Foundation season 3 is already a third of the way through its run on Apple TV+. And, with a new episode set to drop on the tech giant's streaming platform later this week, it won't be long until it reaches its midpoint.
For now, though, you'll want to know when this season's fourth installment will arrive on one of the world's best streaming services.
Below, I'll outline the dates on which we expect Foundation's next entry to drop, and predict what time it'll be released. There's also a full rundown of this season's episodic release schedule at the end of this article, so you can block out some time in your own busy schedule to watch it.
What is the release date for Foundation season 3 episode 4?Dawn will continue to meet with Gaal in secret in Foundation season 3 (Image credit: Apple TV+)This season's fourth episode, titled 'The Stress of Her Regard', will land on Apple's streaming service on Friday, August 1.
Well, that's in some world territories at least. Apple tends to release new episodes of its TV Originals a day earlier in some regions of North and South America. Long story short: Foundation season 3 episode 4 could launch in some parts of these continents, such as the US west coast, on Thursday, July 31.
As for what time it'll debut, that's hard to determine. Apple doesn't publicly state when new episodes will arrive. However, some of the best Apple TV+ shows' latest entries have shown up at 9pm PT on Thursdays and 12am ET on Fridays. It's possible, then, that 'The Stress of Her Regard' will do likewise.
For a breakdown on when I expect season 3 episode 4 to be released, read on (NB: use 12am ET to work out when it'll launch in your country if it isn't listed below):
Want to know when you can stream new episodes of Foundation season 3 in the weeks ahead? Check out the list below:
Republicans in the Texas House have released a proposed new redistricting map that seeks to fulfill President Trump's desire to add up to five additional GOP congressional seats in the state.
(Image credit: Eric Gay)
This review first appeared in issue 360 of PC Pro.
One of the most aggravating things about buying technology is that you can always go one further. Spend a little more, get a little more. And that’s exactly the case if you increase your budget from £1,000 to £1,250, with the Palicomp AMD Destiny offering a significant speed hike over the CyberPowerPC opposite.
The vast majority of this is due to the presence of an RTX 4070 Super card rather than an RTX 4060. At retail, the 4070 Super costs around £600 compared to £300 for its lesser sibling, and it’s roughly twice as fast, too. 3DMark Time Spy is a great measure of this, with the Destiny scoring 18,932 versus 10,518 for the Infinity X145 Elite.
In practice, I can simplify the differences between the two machines in one sentence: the Destiny is a solid option for 4K gaming where the Infinity is not. That’s shown by a trio of results at 4K and High settings in Cyberpunk 2077, Shadow of the Tomb Raider and Dirt 5, where Palicomp’s PC returned averages of 79fps, 114fps and 119fps compared to 39fps, 57fps and 54fps. It doesn’t take a mathematician to spot that the Destiny’s scores are almost exactly twice as fast.
Palicomp has given the RTX 4070 Super a helping hand by using one of the fastest gaming processors on the planet, namely the AMD Ryzen 7 7800X3D. The chip’s X3D suffix indicates the extra Level 3 cache (made possible by a 3D-stacked chipset) that AMD adds over the standard 7800X, which gives an instant boost in most games.
An RTX 4070 Super graphics card takes center stage (Image credit: Future)The downside of AMD’s X3D range of chips is that they don’t have an unlocked multiplier and can’t boost to as high frequencies as their “X” counterparts, so if gaming isn’t your main use then you’re better off with the 7800X (or even the Ryzen 7 7700X, which is roughly 7% faster in multicore tasks). However, in its own right the 7800X3D is a brilliant day-to-day processor, pushing the Destiny to 2,705 and 14,806 in Geekbench 6, 110 and 1,028 in Cinebench 2024, and 1,765 and 17,017 in Cinebench R23 (single-core results first, multicore second).
An overall score of 8,772 in PCMark is also one of the highest I’ve seen, and here I must give due praise to the supporting cast of 32GB DDR5 RAM and a 1TB Adata S70 Blade SSD. The latter is particularly fast, with an astounding 7,476MB/sec sequential reads transfer rate in CrystalDiskMark and a none-too-shoddy 5,461MB/sec result in the writing test.
(Image credit: Future)If you’re thinking 1TB is too small for your needs, there’s one empty M.2 slot on the AMD TUF Gaming motherboard, while removing the right side of the chassis reveals two mounting spaces for 2.5in SSDs and one 3.5in hard disk caddy. There’s also a rat’s nest of cables, but at least that allows Palicomp to keep the visible side of the Destiny relatively empty. This is mainly for aesthetic reasons, as on-board upgrades are minimal: only one PCI-E x1 slot is accessible.
You can at least add to the memory, with two unimpeded DIMM sockets lying empty despite the size of the air cooler. DeepCool isn’t a name I’m familiar with, and while it does its job well – together with the total of six RGB fans on the front, top and back of this PC – it contributes to a continual hum from the Destiny. It’s by no means annoying (more the kind of hum that you only notice when the PC is switched off) but it’s reflected in the idle power draw of 84W compared to 52W for the CyberPowerPC Infinity.
