Feed aggregator

Sam Altman, CEO of OpenAI, the makers of ChatGPT, put his name to a new viral trend on the platform recently when he retweeted it on X (formerly Twitter) with “love this” added in response to a tweet that has got ChatGPT users enthralled.

The original Tweet, from writer Tom Morgan, says simply, 'Ask ChatGPT “From all of our interactions what is one thing that you can tell me about myself that I may not know about myself'.

It’s a simple prompt but many people have been finding it very insightful.

love this: https://t.co/Jh66ElOmfAOctober 13, 2024

Posted on the Reddit group called ChatGPT, the post had already racked up 10K upvotes and generated 1.8K comments at the time of writing. From the user responses it seems that this particular ChatGPT prompt has found a way of resonating with people, providing an instant peek into their own psychological makeup.

Crying right now

Before you dive over to ChatGPT to try it out, it’s worth noting that this prompt only works if you have subscribed to the paid-for version of ChatGPT, called ChatGPT Plus ($20, £16, AU$30), since the free version has no memory of you from the last time you used it.

Responses on Reddit ranged from Newmoonlightavenger who said simply “It was the best thing anyone has ever said about me” to Jimmylegs50 who wrote, “Crying. I really needed to hear this right now. Thanks, OP."

User PopeAsthetic wrote, 'Wow I did it, and GPT gave me the most profound advice and reflection of myself that I’ve ever received. Even told me I seem to have a desire for control, while at the same time having a desire to let go of control. I’ve never thought about it like that.'

(Image credit: Getty Images) Roast me like one of your French girls

Of course, as is typical of chatbots, people instantly started to experiment with interesting new ways of getting ChatGPT to judge you. “Whatever you do do not, I repeat *do not* ask it to roast you based on all your interactions (unless you have thick skin)” wrote Cagnazzo82, and a flurry of people instantly did just that. It turns out that ChatGPT doesn’t mess around when you ask it to roast you, and the results can be quite brutal!

One user, Fluer-de-Mai, asked “roast me like one of your French girls”, a reference to the line from the 1997 film Titanic, where Rose asks Jack to “Paint me like one of your French girls”. The response was particularly funny: “Alright, picture this: You, sprawled out like a masterpiece of half-finished projects and abandoned goals, draped in the finest procrastination. Every curve is a testament to good intentions gone astray, with a side of “I’ll start Monday” energy. You’re basically a work of art – unfinished, perpetually delayed, and admired only by your Netflix account.”

In fact, when I tried the same prompt the results were scarily accurate:

“But let’s not get ahead of ourselves, you’re probably the type to say, 'Draw me like one of your French girls,” only to immediately follow it up with, “But make sure my espresso is ready in exactly 1 minute 45 seconds. And don’t forget to set up the camera – I’m planning a tech review after this sketch.”'

You may also like...

• ChatGPT o1-preview can solve riddles faster than me and I kind of hate it for it
• Apple's Intelligence's free ChatGPT upgrade may come sooner than expected with a Visual Intelligence bonus
• ChatGPT’s Google search rival is set to go live this year – could SearchGPT break Google's hold on the web?

Ckunsa, an indigenous language in Chile, was declared dead 70 years ago. But groups in northern Chile are successfuly reviving the language and teaching it to a new generation.

(Image credit: Cristóbal Olivares for NPR)

The Android Automotive operating system is rapidly becoming the carmakers’ infotainment platform of choice, but Austrian motorcycle manufacturer KTM is about to introduce the connected experience to its motorcycle line-up for the first time.

Highlighting two new TFT displays, complete with new switchgear to operate them, KTM claims it's been exploring ways to make the Human Machine Interface (HMI) more immersive, providing more information without distracting the rider.

Despite the fact that some motorcycle manufacturers have already embraced Android Auto and Apple CarPlay screen mirroring, such as Honda, Indian and Harley-Davidson on their respective cruiser models, this is the first time a manufacturer has taken the next step and fully baked in the Android Automotive platform.

This means the likes of Google Maps comes pre-loaded into the system, while an eSIM and GPS technology allows riders to navigate directly form the dashboard, without the requirement of tethering a smartphone.

