If you've been tempted by something on our best Android phones list and are looking to upgrade your current model, then you may want to do a trade-in to save some money – and it looks as though Google has plans to make that process easier.
As spotted by Android Authority, edits being made in the main Android codebase suggest technicians who are checking that a phone works will be able to put it into a special 'evaluation mode', and run tests from there.
For safety and security reasons, it's recommended that Android phones are factory reset before they're traded in. That means the person testing them has to go through the normal Android setup process to take a look at them – for every single phone.
This new mode, once it's enabled, would bypass that and mean a phone could be evaluated in a quicker time – and that in turn would mean you'd get a faster decision on the trade-in value of your smartphone when you upgraded.
The Android Debug Bridge Android 15 is ready to roll (Image credit: Shutterstock / Mojahid Mottakin)These changes won't actually show up for users at all – Android will still look exactly the same. What will change are the options available in the Android Debug Bridge (ADB), a comprehensive software tool used by developers and technicians.
Once connected to the ADB, that's when the new ability to skip the setup process will come into play. This isn't something you're going to be able to test yourself, unless you go to the trouble of installing the debugging software on Windows or macOS.
We're expecting the finished version of Android 15 to start making its way to Pixel phones including the Google Pixel 9 at some point this month, with Samsung Galaxy devices expected to be upgraded early next year.
Beyond that, we've already started hearing rumors about what Android 16 might bring with it. However, it's not clear if this trade-in update for Android will be associated with a major release, or get pushed out separately.
You might also likeThere are plans to open the world's first permanent museum for AI art next year. Dataland, which is the brainchild of new media artist Refik Anadol, will be a 20,000 square-foot space that sits alongside galleries such as MOCA in the famed Frank Gehry-designed Grand L.A development, a culture hotspot in Los Angeles, California.
Anadol describes Dataland, which will host digital artworks and installations expressed through its own Large Nature Model (LNM), as a place "where human imagination meets the creative potential of machines".
Refik Anadol Studio's Large Nature Model is the world's first open source AI model based solely on nature data. The model is trained on data secured by permission from partners, including millions of specimens, objects, images and sounds from the Smithsonian, London’s Natural History Museum and the Cornell Lab of Ornithology, in addition to another half a billion images of nature.
The resulting multi-dimensional data visualization installations are mesmerizing and immersive, a "living museum made of pixels and voxels". You can see an example below.
Not only is the Large Nature Model's data ethically collected (as opposed to scraping data without permission, which is the approach of many AI image generators), but it is run on Google servers in Oregon that only use renewable energy, leading Anadol to label his model 'ethical AI'.
In an interview with the Los Angeles Times, Anadol says, “We are blending Gehry’s building with AI’s infrastructure and technology, and this never-seen-before art form”. He also explains that he is still working out what to call this new art form. It's "not AR, not VR, not XR... the best name so far, and people love it, is generative reality.”
So is GR the future of art?
A post shared by DATALAND.ART (@datalandmuseum)
A photo posted by on
Is co-creating with AI the future of art?For at least two years there has a been a tidal wave of AI-generated images hitting the mainstream photography world. An AI-generated image tricked judges to win a photo contest, and I spoke with award winning artists and photo contest organizers who had mixed feelings about AI in the art world. Some feel the rise of AI, especially in the arts, equates to the demise of what makes us human, while others are open minded about AI's role in creativity.
Anadol is more than open minded. He embraces AI, describing its role in art to the Los Angeles Times as a "co-creator and co-being".
His fascinating artwork isn't just the result of a few word prompts into a generator, created using anyone's data without their knowledge. Data is ethically sourced from partners and painstakingly labeled. In one example Anadol says they have 75 million flowers, which took a year to label, all of which are available for other artists and researchers to use too.
The Dataland team also personally collects data from unique locations around the world, such as rainforests, using various technologies including LiDAR, photogrammetry, ambisonic audio, high-resolution images and even scents.
Anadol's practice makes for a compelling antidote to the all-too-often sinister and unethical use of AI in art. It's a multi-sensory art form with the power to educate that some fans are even labeling a 'movement'.
I'm relatively open minded about AI in art, though I prefer to celebrate human creativity, but I haven't embraced AI and its presence in my daily life with the same passion as Anadol. However, his concepts have opened my mind to the positive ways AI can be used for creative purposes, and I'm super keen to check out Dataland, whenever it opens next year.
