Feed aggregator

• AMD’s RX 9060 XT is rumored for a May 21 launch at Computex
• The graphics card won’t go on sale until June, though, we’re told
• We do know that AMD intends to launch an RX 9060 offering at some point before the end of Q2, so June would be leaving it late

AMD’s RX 9060 XT is the next in line for the RDNA 4 range, rumor has it – at least outside of Asia (I’ll come back to that point later) – and we’ve just heard a possible release date for this GPU which will sit underneath the RX 9070 models.

VideoCardz reports that Chinese tech site Benchlife has the scoop on this one, claiming that the RX 9060 XT will be launched on May 21 (apply seasoning now).

This backs up a previous nugget of speculation from the grapevine contending that AMD is shooting for a mid-May launch. And it makes sense given that Team Red has its Computex keynote on May 21, which would be an obvious time to make a revelation like this.

I should clarify that this is just the reveal of the RX 9060 XT, we’re told, and these graphics cards won’t go on sale until later – in June, in fact, according to the report. (Assuming no translation issues are clouding the air here, which is always possible when the original material is in another language, Chinese in this case).

AMD itself has only confirmed that RX 9060 models are coming in Q2 2025, so June is at the late end of that timeframe.

(Image credit: Shutterstock / Gorodenkoff) Analysis: A sound bet for Computex

I wouldn’t bet against a Computex launch, because AMD has even said it will “announce key products and technology advancements” in the sphere of gaming at the show (among other areas, including AI PCs too). So, it seems likely enough, then, although arguably it’s also an easy target for an educated guess, if that’s what Benchlife is engaging in here.

The date also seems well placed to fire a shot back at Nvidia, in theory. The rumored release date for the RTX 5060 is May 19, so if that turns out to be correct, then AMD is perhaps looking to steal a bit of the thunder from Nvidia’s release here with a launch just a couple of days after. (Even if going by the rumored specs, the 9060 XT is going to be more pitched at the RTX 5060 Ti, rather than the vanilla RTX 5060).

So, again, all this makes some sense, and while I certainly think we should remain cautious, I’d be inclined to give Benchlife the benefit of the doubt here.

Pricing will of course be key, and we don’t yet have rumored pricing for the RX 9060 XT. Another key question that remains to be answered is whether AMD will release the RX 9060 XT in two flavors: 16GB and 8GB. Previous buzz has indicated that this might be the case, but more recent rumors are asserting that the 8GB model could be canceled before it flies the nest (due to the reaction to the RTX 5060 Ti 8GB).

The RX 9060 XT might be the next GPU to emerge globally as a more affordable RDNA 4 graphics card, but before then, gamers in Asia will get another product – the RX 9070 GRE. Like the GRE variants before it, this might be sold in other regions eventually, but for now, it’s a China-only prospect, with AMD having confirmed a May 8 on-sale date for this GPU.

You might also like...

• Where to buy AMD Radeon RX 9070 XT and RX 9070: the best retailers in the US and UK to check for stock
• Microsoft is digging its own grave with Windows 11, and it has to stop
• I've reviewed three generations of 3D V-cache processors, and the AMD Ryzen 9 9950X3D is the best there is

• OpenAI has rolled back the ChatGPT 4o update that caused the chatbot's AI personality to be 'sycophantic'
• CEO Sam Altman claims more fixes are coming
• We could see a personality selection option in ChatGPT in the future

Following Sam Altman claiming ChatGPT 4o's personality had become 'annoying', OpenAI has now made changes to the AI model, rolling back last week's update.

The company posted a blog post on its website highlighting the issue that had caused ChatGPT to become more 'flattering or agreeable – often described as sycophantic.'

Users found the update infuriating, with CEO Sam Altman sharing on X earlier this week that the company was working on a fix for the 'annoying' behaviour.

the last couple of GPT-4o updates have made the personality too sycophant-y and annoying (even though there are some very good parts of it), and we are working on fixes asap, some today and some this week.at some point will share our learnings from this, it's been interesting.April 27, 2025

In the blog post, OpenAI says, "In last week’s GPT‑4o update, we made adjustments aimed at improving the model’s default personality to make it feel more intuitive and effective across a variety of tasks."