The Kolink Observatory HF Mesh chassis is basic, but does allow all those RGB lights to shine through and has a handy hinged door to make accessing the innards simple. Palicomp also saves money by choosing a CIT 80 Plus Bronze-rated power supply, rather than a bigger name with better efficiency ratings, but at 700W it won’t hold you back.
The Destiny is designed with 4K gaming in mind (Image credit: Future)One of Palicomp’s perks is that you can pick from a variety of components at time of purchase, from CPU to GPU to PSU, and you can also upgrade the warranty. I would consider doing this, as it’s one area where this PC falls behind CyberPowerPC – only the first year includes parts, and you have to pay for courier costs. £108 buys three years of C&R cover. You may also be tempted to pay for faster delivery than the standard ten working days, with £75 accelerating that to the next business day.
Overall, the Destiny isn’t as refined a package as the CyberPowerPC opposite, but there’s no arguing with its powerful components, which are all geared towards one thing: making games run fast.
For on the go play, we've also reviewed the best laptops for work and gaming
A powerful 8.8-magnitude earthquake in Russia's Far East is sending tsunami waves to Hawaii, Alaska, California and Oregon. More countries, like Peru and Chile, are bracing for impact.
(Image credit: Michelle Bir)
Google has just admitted that the UK government hasn't requested access to end-to-end encrypted user data.
Meanwhile, Apple has been hit by a Technical Capability Notice (TCN) under the 2016 Investigatory Powers Act (IPA), forcing it to shutter its iCloud Advanced Data Protection feature in the UK as a result.
The backlash that followed the UK's request for access to end-to-end encrypted data from Apple echoed throughout the rest of the world. It now turns out that the request may have been more targeted than it first seemed, with Google seemingly immune to it for now.
"We haven't received a technical capabilities notice"(Image credit: Shutterstock / nikkimeel)In February this year, the UK authorities requested that Apple break its end-to-end encryption policies in the Advanced Data Protection (ADP) feature. ADP isn't on by default, but when enabled, it adds an extra layer of security. Not even Apple itself can access the data that's been encrypted in this way; it's completely private.
Unsurprisingly, the order was not met with a warm reception. Mounting scrutiny of the UK's Investigatory Powers Act (IPA) led US senators to investigate whether other companies have also received similar requests.
According to TechCrunch, Google refused to answer questions about any involvement from the UK government when prompted by US Senator Roy Wyden. Companies that are subject to government surveillance orders are unable to disclose them under UK law.
However, Wyden disclosed that at least one technology giant confirmed that it hasn't received such a notice. That turned out to be Meta, which told Wyden's office back in March that it hadn't been served an order to backdoor its encryption services.
Although Google remained silent, it appears to have broken that vow of silence in a statement to TechCrunch. Karl Ryan, Google spokesperson, said: "We haven't received a technical capabilities notice."
That's as good a confirmation as we're going to get in this situation. If Google had received such a notice, it would imply that the UK government was surveying whether a backdoor could be added to its end-to-end encryption or not.
Ryan also told TechCrunch: "We have never built any mechanism or 'backdoor' to circumvent end-to-end encryption in our products. If we say a product is end-to-end encrypted, it is."
The UK government might still back downWhen asked to build a backdoor in its ADP service, Apple instead chose to turn off the feature for users in the UK, leaving them without access to additional data protection.
Although using one of the best VPN services can help boost Brits’ online privacy, it's certainly not going to replace iCloud’s end-to-end encryption protection that users in other countries are able to benefit from.
The order was widely criticized worldwide, with US lawmakers warning against "systemic vulnerabilities." Meanwhile, Apple decided to challenge the request in Court.
More recently, two senior British officials disclosed that the UK government might have to give up on pursuing encryption backdoors due to pushback from the US government. "They don't want us messing with their tech companies," they said.
No matter where this case ultimately leads, the fact that Apple was involved, but Google walked away unscathed, is an interesting development. Google’s hands-off status may offer reassurance that the UK’s encryption demands aren’t as sweeping as they first appeared — at least not yet.
You might also likeDJI is well-known for their excellent lineup of drones, videography gimbals and actions cams etc, but all these devices have one thing in common. They are all powered by batteries.
And that's where the DJI Power 2000 comes in, letting you charge your DJI drone (amongst other devices) while out in the field. It's not the first from the company, but is it the best portable power station for drone operators? I put it to the test.
FEATURES AT A GLANCE2048Wh capacity from LiFePO4 batteries
3000W continuous output
4 x AC sockets, 4 x USB-A and 4 x USB-C
0-80% charge in 45minutes
Multiple methods of charging
Fast 10ms response UPS
For drone operators, 30 minutes of flight can equal hours of charging unless they were to buy several expensive, propriety batteries which only have one use, drone power.
DJI may now have the answer for drone operators who not only want to fast charge their drone batteries but other things as well with added flexibility.
Not wanting to miss out on the action, about a year ago DJI brought out their Power 1000 Station. Now they have released the Power 2000 increasing everything for those who need more which we are reviewing with some accessories.