You might also likeThe official launch of Parallels Desktop 20 has introduced a number of advanced AI capabilities to the virtualization platform.
Offering compatibility across macOS Sequoia and Windows 11 24H2, Parallels Desktop 20 brings enhanced performance for users working on macOS, Windows, and Linux environments.
The latest edition introduces secure, downloadable AI-ready virtual machines (VMs), making AI development easier and more efficient. Notably, Parallels Desktop 20 is the only virtualization software authorized by Microsoft to run Windows on Apple silicon, meaning developers have flexibility and choice across various platforms.
AI-Powered virtual machines for offline useParallels Desktop 20 introduces a new AI Package offering a virtual machine pre-loaded with 14 AI development toolsets, including sample code and instructions, aimed at streamlining AI application development.
With a single click, users can download the virtual machine pre-configured for AI development, including support for running small language models inside the VM. This tool also allows users to customize resources and further lock down the working environment by disabling network access.
Parallels Desktop 20 is fully compatible with macOS Sequoia, Apple's latest operating system, and users can take advantage of macOS Sequoia's new AI-powered writing tools while working with Windows apps within the virtual environment. Another useful feature is the ability to sign in to Apple ID directly within macOS virtual machines running on Apple silicon.
(Image credit: Parallels)On the Windows side, the company claims Parallels Desktop 20 brings up to an 80% boost in performance for some workloads due to the improvements made in the Prism emulator, which is built into Windows 11 on Arm.
Parallels Desktop 20 also introduces a new shared folder technology, which enhances compatibility and performance when running applications on various platforms. The update boosts performance with apps like Mathematica and NinjaTrader. Parallels claim that for some operations on Mac files in Linux VMs, the speeds are now four times faster.
There is also an Enterprise Edition of the new Desktop 20 which brings a new Management Portal that allows enterprises to monitor and manage virtual machines while providing advanced security policies. The Enterprise Edition also has enhanced licensing options, including support for Single Sign-On (SSO) and volume licensing.
This software has achieved SOC 2 Type 2 compliance, a critical security standard. It undergoes regular third-party penetration testing to ensure that it remains secure and resilient against cyber threats. For IT professionals and developers, Parallels Desktop 20 brings upgrades such as Enhanced DevOps services allowing developers to build software in Windows, macOS, and Linux VMs.
The new update also brings enhancements to the Visual Studio Code Extension, which now integrates with Microsoft Copilot to allow users to control virtual machines through natural language commands. In addition, the Snapshots feature and OCR-powered Packer for macOS VMs on Apple silicon have been upgraded.
The Parallels Desktop 20 for Mac is available for purchase or upgrade through the Parallels website and authorized resellers. The software is offered in Standard, Pro, and Business editions, with the Parallels AI Package included at no additional cost in the Pro, Business, and Enterprise editions in 2024.
“As PCs become more AI-capable, we believe AI will soon be standard on every desktop. This shift challenges developers to update their applications to fully leverage AI-enabled PCs,” said Prashant Ketkar, CTO at Parallels.
“That’s why we created the Parallels AI Package: to equip development teams—whether experts or beginners—with accessible AI models and code suggestions. This enables ISVs to build AI-enabled applications in minutes, significantly boosting productivity for every software development team using a Mac."
More from TechRadar ProThis week Amazon's early Prime-exclusive sale gave us some great deals, the Oura Ring 4 was announced, and Sonos said its app is almost fixed.
In case you missed these or any of the other biggest tech news stories of the week we've gathered the eight most important ones here so you can get up to speed.
Once that's done you might want to catch up on the new Mac computers Apple is likely gearing up to announce later this month, and you'll want to check out our picks for the seven new movies and TV shows to stream this weekend.
8. Voting for the TechRadar Choice Awards 2024 opened (Image credit: Future)The TechRadar awards are back for 2024, and this year you can help decide the winner in a record 90 categories, split among the following sections:
You can vote in every category, or just a few, but make sure all of your votes are cast before 00:01am PST / 03:01am EST / 08:01am BST on Friday October 11th 2024, otherwise it won’t count.
Once we’re done counting the votes we’ll combine the scores with the rankings of our expert judges to decide the final winners – with the one exception being the Readers' Choice Award which is 100% decided by you. Winners in every category will be announced during our big TechRadar Choice Awards Week, starting Monday, October 28th.