"However, in this update, we focused too much on short-term feedback, and did not fully account for how users’ interactions with ChatGPT evolve over time. As a result, GPT‑4o skewed towards responses that were overly supportive but disingenuous."

In an effort to prevent this from happening moving forward, the company has highlighted how it looks to address sycophancy in the future.

For starters, training will be used to 'steer the model away from sycophancy' as well as added guardrails put in place to 'increase honesty and transparency'.

OpenAI also plans to add more ways for users to provide feedback on updates before they are rolled out to the wider public.

More personality updates to come

we started rolling back the latest update to GPT-4o last night it's now 100% rolled back for free users and we'll update again when it's finished for paid users, hopefully later today we're working on additional fixes to model personality and will share more in the coming daysApril 29, 2025

Following the rollback of the ChatGPT 4o update, Altman wrote on X, "we're working on additional fixes to model personality and will share more in the coming days."

Putting two and two together, it looks like users might get a ChatGPT personality selection tool in the future. OpenAI's blog post mentions the company is "building new, easier ways for users to shape ChatGPT's behavior" beyond custom instructions.

If you want to go further and change ChatGPT's personality yourself, we recommend trying the following prompt Reddit user TedHoliday:

"Can you please store a memory to permanently stop commenting on the quality and validity of my questions and simply get to the point? I don't want to be manipulated with flattery by a robot, and I don't want to have to skip past the garbage to get to the answer l'm looking for."

You might also like

• ChatGPT started speaking like a demon mid-conversation, and it's both hilarious and terrifying
• Character.AI's newest feature can bring a picture to uncanny life
• ChatGPT news just got a major upgrade from The Washington Post

• Surfshark has unveiled a privacy-oriented, free public DNS service
• The service promises never to log or track any activity
• Surfshark plans to put its DNS no-log policy under test soon with an independent audit to validate its claims

Surfshark has just become the first VPN provider to unveil a privacy-first public DNS service.

Unlike DNS servers that your Internet Service Provider (ISP) offers you by default, Surfshark DNS promises never to log or track any activity, and it's completely free to use.

Already offering one of the best VPN services on the market, Surfshark's new service seeks to help privacy-conscious individuals and organizations avoid online tracking.

Surfshark public DNS

"By offering a free, privacy-oriented DNS service, we are not only seeking Surfshark’s commitment to make the internet a safer place for everyone but also encouraging individuals, organizations, as well as NGOs to take their first steps towards enhancing online privacy," said Karolis Kaciulis, Leading System Engineer at Surfshark.

Short for Domain Name System, DNS acts as the phone book of the Internet. It translates human-friendly domain names that we all type online into numerical IP addresses that computers can understand. DNS servers—dedicated machines that handle and respond to DNS requests—are then a vital part of how the Internet works.

Yet, DNS servers that ISPs offer by default aren't necessarily private.

ISPs may collect and log users' DNS queries for user identification. They can also monitor DNS traffic and even block specific domains. Users can then be subjected to targeted advertising, too.

Surfshark DNS promises to be different, though. It operates under a strict no-logs policy, which means no collection, storage, or sharing of browsing activity. It also supports secure DNS protocols such as DoT, DoH, and DoQ to keep browsing activity private.

We're introducing our free privacy-oriented DNS service, designed for those who value extra privacy without compromising the speeds of their daily browsing. Check it out. https://t.co/YaU0xvw5nfApril 28, 2025

Many tech enthusiasts are used to opting for alternative public DNS servers to improve their overall internet performance. Yet, none of them, Kaciulis explains, are provided by companies primarily focused on security and privacy.

He told TechRadar: "Recognizing this gap, Surfshark decided to offer its own public DNS servers, aligning with its core mission of providing a more secure and private connection."

