Feed aggregator

Never overcook your Thanksgiving turkey with this smart kitchen gadget by your side.

Save oven space, time and energy by adding these air fryer dishes to your turkey day toolkit.

If you haven't already fallen in love with colorful home lighting, you will after this Black Friday deal.

Track your sleep, workouts and more with the Galaxy Watch 6, available for 53% off -- the lowest price of the year ahead of Black Friday.

• After many delays, Windows Insiders can now try out Recall
• Windows Recall was pushed back after privacy concerns, which are addressed
• You'll need a Snapdragon Copilot+ PC to give it a go

Windows Recall feature’s long road to actually being rolled out is finally coming to an end – sort of. After several delays, all centered around some major security and privacy concerns around the feature that was poised to take screenshots of whatever was on your PC's screen every few seconds, Microsoft is ready for folks to give it a shot.

As part of the just-released Windows 11 Insider Preview Build 26120.2415 (KB5046723) for those enrolled in the ‘Dev Channel,’ you’ll be able to give Recall – which is still labeled as (Preview) – a go and try out the long-awaited and delayed, feature. You will need a Copilot+ PC powered by a Qualcomm Snapdragon chip to give it a go. Those with an Intel- or AMD-powered Copilot+ PC chip must wait for a future release that tosses in support.

And, of course, it's not final software, so akin to a beta or any other preview, you'll need to be okay with bugs, slowdowns, and crashes. Suffice it to say, it's advised not to install it on your main computer.

(Image credit: Microsoft)

Recall, as you might suspect, is entirely optional, but the gist is that it can help you find items from the past or things related to a specific search term. Since it will be taking screenshots and recording what’s on your screen, it does, as the name suggests, let you recall information … or at least that's the idea.

If you opt into Recall and Click to Do, the models for these tools will be downloaded, and you’ll go through a setup process explaining the features. Microsoft has made some changes based on the security and privacy concerns around Recall. If it detects confidential information such as “​​credit card details, passwords, and personal identification numbers,” then it won’t take a snapshot.

The blog post on Windows Insider detailing this rollout notes that “Recall won’t save or store those snapshots,” and a box to send feedback is built-in. You can also opt to exclude specific applications and websites from being captured in a Recall ‘snapshot.’ Certainly, all of this is a step in the right direction and closer to a full consumer release. Furthermore, if you opt-in and are using Recall, you can go through the library of ‘Snapshots’ and delete them as you choose.

Microsoft can’t see these snapshots either, with the announcement stating, “Insiders and Recall users, we want you to know your snapshots are truly yours. We do not send your snapshots off your PC to Microsoft or third parties and don’t use them for training purposes. Microsoft can’t access the keys to view your encrypted data, so we can’t restore your snapshots if you remove Windows Hello or restore your snapshots if you need to reset your PC or move to a new PC.” And to use Recall, you’ll need to authenticate with Windows Hello or another biometric.

(Image credit: Microsoft)

Recall may prove to be a feature that is quite helpful, especially for finding things you remember seeing or writing but can’t precisely place. Click to Do seems like a lighter lift for Microsoft that can identify text and visuals within a snapshot – it will eventually be available outside Recall – to suggest actions that can be accomplished automatically with a click. Clever naming.

It can take text from a snapshot, and push that through to an application or search the web. For photos, it’s a bit more impressive with the promised capability of adding a blur or bokeh effect and even erasing unwanted objects from the image.

If you want to give Recall and Click to Do a go, they are now part of the latest Windows 11 Insider Preview Build, which is out now for Dev. You can find full instructions on how to install it on the Windows Insider blog.

Still, it seems that Microsoft is right on track with its updated timeline for Recall, as the company said folks would be able to try it in November 2024, and at a quick calendar glance, it’s November 22, 2024 – which, by the way, is not Black Friday.

You might also like

• A Chrome-killing browser will be OpenAI's next big shot at Google, according to a new report
• Can’t uninstall or update your Microsoft Store apps? Weird Windows 10 bug has just been fixed, thankfully
• Microsoft is again nagging Windows 10 users to buy a new machine – a Copilot+ PC to be precise, for the ‘ultimate Windows 11 experience’

• Amazon just invested another $4B in Anthropic
• Most believe Claude will be Alexa AI's brains
• Why isn't Amazon sharing updates about its popular chatbot?

Amazon Alexa's 10-year anniversary passed without fanfare and, in particular, without the major AI update we were promised in 2023. Now, though, as Amazon invests over $4b (on top of a previous $4b+) in Claude AI maker Anthropic, the question becomes more urgent: What is Amazon's AI strategy, and when will consumers and their best Echos get the long-promised AI update? More pertinently, could that change ultimately be Claude in disguise?