7. Amazon Big Deals Day early sales began (Image credit: Future)Amazon Big Deals Day – a sort of Prime Day 2 ahead of this year’s Black Friday sale – kicks off officially on October 8, but that hasn’t stopped plenty of discounts going live as part of the early sale across Amazon US and Amazon UK.
There are plenty of discounts across a wide range of products from Amazon’s own tech – even the brand new Amazon Fire HD 8 tablets are up to 50% off – as well as some of our favorite gadgets like LG TVs and Ninja kitchen appliances.
Just like with Prime Day, however, you must be a Prime member to access the exclusive Big Deals Day discounts – though some items are on sale for everyone, so you might be able to save big on the item you want without shelling out for a Prime subscription.
6. Sonos’ app apology tour has an end in sight (Image credit: Sonos)Sonos’ app launch in May was one of the biggest tech blunders of 2024, but this week it announced that 100% of the app’s missing features will be restored “in the coming weeks” – and that 80% have already made their return.
What’s more Sonos said that it’ll be extending the warranty of all Sonos products that are still under warranty by a year by way of an apology to disgruntled customers, and it now has made four pledges with the goal of "addressing the root causes of the problems with the app release" to ensure it never happens again.
If you read about what reportedly went wrong with the Sonos app launch some of the new commitments sound like they’re simply trying to change the company culture back to what it was before its recent major shake-up, but words are words and actions are actions. On that note, we’ll have to see if Sonos sticks to its promise and can start to claw back consumer confidence after what can be generously described as a rough few months.
5. Disney responded to fans after axing highly rated shows and movies (Image credit: Shutterstock / Ivan Marc)We’re back here again, folks. The physical versus digital media debate has reared its ugly head once more after fans noticed that Disney Plus and Max have removed some critically acclaimed and popular movies and TV shows from their back catalogs this week.
After its latest streaming library cull drew particular fan ire, Disney responded by saying the removals accounted for “less than 2% of our total titles”, which has gone down as well as you might expect with many of its subscribers. As more and more films and series are permanently axed from the world’s best streaming services, it might be time for you to restart buying DVDs and/or Blu-rays.
4. We reviewed the Samsung Galaxy S24 FE (Image credit: Future)Samsung launched the Galaxy S24 FE alongside the Galaxy Tab S10 Ultra last week, and we’ve since put the new phone through our full review process. The verdict? It’s a “shining example of how to do a more budget-friendly phone right.” For a cheaper price than the standard Galaxy S24, the Galaxy S24 FE delivers a high-quality display, strong performance levels, and a capable suite of cameras – not to mention the full gamut of Galaxy AI features that have become synonymous with the best Samsung phones.
According to our tester, Samsung’s latest FE is an “AI-powered phone that's light on sacrifices, heavy on value,” so if you’re in the market for a near-flagship upgrade that won’t break the bank, the Galaxy S24 FE is absolutely worth considering.
Several new challengers may have entered the best smart ring, er, ring, but Oura isn’t about to go down without a fight. This week, it introduced the Oura Ring 4. It’s not a major redesign on the outside – we suppose there’s only so much you can do to tweak a simple band shape – but inside Oura is improving its device’s measuring accuracy (no matter its orientation), making the fit more comfortable, and giving the app a refresh too.
This rejigged sensor array – dubbed Smart Sensing – sounds like it could be a major convenience win for users. If you’ve never used a smart ring before, a quick TL;DR is they have a specific orientation that they like to be in so that the sensors are in the right place to take accurate measurements. If the ring rotates out of this preferred place then your readings will be less accurate, and even perfectly fitting rings can spin out of place throughout the day when you use your hands or while you toss and turn in your sleep.
The Oura Ring 4 eliminating this need for the ring to always be worn perfectly is a step in the right direction, and we’re even more excited to hear it’s not getting a price hike. It’s ready to preorder now ahead of its October 15 launch in in sizes 4 through 15 (in either Silver, Brushed Silver, Stealth, Black, Rose Gold, and Gold) starting at $349 / £349 (not available in Australia).
2. Gemini Live now won’t cost you a dime to use (Image credit: Shutterstock)Gemini Live – a chatty version of the Google AI – has rolled out to all Android users for free. It was originally reserved only for paying subscribers, but now the voice assistant can be accessed by anyone who downloads the Gemini app – though it can only speak English for now.