A third-party DNS server like Surfshark DNS could improve overall network performance. For example, it is less likely to become overloaded. It also promises to offer a better understanding of geolocation, which can provide users with closer servers.

Such a service can also help users bypass online censorship that ISPs may enforce at the DNS level.

Despite these benefits, Kaciulis doesn't recommend using Surfshark DNS alongside Surfshark's virtual private network (VPN).

"Our VPN inherently includes DNS protection, making a separate DNS configuration redundant for users," he said.

Similarly to its no-log VPN, though, Surfshark now plans to put its DNS no-log policy under test soon with a third-party independent audit.

You might also like

• Surfshark unveils new end-to-end encryption patent to reduce metadata collection
• Surfshark finally adds split tunneling on macOS – making it one of the few VPNs to have it
• Is the future of censorship-resistant VPNs, no VPNs?

Welcome to our 48 hour coverage of World Password Day 2025 - here we will be bringing you all the news, updates, and expert insights into the world of passwords and password security ahead of the big day on May 1st 2025!

Most of us use passwords almost every day to access and protect online accounts, files, and documents. But passwords are only as secure as you make them, and remembering multiple complex passwords can be such a hassle.

Don't worry though, we've rounded up

• advice from our experts to help keep your accounts and passwords secure
• exclusive deals from our password manager partners to keep your data safe
• insights into how passwords are evolving and the future of passwords
• password security content from our extensive archive
• And much more!

Save up to 50% on all three Keeper Plans

Keeper is offering 50% off on its Personal and Family plan and 30% off on its Business plan. The Personal plan has everything a password manager should have, including support for unlimited devices. The Family plan can be shared between 5 users and comes with 10GB of cloud storage. Businesses can benefit from the Business Starter plan and access user activity reporting, plus, as a bonus, each member gets a free family plan.

Preferred partner (What does this mean?)View Deal

Get 60% off RoboForm Premium Plan

RoboForm Premium offers unlimited password storage, one-click login and autofill, password sharing, two-factor authentication (2FA), cloud backup, and emergency access for trusted contacts.

Preferred partner (What does this mean?)View Deal

Save up to 20% on NordPass Business

Use the code PASSDAY20 to get 20% off NordPass Business for World Password Day. With the promo code, expect to pay $3.59 per user per month for a 2-year subscription.

Preferred partner (What does this mean?)View Deal

For days, police say, Samuel Patrick Groft cruised through the streets of Los Angeles on his bicycle, single-handedly chopping down about a dozen city trees with an electric chainsaw in three different neighborhoods.

(Image credit: Jae C. Hong)

From foreign policy and tariffs to immigration changes and targeting of DEI, here's a look back at some of the major moves made in the past 100 days of President Trump's second administration.

(Image credit: Alex Brandon)

At the northern end of the Black Sea, Crimea sits at the crossroads of Europe and Asia, having been at various times in its long history either coveted, conquered or controlled by various powers.

(Image credit: Roger Fenton/Getty Images)

Kids – even some young kids – are being exposed to an unprecedented amount of pornography online and a lot of it is violent and misogynistic.  There are tools parents can use to block this content.

The case could transform public education in the Unites States.

(Image credit: Win McNamee)

Researchers at the Department of Veterans Affairs warn that crucial medical research is in jeopardy unless the Trump administration reverses course on cuts.

(Image credit: Chip Somodevilla)

NPR's Steve Inskeep speaks with Trump ally Steve Bannon about the president's agenda at home and abroad.

(Image credit: Nickolai Hammar/NPR)

President Trump promised a new "golden age" for the United States. But his first 100 days in office have left the economy looking tarnished.

(Image credit: Chip Somodevilla)

Telecommunications is an inherently data-driven industry. Traditionally, great focus has been placed on gaining data visibility into the core and transit areas of the network. This approach has provided valuable insights for throughput optimizations, speed, and general performance.