KTM’s V80 (vertical 8-inch) and H88 (horizontal 8.8-inch) touchscreen displays have been designed for easy operation with gloved hands, while a customizable layout allows the rider to split the screen according to their preference.

The company also states that it has introduced a revised CCU3.0 (Connectivity Control Unit) connectivity unit with 32GB of storage and 3GB RAM, which will let you save offline playlists and synchronize phone contacts, for example.

This new control unit also ensures seamless reconnection with a paired smartphone on every ignition cycle, allowing riders to pick up where they left off or navigate via a smartphone and then send routing to the infotainment system.

There is no definitive word on a release date for the technology, with KTM stating it will initially be rolled out on its ‘premium’ models, which we would expect to include the likes of the Super Adventure and Super Duke GT range.

Analysis: Taking the stress out of touring

(Image credit: KTM)

As with the many automakers using Android Automotive, such as Renault, Volvo and Ford, KTM’s newest operating system is designed retain the company’s branding and general feel, but also allow access to a selection of built-in Google-approved apps.

However, there is no mention of access to the wider Google Play store, nor is there any official note on the sort of apps that riders will be able to download and use natively on both the V80 and H88 systems.

This is compounded by the fact that none of the images accompanying the official press release show any third-party apps in action, so we will have to assume that a number of these are still being signed off for use on a motorcycle.

Regardless, KTM's system opens up more possibilities for riders covering big distances to pair Bluetooth headphone and speaker systems to enjoy audio or communicate with a pillion passenger. Riders will also be able to navigate using their favored route-planning tool, such as Waze or bike-specific services like Rever, Detecht and Calimoto, without the need to pair a phone.

While it's nice to experience the open road and get well and truly lost on two wheels, sometimes it's also good to get to a destination without the need to constantly check a smartphone.

You might also like

• I rode the world’s first fully electric adventure motorcycle – and it’s the future aside from one big drawback
• The 7 hottest new electric motorcycles and scooters to boost your commute
• I rode the new Maeving RM1S – and it's the first truly convincing 125cc electric motorbike I've tried

AI is making cybersecurity threats harder to detect, but security professionals believe they have the right tools to fight back, a new survey from Keeper Security has claimed.

The report says the vast majority of IT and security leaders (84%) believe AI-powered tools have made phishing and smishing (SMS-based phishing) more difficult to detect. In the days before AI, phishing emails were full of inconsistencies, as well as spelling and grammar errors, making them somewhat easy pickings for those with even a hint of vigilance.

However, as AI usage has skyrocketed, phishing emails have become almost indistinguishable from legitimate emails, making detection infinitely more difficult. As a result, roughly half (51%) of security leaders see AI-powered attacks as the most serious threat they are facing, with 35% adding that they are least prepared to tackle these attacks, compared to other threats.

The business strikes back

But businesses are not sitting idly. Four in five (81%) implemented AI usage policies for employees, and confidence is relatively high. More than three-quarters (77%) said they were quite familiar with best practices for AI security. Furthermore, businesses were seen adopting data encryption (51%) into their security strategies, since encryption helps protect sensitive data from unauthorized access - a crucial step in defending agastin AI attacks.

Other notable mentions include employee training and awareness, and advanced threat detection systems, both of which were cited by many (45% and 41% respectively) as important tools in detecting and responding to sophisticated AI-drive threats.

“The emergence of AI-driven cyber attacks presents new challenges, but the fundamental cybersecurity practices – such as data encryption, employee training and advanced threat detection – remain essential,” Keeper said. “Organizations must ensure these foundational measures are consistently updated and adapted to meet emerging threats.”

More from TechRadar Pro

• AI models could be hacked by a whole new type of Skeleton Key attacks, Microsoft warns
• Here's a list of the best firewalls today
• These are the best endpoint protection tools right now

True generative AI video editing has arrived on Premiere Pro. At this year’s Adobe Max, the company has revealed the new genAI video tools are now available in beta, including the first generative video model designed to be safe for commercial use.

As we reported last month, the latest update adds a whole suite of genAI tools. Generative Extend is the headline feature, letting users increase the length of video and audio clips.