(Image credit: Future)DJI Power 2000: PackagingThe power station and its accessories were all packaged individually in non-descript, plain cardboard boxes with only simple outlines of the products and what they were printed on the box. No fancy glossy or satin box with listed features, photos, pictures or illustrations.
Removing the Power 2000 from the box, we were given only the quick start guide, certificate of conformity and AC power cable, no other accessories were provided.
(Image credit: Future)DJI Power 2000: The power stationOnce out of the box, the Power 2000 measures 448 × 225 × 324mm or 17.6 x 8.9 x 12.7 inches (L x W x H) and weighing in at 22kg (or 48.5lbs).
The Power 2000 has two large, grippy handles on the top to facilitate portability
All the sockets, plugs, ports etc are at the front of the Power 2000 as is the status screen. The sides and back are bereft of anything much save for some threaded holes which we will come back to later.
At the front of this UK model are the following sockets/plugs:
4 x UK Type G sockets outputting 230V at 50Hz AC pure sine wave
4 x USB-C sockets, 2 x max. 140W and 2 x max 65W
4 x USB-A sockets, each able to output 24W
1 x AC charging socket
2 x SDC multi use sockets
And the status screen
(Image credit: Future)DJI Power 2000: ChargingSwitching on the Power 2000 the first time showed us a state of charge of 51%. We were pleased to see that the AC charging socket was an industry standard C13/14 connector.
But it was here that we stumbled on the first hurdle, the cable has screws on either side to prevent accidental disconnection, but the screws were too small, tool free thumb screws would have been better.
Next to the AC charge socket is a small switch that allows you to set it for fast or slow charge, on slow charge the Power 2000 was charging at 1300W and on fast nearly double that at 2200W.
The AC charging socket at the front wasn’t well received, having reviewed several power stations, they often segregated charge on the side or back and discharging sockets on the front.
Also, to the right of the AC socket is a mysterious multi pin socket, is not documented in any of the manuals or software, our assumption, it’s a terminal used by DJI for servicing.
Charging it using solar panels sent for this review, it was noted that this was a third-party panel made by “IBCPower”. Fortunately, the panel uses MC4 connectors. Using the provided MC4 to XT60 cable means that the panels can be used for other power stations.
Plugging in the panels into the Power 2000 is a nuisance. The panels plug into the provided DJI Solar panel adapter module (MPPT). A small adapter box with three XT60 sockets, this allows you to plug three panels to the Power 2000 via this box which is plugged into the Power 2000’s large SDC sockets to charge it.
With only one 200W panel on a moderately bright day, we were only seeing 180W input with already 89% charged, an estimated 92minutes to full charge. Charging from 0% to 100% using only one panel would likely take over 10 hours.
Mentioned before were the screw holes on the sides and back of the Power 2000. You can see that the MPPT box has screw holes through it and it is here that the MPPT box is screwed on to the side of the Power 2000 to secure it in place so you don’t lose it. Screws and a hex wrench are provided.
(Image credit: Future)DJI Power 2000: In useThe status screen is your typical affair that shows all the useful information that you will need at glance such as state of charge, input and output wattage. Output voltage and AC Hz, USB as well as Wi-Fi or Bluetooth connections.
Putting the Power 2000 through its paces, using it to power a variety of high draw household devices such as a rice cooker, mini-induction cooker, hair dryer, air fryer. The Power 2000 was able to power all of them separately and remained silent throughout.
Using the USB is about as ubiquitous as it is, it was able to charge a variety of phones, a Lenovo ThinkPad, MacBook Air laptops with ease.
The Power 2000 has a combined output power of 3000W continuous power.
A feature that stood out to us was the Uninterruptable Power Supply or UPS feature, whilst other power stations are capable of this, most advertise a 20 to 30ms response time or require setting up. The response time is the time the power station detects grid outage and switches over to battery.
The Power 2000 boasts a fast 10ms response time. UPS mode is automatically activated when the Power 2000 is plugged into mains, AC is on and power drawn.
We tested this by plugging in a typical home office setup, into the Power 2000 via multiplugs. The setup drew around 200W, we let the set up drain the Power 2000 for about two hours taking it from 89% to 61%.
We then switched the mains on to charge the Power 2000 back to 100%, you can see the UPS logo (small, orange battery with arrows under the “INPUT”) automatically activated and the Power 2000 began charging at 1360W in slow mode.
Once at 100%, we flicked the mains switch again whilst keeping an eye on the setup, not a flicker. The set up carried on as if there were no power outage, the Power 2000 drew power from the batteries and the UPS mode switched off.
(Image credit: Future)One prominent feature that stands the Power 2000 different to other power stations is the inclusion of two SDC ports on the far upper right.
These are multi use sockets able to charge or discharge from the Power 2000 using DC power, the connectors are essentially QS10 sockets/connectors which are high current anti spark connections.
We were only able to use this to charge the Power 2000 via the afore mentioned MPPT solar adapter box.
The SDC connections would allow a user to plug in add on batteries to the Power 2000 or charge it faster using separately available cables and adapters capable of up to 1800W of input.