Regardless of promises of speed, customers are mainly concerned about whether or not they’re able to watch streaming services, play live games, and connect with colleagues and loved ones via video calling - sometimes simultaneously. To solve for optimized performance and a great customer experience, telecommunication companies are shifting their focus toward further enhancing their last-mile visibility.

Historically, service providers have used the last-mile data they collect to identify the worst-performing connections and pinpoint the group of customers most at risk of churning to another provider. However, if approached correctly, last-mile performance intelligence can offer more than just a way to mitigate churn.

Today, when consumers visit the marketing and sign-up pages of telcos or internet service providers, they are typically presented with maps showing broad network coverage and similar pricing between providers, indicative of both the competitiveness of the telecommunications landscape and the utility-like nature of many of its services. On paper, there is often little to differentiate providers from a marketing perspective.

This is where the opportunity lies. Service providers are in a unique position by virtue of owning and operating the networks that connect users with the applications and services they’ve come to rely on in their daily lives. By rethinking their relationship with last-mile data, additional benefits - beyond customer retention - are ripe for the taking when it comes to intelligent fault diagnosis, customer communication, and differentiated services.

How incomplete last-mile data can impact decisions

One of the most important aspects of data-driven decision-making is that the source data underpinning the decision is clean, comprehensive, and has good lineage. This goes to its quality and reliability as the foundation for decision-making.

With more comprehensive last-mile data, service providers and carriers can make better informed decisions regarding new product launches, network planning, capacity, and capital allocation. For example, telcos and service providers will make significant moves around optimization of services for the needs of specific market segments - such as gaming, or industrial companies with IoT needs, and so on.

This kind of optimization has long been a promise of network slicing technology, and many telcos and providers now have this in their arsenal. Yet, the biggest challenge may be determining the best customer cohorts to target with this level of optimized services. This is a problem that is addressable in part by having more comprehensive last-mile data to analyze and base these decisions on.

Incomplete last-mile data also makes it harder to determine where a proposed investment in network infrastructure or capacity would have the biggest performance improvement. By contrast, in circumstances where the telco or service provider is collecting last-mile data from every active service, from every home or premises, they have a much more comprehensive understanding of how their services land, and where improvements most need to be made.

Last-mile intelligence as a differentiator

Telcos and service providers have always had options available to them to collect last-mile data: whether that’s TR-069 on customer premises equipment, or specific broadband monitoring boxes.

The key opportunity ahead lies in the improved capability within service provider environments to correlate and use last-mile intelligence.

For instance, last-mile network intelligence can significantly enhance the ability to deploy maintenance strategies proactively. By continuously monitoring the last-mile connections, service providers can identify potential issues before they escalate into major problems, reducing downtime and improving overall service reliability. This proactive approach not only helps in maintaining customer satisfaction but also reduces operational costs associated with emergency repairs and customer support.

Service providers can also use last-mile insight to analyze usage patterns and preferences to tailor their offerings for a higher level of personalization that can lead to deeper customer satisfaction and loyalty, as customers feel their specific needs are being met.

In a market where pricing and coverage are often similar, last-mile intelligence can be a key differentiator. Service providers that can demonstrate superior last-mile performance through transparent reporting, value-add services, and customer testimonials can gain a competitive edge. This transparency builds trust and can be a deciding factor for customers choosing between providers.

Where will richer data take you?

With insights into all last-mile connections, service providers can access a valuable dataset that has long been overlooked. This knowledge can enhance customer experience and inform a wider range of commercial decisions, resulting in comprehensive strategies that are supported by a clearer understanding of the last-mile situation and the needs of their subscribers.

We rate the best network monitoring tool.

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

With fiber internet available to over half the city, Plano's internet choices are more plentiful and affordable than much of the state.

This review first appeared in issue 342 of PC Pro.

Lenovo’s P series of laptops is designed for speed, with more expensive configurations than this offering discrete workstation-class Nvidia processors to tie in with its ISV certification. The P14s is a slimline version, which means less powerful chips and – in this case at least – integrated graphics. But don’t imagine it’s slow: turn to the graphs and you will see the ThinkPad P14s vying for top position in many tests.