But there’s much more on offer as Adobe pushes its Firefly AI deeper into the video editing software.

What's new from Adobe Firefly Video?

With the release of the first set of Firefly-powered video editing workflows, Adobe has confirmed several core focuses.

First, dissatisfied with the quality of previous results, Adobe has R&D’d the latest version to the Nth degree. As well as improved video quality, the model has, the company said, been trained on Adobe Stock and public domain data - and not user data or media found online. Adobe trusts the safeguarded training, alongside the indemnification available to enterprise customers, makes this the first generative video model designed to be commercially safe, and more attractive to professionals looking to use AI without fear of copyright infringement.

That doesn’t mean Adobe’s forgotten the core of the experience. In a virtual press conference attended by TechRadar Pro, Alexandru Costin, Vice President, Generative AI and Sensei at Adobe, explained that users “told us editing is more important than pure generation. If you look at the success of Firefly Image, the most use we get inside Photoshop is with Generative Fill because we’re serving an actual customer workflow. So, with video, we’ve decided to focus more on generative editing.”

So, what does that look like in practice?

Generative Extend is the clearest, and most useful example coming to the beta. This tool lets users extend existing video and audio clips to match the soundtrack or alter the pacing, even without enough coverage.

Image to Video and Text to Video have also arrived in earnest - as one would expect to find in any self-respecting AI video editor. By the looks of things, it works in a similar fashion that that found elsewhere across the Creative Cloud ecosystem - with, like any good movie, a twist. Here, users can effectively become the director with creative control over shot size, angle, motion, and zoom. Using the new models, the company also showcased examples of text graphics, B-roll content, and overlaying AI-generated atmospheric elements like solar flares to existing footage.

The latest updates build on last month's set of beta tools, including a new context-aware properties panel that adds most needed tools into one place to speed up workflows. There’s a new Color Management that, Adobe said, “fundamentally transforms the core color engine.” And general performance sees an improvement, ProRes exports, for example, are now three times faster than before.

We’ll be reviewing the latest version of Premiere Pro soon, and we’re keen to see how well the new video tools complement the editing process. In the meantime, users can try out Adobe’s new tools in beta by clicking here.

More from TechRadar Pro

• Get started with the best video editing software for beginners
• Looking for something different? See our review of the best Adobe Premiere Pro alternatives
• New Adobe Premiere Pro updates will make your video content pop

Our writer tried some of the most popular insect repellents during runs, hikes, and evenings. These are the ones we recommend.

By proving that its Super Heavy booster can return to Earth and land, SpaceX has moved closer to creating a reusable interplanetary transport system.

Thomas Was Alone creator Mike Bithell heads the next Tron game, an isometric action adventure with lots of disc combat.

It's by far the best value in VR. But the question is, should you upgrade to it or go with Meta's even better model?

At less than a tenth of the price of an Apple Vision Pro, Meta's new budget VR headset surprisingly delivers... with some corner-cutting. But should you buy this or the Quest 3?

This is everything announced during Monday's Adobe Max keynote, including new features in Premiere Pro, Photoshop, Illustrator and Firefly AI.

New research has claimed nearly as much as two thirds (62%) of SMBs have been impacted by a ransomware attack that originated through software supply chain vulnerabilities.

The report from OpenText notes this is creating apprehension with businesses, as 90% plan to increase collaboration with software suppliers to try and bolster their security practices.

Weaknesses in supply chains are a significant worry for SMBs, with 91% of respondents concerned about ransomware attacks on their organization’s downstream software supply chain or third-party and connected partners - with nearly half (49%) concerned enough to consider vendor changes.

AI in ransomware

Ransomware remains a serious issue for businesses, with 48% of respondents experiencing an attack, and 73% of those being hit within the last year. This affects SMBs more than large enterprises, but only by a small margin, with 76% of SMBs experiencing a ransomware attack in the past year, versus 70% of large enterprises.

Recovery after an attack is a mixed bag. The positive news is that 97% of respondents successfully restored their company’s data, but 46% paid the ransom to do so. The average ransom is rising too, with 31% of those surveyed paying over $1 million.