The other is to allow fast charging of DJI drone batteries using one of the appropriate adapter cables. It can charge a DJI Mavic 3 series battery from 10% to 95% in about 32minutes. The Mavic 3, Mavic 4, Air 3, Matrice 30, Matrice 4 and Inspire 3 drones, each cable costs £19 at time of review.
Also provided as part of this review is the DJI Power Series Hand Truck.
Essentially a hand trolley with a sculpted foot for the DJI Power Series batteries so that a power station such as the Power 2000 sits nicely at the base. The trolley supports a further two add on batteries to stack on top of the main one with a simple ratchet strap to batten them down.
Also, we mentioned the screws on the back of the Power 2000, these are for the middle handle retaining arm to screw into so that the Power 2000 has at least two points of anchorage to the trolley.
The trolley has 12cm heavy duty rubberised wheels, it can also be folded up for storage but the handle cannot be shortened or extended. The handle height is 98cm high.
(Image credit: Future)DJI Power 2000: The appThis is one place we think that DJI could vastly improve upon and was one where we had some concerns.
Most brands of power stations such as Jackery or Bluettti have apps that go beyond just simply controlling the power station. They are the place where you access your profile/account for that brand as many have loyalty schemes. It is the app that allows you to access your power stations hidden features, settings to tweak to your liking, even if just once.
The competition, mentioned above, have well translated apps that are downloaded from your app shop of choice, here we used the Google Play Store.
On setting up the Power 2000, we searched the Google Play Store to get the DJI Home app and could not find it. It was then discovered that you have to manually download and install the APK file from the web, not exactly what we like doing as it isn’t as curated as it would be from the Play Store.
On firing it up, the first menu is a DJI Product Improvement Project that gives us the choice to disagree or “加入”? It was in Chinese but we were able to translate this as “Join In”. Not off to a good start.
Detecting the battery and setting up was straight forward but once in the app you can see the Power 2000, see where/what is drawing power, state of charge etc. There are limit options you can tweak.
DJI Power 2000: PricingCurrently the Power 2000 is on retail in the UK DJI store for £1185
DJI Power 2000: Final verdictThe Power 2000 is a solid power station but needs some improvement here and there.
If you have already invested in one of DJIs above mentioned drones and are in the market for a power station, then the Power 2000 is the one for you. The SDC connectors allows you fast charging of your drone batteries will allow you to fly more for longer.
The UPS feature was among the fastest response and easiest to set up and its near silent operation means that it can happily work silently in your home office too.
However, it is let down by some poor decisions such as the AC charging socket at the front and the woeful app, both in getting it and it being a glorified status screen if anything.
For more power solutions, we've reviewed the best UPS (Uninterruptible Power Supply).
A vulnerability in SAP NetWeaver is being exploited to deploy Linux malware capable of running arbitrary system commands and deploying additional payloads, experts have warned.
Security researchers from Palo Alto Networks’ Unit 42 discovered a piece of malware called Auto-Color, a Linux backdoor, dubbed for its ability to rename itself after installation.
The researchers found it was capable of opening reverse shells, executing arbitrary system commands, acting as a proxy, uploading and modifying files, as well as adjusting settings dynamically. It was also discovered that the backdoor remains mostly dormant if its C2 server is unreachable, effectively evading detection by staying inactive until the operator instructions arrive.
Salt TyphoonHowever, the researchers weren’t able to determine the initial infection vector - how the malware made it onto target endpoints remained a mystery - until now.
Responding to an incident in April 2025, cybersecurity experts from Darktrace investigated an Auto-Color infection on a US-based chemicals company. They were able to determine that the initial infection vector was a critical vulnerability in SAP NetWeaver, a technology platform developed that serves as the technical foundation for many SAP applications.
The vulnerability was found in the platform’s Visual Composer Metadata Uploader element, which was not protected with a proper authorization. As a result, unauthenticated agents were allowed to upload potentially malicious executable binaries that could do severe damage. It is tracked as CVE-2025-31324, and was given a severity score of 9.8/10 - critical.
SAP fixed the issue in late April 2025, but at the time, multiple security firms were already seeing attacks in the wild. ReliaQuest, Onapsis, watchTowr, Mandiant, all reported observing threat actors leveraging this flaw, and among them - Chinese state-sponsored groups, as well.
Given the destructive potential of the flaw, and the fact that a patch is available for months now, Linux admins are advised to apply it without hesitation and mitigate potential threats.
Via BleepingComputer
You might also likePresident Trump said he would pose an additional penalty on India for its trade relations with Russia, which Trump is trying to pressure over its war with Ukraine.
(Image credit: Andrew Caballero-Reynolds)
For users of Dropbox Passwords, it's time to take a look at the best password managers and find a new service to use, after the cloud storage company revealed it will soon discontinue the service.
The end of service for Dropbox Passwords will be October 28, 2025, giving users ample time to find a new credential manager to suit their needs.
In the announcement, Dropbox said it was discontinuing the tool, “as part of our efforts to focus on enhancing other features in our core product.”