After choosing one of ten voices, you can start conversing with the AI in a more natural way than you might be used to with Google Assistant or Alexa. Gemini will even respond in more human-like ways using casual language, or sounding like it's brainstorming, rather than just responding to direct instructions.
If you’ve been waiting to try out Gemini Live now’s the best time to.
1. Verizon had a serious outage (Image credit: Getty)If you live in the US and use Verizon, here's one story you won’t have missed; from around 10 AM ET on September 30 the Verizon network experienced some serious issues until it was fixed over nine hours later.
Affected customers couldn’t make calls, send text messages, or use data – or receive calls or messages from unaffected customers using AT&T and T-Mobile.
The FCC reportedly began an investigation into what caused the outage, however at the time of writing we don’t know what went wrong and might not find out for some time.
A recent investigation by Acronis Threat Research Unit (TRU) has revealed an intricate attack which utilized an old version of Microsoft Word as a conduit for installing a persistent backdoor on infected systems.
WordDrone focuses on companies in Taiwan, particularly those involved in the drone manufacturing industry. The investigation revealed that the malware had been installed on systems in companies working in Taiwan's growing drone industry, which has seen significant government investment since 2022.
Taiwan's strategic position in both the technological and military sectors likely made these organizations attractive targets for espionage or supply chain attacks.
Microsoft Word vulnerabilitiesThe attackers use a technique known as DLL side-loading to install malware through a compromised version of Microsoft Word 2010. It installs three primary files to the target system which are a legitimate copy of Winword (Microsoft Word), a maliciously crafted wwlib.dll file, and a file with a random name and extension.
The legitimate Winword application is used to side-load the malicious DLL, which serves as a loader for the actual payload hidden within the encrypted random-named file.
DLL side-loading is a technique that exploits how Windows applications load libraries. In this case, the attackers take advantage of an older version of Microsoft Word, which had a vulnerability allowing it to load a malicious DLL file disguised as a legitimate part of the Microsoft Office installation. The malicious wwlib.dll file acts as a loader, decrypting and executing the actual malware payload hidden in another encrypted file. This use of DLL side-loading makes it difficult for traditional security tools to detect the attack.
The attackers go as far as digitally signing some of the malicious DLLs with certificates that had only recently expired. This tactic allows the malware to evade detection by security systems that fully trust signed binaries.
Once the attack is triggered, a series of malicious actions unfold. The attack begins with the execution of a shellcode stub, which decompresses and self-injects a component known as install.dll. This component establishes persistence on the target system and initiates the next phase by executing ClientEndPoint.dll, which serves as the core of the backdoor functionality.
After installation, the malware prioritizes maintaining persistence on the infected system, utilizing the install.dll component to achieve this. This component supports three operational methods: installing the host process as a service, setting it up as a scheduled task, or injecting the next stage without establishing persistence. These options allow the malware to remain active and evade detection, ensuring it can continue its malicious activities even after the system reboots.
The final stage of the attack begins with two important tasks. First, the malware performs NTDLL unhooking, a technique used to remove potential hooks placed by security software. The malware ensures that no hooks can interfere with its malicious operations by loading a fresh instance of the NTDLL library. Second, the malware uses a technique known as EDR silencing to neutralize popular Endpoint Detection and Response (EDR) tools. It scans the process list for known security tools and adds blocking rules to the Windows Firewall for any matches. This effectively disables the ability of security software to detect or prevent further malicious activity.
One of the more sophisticated aspects of the malware is its ability to communicate with a Command-and-Control (C2) server. The configuration for C2 communication is embedded in the malware and it's based on a time-based schedule. A bit array in the configuration represents every hour in a week, and if a specific hour is marked as active, the malware would attempt to establish a connection with the C2 server.
The malware also supports multiple protocols for communication, including TCP, TLS, HTTP, HTTPS, and WebSocket. Once communication is established, the malware could receive additional commands or payloads from the C2 server. The custom binary format used in the communication made it more difficult to detect and analyze the traffic.
The initial access vector for the attack remains unclear, but investigators noted that the first appearance of malicious files was in the folder of a popular Taiwanese ERP software. This raised the possibility of a supply chain attack, where the attackers compromised the ERP software to distribute the malware.
More from TechRadar Pro