Amazon announced the fresh investment on Friday in a blog post and characterized it as a deepening "strategic collaboration." It is a two-way street in that Anthropic is set to use the vast Amazon Web Services server army as its "primary training partner." That's a big deal because one of the key stumbling blocks for these fast-developing generative AI platforms and large language models (LLM) is the processing power needed to train them.

Nvidia, now the world's most valuable company, has made a huge business out of this. Anthropic, still a startup, needs a tech giant like Amazon to back its AI aspirations.

It is, though, the other side of the deal that interests me and likely interests most consumers, and that's Amazon's growing investment in the Anthropic AI platform, specifically Claude models. Currently, they're underpinning much of the cloud-based AI Amazon provides to partners through its Amazon Bedrock platform. The missing connection, or at least not spelled out, is how this might affect Alexa, which doesn't even warrant a mention in Amazon's blog post.

Still waiting on Alexa AI

Launched a decade ago alongside the Echo, Alexa was not the first digital assistant, but most regarded it as better than the first mover, Apple Siri. Echo's instant popularity and pervasiveness in the home meant that millions of people who had never talked to a digital assistant were soon engaged with something that at least sounded like a basic AI chatbot.

Today's Alexa is generations behind leading chatbots like ChatGPT, Gemini, Claude, and Copilot. This is even though Amazon demonstrated a far more intelligent and conversational Alexa AI in September 2023.

The absence of any Alexa mention in the Anthropic news leads me to believe somewhat counterintuitively recent reports that the next Alexa AI will be powered almost entirely by Claude. This will start, I think, with Amazon finally buying Anthropic. I know, it doesn't have to; Microsoft has gotten away with holding OpenAI firmly at arm's length, pouring in money, and getting first dibs on the best stuff without making it an official Microsoft brand.

Anthropic and Claude do not have the brand cache of OpenAI and ChatGPT. Alexa is the bigger and more interesting brand, so why not make Anthropic part of Amazon and Claude the true brains of one of the best-known and most widely used digital assistants on the planet?

I've asked Amazon to comment on the progress of Alexa AI and how Claude might factor in but have yet to hear back. In the meantime, I asked Claude, "Will you power the next version of Amazon Alexa?"

As is often the case with these generative AI chatbots, the answer arrived swiftly:

I apologize, but I cannot confirm or speculate about future product partnerships between Anthropic and Amazon. While I'm aware of developments in AI technology, I do not have inside information about specific business arrangements. If you're interested in the latest information about Amazon Alexa's technology, I recommend checking Amazon's official announcements or press releases.​​​​​​​​​​​​​​​​

When I asked Alexa the same question, it cited the Reuters report from August:

(Image credit: Future)

At least Alexa knows which way the wind is blowing. Now, we just have to wait for Amazon to confirm and then finally deliver the Alexa AI we all deserve.

You migth also like

• Amazon might ask you to pay for the best Alexa
• Beware, your home is spying on you
• People are using their Alexa speakers as glorified timers

Since the Taliban took power 2021, Afghanistan has not been invited to big climate conferences. And money for projects addressing climate-related issues has been frozen. Are things about to change?

(Image credit: Lynsey Addario/Getty Images)

• NoName057 continue DDoS attacks against Taiwanese targets
• Multiple sectors and critical infrastructure were hit by the attacks
• No significant disruption was noted, and many services were restored

In an apparent escalation of recent cyber warfare, Taiwan’s government and major corporations have been targeted by a second wave of Distributed Denial of Service (DDoS) attacks by the pro-Russian hacker group NoName057.

These attacks, which began in early September 2024 and surged again in early October, have affected a wide range of companies including several high-tech firms and critical infrastructure organizations.

The most recent wave of attacks occurred from October 5 to 6, affecting several major companies such as Formosa Plastics, Wistron, and United Microelectronics, all of whom reported their websites were targeted by DDoS attacks, but were quick to restore services and resume normal operations.

Widespread targets

This second wave of DDoS attacks appears to be a continuation of earlier cyber incidents that occurred in September 2024. NoName057 has aggressively targeted Taiwan, claiming responsibility for multiple cyber-attacks. These attacks have also expanded beyond the corporate sector to include essential public services and government entities.

According to NoName057, targets over the first weekend of October included municipal governments, public offices, judicial units, and airports such as Taoyuan International and Songshan Airports. Furthermore, the group claimed attacks on key government databases, including the National Legal Database, showcasing the hackers’ intent to disrupt critical infrastructure in Taiwan.