That’s despite the fact it includes a Ryzen 5 processor with six cores compared to the 12 found in the Core i7-1260P, with AMD’s strength due to them all being fast performance cores. Lenovo supports the AMD Ryzen Pro chip with 16GB of high-bandwidth LPDDR5 memory, although this is soldered to the board; if you want more, upgrade to 32GB at the time of ordering for an extra £90. You can replace the 512GB SSD yourself, with a handful of crosshead screws in your way, but doing so will void the generous three-year warranty.

This laptop’s greatest performance came in our battery tests, lasting for over 16 hours in both our video-rundown and light-use benchmarks. Left idling, it kept going for an astonishing 20 hours. It’s reasonably light at 1.4kg, albeit no match for the similarly long-lasting HP Elite Dragonfly G3 (1.1kg) and Dynabook Portégé X40 (1kg).

The P14s’ keyboard lacks the finesse of the ThinkPad T14s, with a harsher feel to the keys and no backlight, but the glass-coated trackpad is just as glorious and they share near-identical IPS panels. Both provide 1,920 x 1,200 pixels with a matte finish and, while it lacks the vividness of an OLED screen, the P14s covered a fine 81% of the DCI-P3 gamut compared to 71% for its sibling. With a contrast of over 1,800:1, clean whites and a peak of 409cd/m2, it’s one of the best screens here.

Lenovo squeezes six connectors onto the left edge: a 3.5mm jack, HDMI, USB-A port, two USB-C ports (not Thunderbolt 4, merely USB 3.2 Gen 2) and an always useful Ethernet port. The right-hand side is dominated by the vent necessary to keep the processor cool, but Lenovo still finds room for a second USB-A port and a smart card slot.

This isn’t an exciting laptop – music sounds harsh, for instance – but it has plenty to offer business users, including a high-quality 1080p webcam, which supports Windows Hello, within the thick top bezel. If you need to square the circle of strong performance and supreme battery life, it’s the best choice here.

We've listed the best laptops for graphic design and the best laptops for drawing and digital art.

Data-as-a-Service (DaaS) has become a hot commodity, with companies around the world enjoying access to multiple rapidly-growing, full-spectrum data marketplaces provided by the likes of IBM and Google, as well as S&P Global, Snowflake, and others.

What is underpinning their rising popularity is the fact that hedge funds, investors, and financial services companies are increasingly discovering how they can leverage these rich datasets to make informed decisions.

Governments around the world have also been quick to adopt DaaS practices, especially when it comes to data monetization drives in such fields as health care and smart city projects.

My aim with this article is to explain what DaaS is, what it is packing “under the hood”, and what data collection alternatives are currently available to businesses and organizations.

What is DaaS?

In a nutshell, Data-as-a-Service is a cloud-based type of software that provides users with multi-source data on demand via APIs, rather than as a standalone product. In other words, by paying a subscription fee based on data usage, businesses gain access to numerous data sets rated by other users, making it easy for them to find what they need.

DaaS facilitates the consolidation of enterprise data in one place and, unlike other data management frameworks, doesn’t require users to have extensive on-premises IT infrastructure or expertise to store, manage, retrieve, and otherwise handle massive amounts of data. As of 2024, there are a total of 60 publicly traded DaaS companies.

As with any other technology, DaaS comes with its own set of pros and cons. On the pro side, it has been used to reduce licensing costs, streamline workloads by leveraging cloud services, speed up software development, create enterprise benchmarking reports, and boost the efficiency of business intelligence.

DaaS’s cons, meanwhile, include risks like data privacy breaches, security violations when dealing with sensitive data, and the low quality of granular, niche data types.

What powers DaaS?

DaaS makes extensive use of alternative data. How do we define that though? Traditional data can be defined as pretty much all publicly available, structured data produced under legal and official supervision. This would include data from statistics departments, press releases, financial statements, and so forth.

