying to sell it back, lease it back” and putting the onus on the customer to “get as much out of the server as possible.”