So long, Dropbox PasswordsDropbox Passwords users will still have ample time to access their saved usernames, passwords, and stored credit card information for export until October 28, but there will be some key changes before then.
On August 28, 2025 the password manager will become view-only on both mobile devices and on through the browser extension. You won’t be able to add any more credentials nor use the autofill feature from this date.
On September 11, the mobile app will be depreciated and will no longer be available for use. Your data will remain available through the browser extension.
The fateful date of October 28 will see all Dropbox Password data securely deleted and the dark web monitoring feature will cease to function.
In its blog post, Dropbox has provided some key information on how to export your data from both the app and browser extension. Mobile app users can follow these instructions to export their data:
To export data from the browser extension, do the following:
Levoit's Sprout is an air purifier geared specifically towards children, and it has some nifty tricks up its sleeve. Not only does it purify air, but it also offers various extra features intended to help create the perfect safe, soothing sleep environment. Levoit calls it a 'three in one peacemaker'.
Today's best air purifiers are excellent at removing impurities from the air, and some even double up as fans, but I've never seen one that's designed to act as a sleep aid. I'm actually a little miffed that this one's really for children, because I'd quite like one in my bedroom.
There's a true HEPA filter, which can capture all kinds of nasties – including dust, pollen, smoke and viruses – from the air. On top of that, it's able to track the levels various pollutant types (PM1.0, PM2.5, PM10, TVOC and CO2) as well as monitoring the temperature and humidity in the room.
Soothing sleepThe Sprout doubles up as a night light, emitting soft, blue-light-free ambient illumination, and offers five different white noise settings to help create a soothing environment to drop off in. The curvy shape with rounded edges was created with nursery décor in mind.
The Sprout purifier connects to the VeSync app, where you can view air quality reports in real time, create personalized schedules, control the appliance remotely, and check how soon the filter needs replacing. It's also compatible with Amazon Alexa if you want to go hands-free.
The Levoit Sprout is available to purchase direct from Levoit or via Amazon US or Amazon UK, and costs $279.99 / £279.99 at list price. The good news for UK shoppers is that there's a deal available to take the price down:
There's £50 off the Levoit Sprout right now, taking the price of this child-friendly air purifier and sleep aid down to under £230. This multitasking appliance purifies the air and also has nightlight and white noise functions, plus the curvy design will fit seamlessly into a nursery or child's room.
The same deal is available direct from Levoit, via a coupon code.View Deal
You might also like...Far-Flung Postcards is a weekly series in which NPR's international team shares moments from their lives and work around the world.
AI agents were once theoretical, but now they are a tangible force reshaping the modern threat landscape. Also known as Computer-Using Agents (CUAs), these advanced AI bots can use applications and browse the internet to complete complex, often time-consuming tasks with minimal or no human oversight. Their rapid evolution is unlocking new efficiencies across a variety of sectors with automation and analysis, enabling more informed decision-making.
But this leap forward comes with a caveat. As they grow more capable, AI agents introduce a new class of cybersecurity threats. Malicious actors can hijack these tools to orchestrate sophisticated cyberattacks, exploiting predictable patterns of human behavior to infiltrate systems and exfiltrate sensitive data.
From theory to realityTo move beyond theory and speculation, our team undertook a series of controlled experiments to assess how agentic AI could be weaponized. We found that these agents can automate a wide range of malicious tasks on behalf of threat actors when instructed correctly.
This includes, but is not limited to, credential stuffing and reconnaissance, which previously required significant human effort. To make matters worse, they can even perform outright cyberattacks by guessing passwords and sending out phishing emails en masse.
This marks a watershed moment in cybersecurity’s fight against AI-powered threats. The automation of attacks significantly lowers the barrier to entry for threat actors, enabling even low-skilled individuals to launch high-impact campaigns. This has the potential to rapidly escalate the scale at which phishing attacks can be carried out.
The growing capabilities of AI agentsThe largest AI players are redefining what agents can do. Platforms like OpenAI’s Operator, alongside various tools developed by Google, Anthropic and Meta, all have their own strengths and limitations, but share one critical feature. The ability to carry out real-world actions based on very simple text prompts.
This functionality is a double-edged sword. In the hands of responsible users, it can drive innovation and productivity. But in the wrong hands, it becomes a powerful weapon, one that can turn a novice attacker into a formidable threat.
The good news is that widespread abuse of these tools is not yet common. However, that window is closing fast. The simplicity and accessibility of agentic AI make it an ideal tool for amplifying social engineering attacks.
Automating reconnaissance at scaleTo illustrate the real-world implications, we investigated whether agentic AI could be utilized to automate the collection of information for targeted attacks. Using OpenAI’s Operator, which features a sandboxed browser and possesses uniquely autonomous behavior, we issued a simple prompt: identify new employees at a specific company.
Within minutes, the agent accessed LinkedIn, analyzed recent company posts and profile updates, and compiled a list of new joiners from the past 90 days. It extracted names, roles, and start dates, all the information needed to craft highly targeted phishing campaigns. And, it did this in the blink of an eye.