Alternative data, on the other hand, is typically unstructured, stored in multiple different formats (from blocks of text to video clips), and extracted for specific purposes. Some businesses — particularly financial services and investment companies — are highly reliant on exactly this type of data already. A key reason for this is that alternative data often contains unique investment signals that aren’t present in its traditional counterpart.

The most common method of collecting alternative data is called web scraping, which dates back to around 1989. In basic terms, web scraping uses specialized software to copy large amounts of unstructured public data from websites automatically and transfer it to a central database or spreadsheet, where it is converted into structured data, for later analysis. This form of data gathering is arguably the main engine behind the power of DaaS marketplaces.

Web scraping has been rapidly rising in popularity over the past decade. As global competition continues to heat up, more and more companies are turning to web scraping to make better-informed business decisions and gain an edge over their market opponents. And with vast amounts of data being generated every single day — around 2.5 quintillion bytes, to be precise — the ways of collecting alternative data and putting it to good use are potentially limitless.

To sum up, DaaS marketplaces often use web scraping to collect alternative data, which may contain unique signals, and provide their customers with powerful cloud infrastructure they can use to analyze it according to their individual needs.

SaaS, DaaS, or… ?

Businesses and organizations have three ways of acquiring the alternative data they need. Let’s take a brief look at each in turn.

Web scraping with SaaS

The first method entails the use of web scraping software (e.g., via a customized scraper API) and an ethical proxy network from a reliable provider. This in-house approach comes with a substantial learning curve and requires a good deal of maintenance — e.g., when a website changes its HTML structure and content, the scraping pipelines can break. There are also potential legal issues, such as accidentally scraping sensitive data.

The main benefit of web scraping with SaaS is that you actually control the data you get. Which means that it’s easier to control its quality, to target specific data, and to maintain security, privacy, and compliance, especially if you’re gathering sensitive information. In addition, you get your data fast, at relatively low cost, and it comes in a structured form.

Buying ready-made datasets

The second method is to buy ready-made datasets that match your needs. While this does mean that you’ll be dependent on a specific vendor (or vendors) for updates and support, as well as somewhat limited customization options, you also get many benefits. For instance, since you’re buying a finished product, you can start using it immediately — no in-house data collection required, and no infrastructure-related overhead.

Ready-made datasets are relatively cheap, as licensing or subscription fees replace substantial upfront investments. Additionally, since you’re getting your data pre-packaged by a reputable vendor, it’s going to be of high quality. This is because companies that sell datasets use various data validation techniques to ensure its accuracy.

Sourcing from a DaaS marketplace

Finally, you may opt for getting your data from a DaaS marketplace, which combines data collection, storage, and management. Since we’ve already covered “vanilla” DaaS, it might be worthwhile mentioning Big Data-as-a-Service (BDaaS) here. In basic terms, what you get with BDaaS is not simply more data, but also a whole data analytics package designed to help companies extract the insights they need.

Given that, according to current projections, the BDaaS market value will reach over $52 billion by 2026, you might want to keep an eye on this in the future. For now, however, if you don’t have much experience in working with data, it’s probably best to sit this one out.

Final word

Ultimately, which method is best depends on your needs. Can you get the data you require in the form of standalone datasets? Does it make sense to collect it yourself using a custom SaaS web scraper? Or perhaps you need access to larger pools of data via something like BDaaS?

Before making that all important decision, make sure you’ve established exactly what type of data you actually require, and what is the easiest, most cost-effective way of obtaining it.

We list the cheapest proxy services.

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

A bill scheduled for House and Senate votes on Wednesday would add an additional 0.75% to the daily room rate tax starting Jan. 1.

(Image credit: Jennifer Sinco Kelleher)

Here are the answers for The New York Times Mini Crossword for April 30.

Trump has made many changes to immigration policy in his first 100 days. Here are five you should know about.

(Image credit: Ariana Cubillos)

Noem was dining in Washington, D.C. when her purse went missing. Authorities believe at least two people worked together on a string of purse robberies in the city.

(Image credit: Alex Brandon)