Although not all organizations have officially disclosed that they were targeted, there were clear signs of disruption. Companies and organizations such as IC design company Shichi, various judicial units and Shixin-KY were also hit by the attack. However, by the evening of October 7, Shixin-KY announced that its website was functioning normally again.

These attacks are not isolated incidents, as NoName057 has a long history of launching cyber-attacks against governments and enterprises across various countries, including Ukraine, France, Lithuania and Czechia.

Before its October attacks on Taiwan, the group targeted Austria and Israel, emphasizing the widespread nature of their malicious activities. Their attacks are usually aimed at disrupting essential services, including government websites, public utilities, and financial institutions.

In 2022, during the visit of U.S. House Speaker Nancy Pelosi to Taiwan, a similar wave of DDoS attacks hit government departments and private companies alike. Likewise, Taiwan’s hosting providers experienced substantial disruption in 2020 due to cyber-attacks, and in 2017, several brokerage websites faced significant downtime due to similar DDoS activities.

Via ITHome

You might also like

• New method for phishing discovered for Android and IPhone users
• Here's a list of the best firewalls today
• These are the best endpoint protection tools right now

Thinking about getting a Nintendo Switch? Black Friday is the best time to do it.

The 2024 elections were a high-water mark for naming and shaming threat actors from foreign governments. There’s still work to be done, though, on how to attribute disinformation campaigns most effectively.

YouTube has augmented its Dream Screen AI tools for short-form videos. You can now use Dream Screen to generate video backgrounds for Shorts using text prompts. Up to now, Dream Screen made image backgrounds, but now it leverages Google DeepMind’s video-generation model, Veo to expand into videos as well.

To set up a Dream Screen video background, you open the Shorts camera, tap the green screen icon, and then select Dream Screen. You can then write whatever text prompt you want and select an animation style. Hit the Create button, and YouTube will produce a set of video backgrounds for you to choose from. After you pick the background you want, you can film your video, and the AI-generated background will be incorporated into the final video.

You might take a video of yourself in a magical forest, recreate scenes from books and movies, or even design a whole opening sequence for a video series you're making. YouTube has also hinted at future updates that will include six-second standalone video clips made with Dream Screen.

level up your Shorts game with Dream Screen ✨you can now generate video backgrounds instantly with just a few words!available now in the United States, Canada, Australia, and New Zealand. try it out ➡️ https://t.co/8w3DiWHyNw https://t.co/oqCGMSdrysNovember 21, 2024

Music dreams

The video backgrounds can mimic various cinematic styles in high-definition, 1080p. The dynamic video backgrounds expand the options for your video backgrounds. Notably, it's a feature that isn't common to YouTube's rivals, such as TikTok. It's not universally available yet, but you can try it out in the United States, Canada, Australia, and New Zealand. YouTube will likely roll it out elsewhere soon.

The new Dream Screen feature also fits with YouTube's other efforts to embed AI throughout the platform. For instance, AI can help inspire new video ideas with YouTube's Brainstorm with Gemini tool and get around rights issues by using AI tool to remove copyrighted music from your video without taking it down completely.

You might also like

• You can start asking AI for your next YouTube video idea
• YouTube’s powerful new AI features give new creators everything except the personality
• YouTube will now take down AI deepfakes of you if you ask

No better time to spruce up that barren living room than right now with these fantastic deals.

It has been over a thousand days since Russia began it's full-scale invasion of Ukraine. The war has had huge geopolitical ramifications, but it has also has affected the lives of millions of ordinary people. Our correspondent in Kyiv tells us how the one couple in Ukraine has weathered the conflict.

(Image credit: Claire Harbage)

Start building your smart home with this massive 66% discount.

It's hot beverage season, and I can't go back after trying this kitchen gadget.

Loved by a CNET wellness expert for its glowing results, the Omnilux Contour Face red light therapy mask is $60 off in honor of Black Friday.

Wisconsin authorities say they have not yet issued a warrant for Ryan Borgwardt, who is believed to be somewhere in Eastern Europe. But that can change if he does not cooperate and return home soon.

(Image credit: Green Lake County Sheriff's Office via AP)

Get a jump on your 2025 reading goal with a discounted e-reader, like the 2024 Kindle Paperwhite.

If you're a Prime member, signing up for the Prime Visa is a no-brainer.

Editors' Choice: The Deco W7200 will cover your whole house with fast speeds for under $170.