Some might be tempted to dismiss this as a simple information-gathering exercise. But this experiment displays that seemingly harmless human behaviors like posting job updates on social media can inadvertently expose organizations to significant cyber risk. What once took hours or days can now be accomplished in minutes, at scale.
Exploiting identity through credential stuffingAnother alarming capability of agentic AI is its potential to facilitate identity-based attacks. Credential stuffing, a method where attackers use previously compromised username and password combinations to gain unauthorized access, is a prime example.
To test this attack vector, we instructed Operator to attempt access to login flows on several popular SaaS platforms, equipping it with a target email address and a publicly available list of breached passwords. Based on this limited information, it was able to get into one of the accounts. This underscores how agentic AI can be used to automate credential abuse, bypass traditional defenses and exploit a weak link in the security chain. Human error.
Injecting heightened urgency into human risk managementOur research confirms that agentic AI is already capable of executing a broad spectrum of malicious activities, from phishing and malware delivery to exposing vulnerabilities. While current capabilities are still in their early stages, the potential is there for automated attacks at scale in the not-so-distant future.
This calls for a fundamental shift in how organizations approach cybersecurity. Historically, the focus has been on protecting systems, not people. However, traditional methods like annual training and awareness campaigns only serve to place the burden on employees. This is an outdated approach, and it papers over the real root causes of human error.
Human-centric cyber risk needs to be proactive. And, it needs to be in real-time. This includes two main steps:
By understanding the human behaviors that create openings for threat actors, businesses can deploy smarter, more effective defenses. This shift from reactive to proactive security is well established for software defense, so there is no good reason human risk should be treated any differently.
Adapt before it’s too lateAgentic AI is not just a technological advancement, it is a vehicle for cyberattacks at scale. As these tools become more powerful and accessible, the cybersecurity community must shift its mindset. The future of cyber defense lies not just in securing systems, but in understanding and protecting the people who use them.
The clock is ticking, and the attackers are already adapting. So should you.
We've featured the best encryption software.
This article was produced as part of TechRadarPro's Expert Insights channel where we feature the best and brightest minds in the technology industry today. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/news/submit-your-story-to-techradar-pro
The world has been captivated by the sheer magic of generative AI. In what felt like an instant, we were handed a technology capable of creating endless streams of enterprise knowledge — from customer support scripts and detailed technical manuals to entire marketing campaigns and even lines of code. The first phase of this revolution was one of chaotic, yet thrilling, experimentation. But the novelty is giving way to a stark reality: Creating an avalanche of information isn’t a strategy. It's a liability.
At the Responsible AI Summit in June 2025, the urgency was palpable. Across every conversation I had, one truth kept resurfacing: The very tools designed to drive innovation and accelerate growth could just as easily expose us to risks. What was once seen as a ‘golden ticket’ of sorts is now being viewed through a sharper lens — one that reveals reputational risks, regulatory landmines, and the potential for operational chaos if left unchecked.
The second act of generative AI: complianceWe’re now entering the second, more critical act of this technological shift. The primary challenge is no longer about generation, but about governance and scale across the entire organization. The siloed, human-centric workflows that have managed our information pipelines for decades are breaking under the strain of this new volume. It's not lost on me that the generative AI boom has made content production quicker and easier than ever. But speed without oversight is a liability.
The strategic imperative has shifted from creating more content to creating the right content, reliably and with purpose, regardless of its function. The companies that thrive will be those that move from frantic experimentation to building a unified, strategic creation engine.
Every piece of content either builds your brand — or breaks it. Whether it’s content that strays from brand guidelines, defies industry standards, or is misaligned with regulatory compliance, these missteps in content expose your organization to legal, financial, and reputational risks. We’re seeing the consequences of poor quality content unfold in real-time.
The infrastructure behind AI intelligenceThis requires building a new kind of digital factory for enterprise intelligence. This isn't just a metaphor; it's an operational necessity built on three pillars.
At the heart of it’s a robust MLOps (Machine Learning Operations) pipeline specifically designed for the nuances of generative models, serving as the factory's core assembly line. With generative AI models demanding continuous tuning, evaluation, and oversight, MLOps is what’s behind running the engine, adapting to evolving enterprise needs, and preventing misuse before it occurs.
Second, a mastery of prompt engineering, which acts as the skilled artisan, ensures the AI’s output is refined and aligned with specific departmental intent. Generative AI will only ever be as good as its inputs and the instructions it’s given. Prompt engineering acts as a quality control in translating business context into meaningful, actionable outputs — each purpose-built for the teams that need it.
Finally, a network of powerful APIs serves as the distribution network, seamlessly integrating this new creation engine into every facet of the business, from technical authoring platforms to conversational AI interfaces. Embedding AI into the platforms that employees are already using, APIs bring new capabilities into everyday, scalable workflows.
Together, these three pillars move you beyond fragmented AI experiments and into operational reality. They create a foundation where AI becomes a native part of how your business runs: Consistently, safely, and at scale.