Research suggests attacks have as much as doubled thanks to AI, and this is reflected in anxieties from SMBs. Over half of all respondents (55%) reported that their company is now more at risk of ransomware thanks to the increased use of AI by threat actors, and 45% have already seen an increase in phishing attacks thanks to AI.

AI is increasingly being used in both cyberattacks and cyberdefence - with the landscape rapidly evolving to accommodate new technologies. AI tools are equipping cybercriminals with new capabilities and methods, making it more difficult for spam filters and antivirus to detect nefarious actors.

More from TechRadar Pro

• Take a look at some of the best malware removal software
• The double-edged sword of AI: Navigating evolving cybersecurity threats
• Check out our pick for best antivirus software

Make charging and organization that much easier with this buy-one-get-one-free offer.

Proton Pass, one of our picks for the best password manager around right now, has launched a new family tier to simplify credential management and storage for parents and children alike.

The Pass Family plan is being offered for a limited time at $3.99 per month (billed annually), and includes all of the features of the Pass Plus plan, but for six users instead of one.

Families will be able to benefit from integrated two-factor authentication (2FA), secure vault and link sharing, dark web monitoring and advanced account protection, all managed from an admin panel.

Family password protection

The new Proton Pass family tier makes sharing credentials for streaming services and household accounts easier and more secure, while also providing vaults for storing personal information such as emergency contact information that can be shared with other users. Each user account is segmented by a unique email address to mitigate data breaches and identity theft.

Password sharing has quickly gone from a big security no to a fact of life, especially within families. Statistics show that 43% of Americans share at least one password with someone else, emphasizing the need for secure password sharing tools.

"Password sharing among family members is common, but it also exposes them to significant cybersecurity risks," said Son Nguyen Kim, Product Lead for Proton Pass. "The Pass Family plan offers a safer, more organized way for families to manage their online identities, combining Proton's high standards of privacy and security with user-friendly features that everyone can use."

More from TechRadar Pro

• These are the best free password managers
• Identities and biometric authentication for Proton Pass have arrived
• Take a look at our guide to the best firewalls

As more and more tech companies enforce office-centered hybrid working policies, many mandating three office days per week, Spotify has bucked the trend and says that working from home is absolutely fine.

Spotify first introduced its ‘work from anywhere’ policy in February 2021 as pandemic restrictions eased, and while the likes of Amazon, Apple, Google and Microsoft have all mandated workers back into the office, the music streaming platform says it has no reason to re-implement in-person working.

The company indicated that it trusts workers to be productive and get the job done, adding that they “aren’t children.”

Spotify says it’s ok with working from home

Chief Human Resources Officer Katarina Berg added: “You can’t spend a lot of time hiring grown-ups and then treat them like children.” She stated that the company had no reason to believe that it should enforce office working policies, but understands why some companies prefer to return to what they know.

Spotify’s remote work policy, which allows workers to choose there they work, has had no noticeable impact on efficiency. That being said, the Swedish firm has partnered with the Stockholm School of Economics to launch a study into the effects of remote work on collaboration and innovation.

While Spotify retains office spaces worldwide, the company encourages employees to come in during a designated ‘core week’ for in-person collaboration and strategy discussions, where projects can be thrashed out without the latency that’s typical of video conferences.

Berg added that this structure appeals to workers as well as reduces the company’s climate impact.

Despite struggling with the tough economic conditions that followed the pandemic, including having to lay off more than 2,000 workers (1,500 in December 2023 alone), the company’s commitment to hybrid working appears to be paying off.

More from TechRadar Pro

• These are the best job sites and best recruitment platforms
• Check out our roundup of the best music streaming services
• Dell is ordering hundreds of workers back to the office full time

A handful of maps get quality-of-life reworks, and mythic skins get new customization options for the cost of more mythic prisms.

While we wait for Apple Intelligence features to roll out in iOS 18, Google has already got some killer AI search features in the Google app.

Get up to speed on the rapidly evolving world of AI with our roundup of the week's developments.

Although they are evenly matched in many ways, Amazon's Alexa smart speakers perform better than Google's Nest in some areas.

Surf the web in private with these browser VPNs.