Defining quality by determinismYet, even with a perfect factory, a fundamental question remains: What’s "good" content when it could be anything from a line in a legal contract to a spoken response from a chatbot? Quality is no longer subjective; it must be defined by accuracy, consistency, and safety. This is where we must install creative guardrails.
The answer lies in a concept that seems at odds with generative AI’s nature: Determinism. While AI’s probabilistic power is the source of its creativity, an enterprise requires certainty. We need a technical document to cite the correct API endpoint every time. We need a support chatbot to follow a specific, compliant troubleshooting protocol. We need our legal boilerplate to be exact.
By embedding deterministic rules into our AI systems, we don’t stifle innovation; we create a safe space for it to flourish. These guardrails are the mechanism that safeguards that whether the output is for marketing, engineering, or customer support, it’s verifiably accurate and trustworthy.
Building the intelligent enterprises of the futureLooking ahead, this controlled, strategic approach will unlock a future of knowledge that’s truly dynamic and multi-modal. We’re on the cusp of creating interactive, voice-navigated repair manuals for technicians in the field, real-time multilingual voice support for global customers, and hyper-personalized onboarding documents for new employees. As data from McKinsey suggests, this level of personalization and efficiency lifts revenues by 5-15% and dramatically improves operational effectiveness, turning a universal business function into a powerful engine for growth.
The era of siloed experimentation with generative AI is over. The competitive advantage now belongs to those who can master it as a unified, enterprise-wide capability. The leaders of the next decade will be the ones building their intelligent information factories today, harnessing the chaos to create value that is not only innovative but also intentional, consistent, and mission-critical. The revolution is here; it’s time to give it direction.
We've featured the best AI website builder.
This article was produced as part of TechRadarPro's Expert Insights channel where we feature the best and brightest minds in the technology industry today. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/news/submit-your-story-to-techradar-pro
Following suspicious activity being detected in the computer networks of the capital St Paul, Minnesota Governor Tim Walz has activated the state's National Guard to help defend against attackers.
The attack led to local officials shutting down the information systems in St Paul as a defensive measure, including network access for a range of internal applications, in order to contain the threat. Mayor Melvin Carter assured residents 911 services were operating as normal, but shared that the city has experienced some ‘back-end’ challenges;
“We recognize that these outages have created inconveniences for residents and city staff alike. While these disruptions are difficult, they are necessary steps to limit exposure, preserve system integrity, and protect sensitive information as our investigation and recovery efforts continue,” Carter continued.
A deliberate effortThe city has been working with Minnesota Information Technology Services as well as an external cybersecurity vendor, but even still the ‘scale and complexity’ of the attack has exceeded ‘both internal and commercial response capabilities’, former Democratic VP candidate Walz explained.
It’s not yet clear if this was a ransomware attack, but a range of networks and services have been intentionally and proactively shut down in order to ‘isolate and secure’ Minnesota systems.
These are ‘necessary steps to limit exposure, preserve system integrity and protect sensitive information’ argues Carter, who reveals that the breach was intentionally caused by a ‘criminal, external threat actor’.
"This was not a system glitch or technical error. This was a deliberate, coordinated digital attack carried out by a sophisticated external actor intentionally and criminally targeting our city’s information infrastructure," Carter explained.
Critical infrastructure is a top target for cybercriminals, who leverage the importance of the services they provide and the sensitive data they hold to request higher ransoms.
The US government recently warned Medusa ransomware has hit hundreds of critical infrastructure targets in similar incidents - although no threat actor has yet taken responsibility for this attack.
You might also like20 years after it originally aired, Deadliest Catch is back once again with more death-defying exploits from a fleet of fearless fishermen. A Discovery Channel original, Deadliest Catch goes out on the network around the world, while US viewers can also tune in via Sling TV and HBO Max. Read on for how to watch Deadliest Catch season 21 online from anywhere with a VPN.
Premiere date: Friday, August 1 at 8pm ET / PT
US broadcast: Discovery Channel via Sling TV
Global streams: HBO Max (US) | Discovery Plus (US, UK & CA)
Use NordVPN to watch any stream
Season 21 sees the intrepid Dutch Harbor voyagers head back to the Bering Sea, looking to bring in a haul of red monster king crabs. Making their way to the abandoned military island of Adak off the unforgiving tip of Alaska, there's a 21st century-style gold rush on to find the finest specimens first – hence the name of the first episode: The Wild West.
Familiar face Captain Sig Hansen and John Hillstrand are back to navigate their ship Time Bandit through the perilous seas and, although the full list of vessels and crew are being kept close to Discovery's chest ahead of the series, Jake Anderson has confirmed separately that he will also be returning.
Below we have all the information you need on where to watch Deadliest Catch season 21 online and stream every episode from wherever you are.
Can I watch Deadliest Catch season 21 for free?Not subscribed to Discovery Plus before? New users of the streaming platform get to try it without paying thanks to its 7-day free trial. The trial period is available in the US and Canada.
How to watch Deadliest Catch season 21 online in the USDeadliest Catch season 21 premieres in the US on Discovery at 8pm ET / PT on Friday, August 1. Episodes will go out in the same slot weekly.
Cord cutters can access Discovery via an OTT service such as our favorite, Sling TV. Sling Blue carries Discovery and starts at just $45.99 a month with 50% off your first month.
Episodes will also be available stream online the day after they air Discovery Plus (from $5.99 per month) and HBO Max (from $9.99 per month).
Have one of these subscriptions but away when Deadliest Catch is on? You can still access your usual streaming services from anywhere by using a VPN.
Get 50% off your first month of Sling TV
Sling TV gives you live TV at an affordable price. The Sling Blue package includes more than 50 channels including ABC, Fox and NBC (in select cities), AMC, Bravo, Food Network, HGTV, Lifetime and USA.
If you’re traveling abroad when Deadliest Catch season 21 episodes air, you’ll be unable to watch the show like you normally would due to annoying regional restrictions. Luckily, there’s an easy solution.
Downloading a VPN will allow you to stream online, no matter where you are. It's a simple bit of software that changes your IP address, meaning that you can access on-demand content or live TV just as if you were at home.
Use a VPN to watch Deadliest Catch season 21 from anywhere.
Editors ChoiceNordVPN – get the world's best VPN
We regularly review all the biggest and best VPN providers and NordVPN is our #1 choice. It unblocked every streaming service in testing and it's very straightforward to use. Speed, security and 24/7 support available if you need – it's got it all.
The best value plan is the two-year deal which sets the price at $3.39 per month, and includes an extra 4 months absolutely FREE. There's also an all-important 30-day no-quibble refund if you decide it's not for you.
- Try NordVPN 100% risk-free for 30 days and get an Amazon Gift Card included right now!VIEW DEAL ON
How to watch Deadliest Catch season 21 online in CanadaNew episodes of Deadliest Catch go out on Discovery in Canada, with the season 21 airing at the same time as south of the border at 8pm ET/PT on Friday, August 1.
You'll be able to stream episodes on Discovery Plus, too, with subscriptions costing CA$5.99 per month for Canadian subscribers.
US viewer in Canada? If you’re away from home, you can access your usual services easily with NordVPN.
How to watch Deadliest Catch season 21 online in the UKDiscovery is the home to Deadliest Catch in the UK and season 21 is slated to get started on Tuesday, August 5 at 9pm BST.
Discovery is available through TV providers such as Sky and Virgin Media. You can also stream its shows on demand through the Discovery Plus streaming service (where you'll also find all 20 other series) for as little as £3.99 per month.
Brit abroad? You can still watch the show as you usually would by using a VPN.
Can I watch Deadliest Catch season 21 online in Australia?Deadliest Catch has previously gone out on Foxtel and the Binge streaming service in Australia. However, season 21 is not yet in their listings.
If you’re visiting Australia from abroad and want to watch on your home service, simply download a VPN to stream Deadliest Catch season 21 just as you would back home.
Deadliest Catch season 21 Need to KnowDeadliest Catch season 21 episode guideDeadliest Catch seasons tend to have around 20 episodes, but season 21 episode names for only the first two have been released at the time of writing:
(US TV premiere dates)
No, Deadliest Catch season 21 and all the other seasons aren't available on Netflix anywhere around the world.
You can catch all the action on Discovery+ in the UK, US and Canada.
VPN services are evaluated and tested by us in view of legal recreational use. For example:a) Access to services from other countries, (subject to the terms and conditions of that service).b) Safeguarding your online security and making your online privacy more robust when abroad.Future plc does not support nor condone the illegal or malicious use of VPN services. We do not endorse nor approve of consuming pirated content that is paid-for.
Nintendo has officially announced the next Nintendo Direct showcase. And in typical fashion, there's not long to wait until we can watch it live. The company usually announces Nintendo Directs just a day or two before they go live, and that's no different here.
Announced via Nintendo's social channels, the next Nintendo Direct is taking place on July 31, 2025 at 6am PT / 9am ET / 2pm BST / 3pm CEST. You'll be able to watch the presentation live as it happens over at Nintendo's regional YouTube channels.
The presentation is due to last around 25 minutes, which is to be expected for a more focused partner showcase. There are still plenty of Nintendo Switch 2 games we're waiting for more details on throughout the rest of this year and next, though these will likely be reserved for a larger first-party Direct.
Join us for a #NintendoDirect Partner Showcase tomorrow, July 31, at 6am PT! Tune in for roughly 25 minutes of information on upcoming #NintendoSwitch2 and #NintendoSwitch games from our publishing partners. Watch here: https://t.co/PvBBmmxGTI pic.twitter.com/8hJBngwXHwJuly 30, 2025
Games we're looking forward to hearing more about include Metroid Prime 4: Beyond, Drag x Drive, Kirby Air Riders, and FromSoftware's The Duskbloods. All except The Duskbloods are slated to release this year, with the multiplayer Bloodborne-like arriving in 2026. Also keep in mind that this is a Partner Direct, meaning Nintendo's first-party titles are less likely to be featured here.
You might also like...