OpenAI is officially removing several older language models from its ChatGPT interface as of February 13, 2026, marking a significant shift in how users interact with its AI. The retirement affects several models, including GPT-4o, GPT-4.1, GPT-4.1 mini, and OpenAI o4-mini, alongside previously announced retirements of GPT-5 Instant and GPT-5 Thinking variants. These changes apply to the ChatGPT product itself, while access via API remains unchanged for the time being.

Known for its expressive tone, multilingual capabilities, and multimodal features, GPT-4o was temporarily brought back after user pushback during the GPT-5 rollout last year. But with most users already gravitating toward newer versions like GPT-5.2, OpenAI says it’s time to sunset these older systems and concentrate its development efforts on more current technology.

What’s changing and why it matters

The official retirement means that from today forward, these legacy models simply won’t be selectable in the ChatGPT model menu for everyday users. Conversations you’ve had that relied on a now-retired model will automatically default to a newer model, like GPT-5.2, moving forward. For custom applications, workflows, and saved chats, the behaviour will remain seamless from the user’s perspective, though behind the scenes, the model powering those interactions will shift.

OpenAI says the move isn’t taken lightly. Feedback from users who favoured GPT-4o’s particular style and warmth helped shape features in its newer models, like the personality and customisation options in GPT-5.1 and GPT-5.2. However, overall usage of the older systems had dwindled to the point where only a fraction of daily users still opted for them. That said, there is a section of the user base that genuinely relied on the emotional support of the 4o model, and they seem to be quite devastated.

There’s also a practical side to this transition. From a safety and maintenance standpoint, having fewer legacy models to support means fewer resources spent on patching, hosting, and updating code for systems most people no longer use. Looking at the bigger picture, though, this change underscores how fast the world of generative AI continues to move. Models come, users adopt them, and eventually even popular systems give way to advances that promise better speed, smarter responses, and more nuanced interaction. Even if some longtime fans aren’t quite ready to say goodbye.

Many things about diamonds seem eternal, including the many engineering problems related to making them work as a silicon replacement in semiconductor technology. Yet much like a diamond exposed to a stream of oxygen-rich air and a roughly 750°C heat source, time will eventually erase all of them. As detailed in a recent [Asianometry] video, over the decades the challenges with creating diamond wafers and finding the right way to dope pure diamond have been slowly solved, even if some challenges still remain today.

Diamond is basically the exact opposite as silicon when it comes to suitability as a semiconductor material, with a large bandgap (5.5 eV vs the 1.2 of silicon), and excellent thermal conductivity characteristics. This means that diamond transistors are very reliable, albeit harder to switch, and heat produced during switching is rapidly carried away instead of risking a meltdown as with silicon semiconductors.

Unlike silicon, however, diamond is much harder to turn into wafers as you cannot simply melt graphite and draw perfectly crystallized diamond out of said molten puddle. The journey of getting to the state-of-the art soon-to-be-4″ wafers grown on iridium alongside the current mosaic method is a good indication of the complete pain in the neck that just this challenge already is.

Doping with silicon semiconductors is done using ion implantation, but diamond has to be special and cannot just have phosphorus and boron implanted like its sibling. The main challenge here is that of availability of charge carriers from this doping, with diamond greedily hanging on to these charge carriers unless you run the transistor at very high temperatures.

Since you can only add so much dopant to a material before it stops being that material, a more subtle solution was sought. At this point we know that ion implantation causes damage to the diamond lattice, so delta-doping – which sandwiches heavily doped diamond between non-doped diamond – was developed instead. This got P-type transistors using boron, but only after we pacified dangling carbon electron bonds with hydrogen atoms and later more stable oxygen.

State-of-the art switching with diamond transistors is currently done with MESFETs, which are metal-semiconductor field-effect transistors, and research is ongoing to improve the design. Much like with silicon carbide it can take a while before all the engineering and production scaling issues have been worked out. It’s quite possible that we’ll see diamond integrated into silicon semiconductors as heatsinks long before that.

Assuming we can make diamond work for semiconductor transistors, it should allow us to pack more and smaller transistors together than even before, opening up many options that are not possible with silicon, especially in more hostile environments like space.

• Samsung begins commercial HBM4 shipments as AI memory competition heats up
• HBM4 reaches 11.7Gbps speeds while pushing higher bandwidth and efficiency gains for data centers
• Samsung scales production plans with roadmap extending to HBM4E and custom memory variants

Samsung says it has not only begun mass production of HBM4 memory, but also shipped the first commercial units to customers, claiming an industry first for the new high bandwidth memory generation.

HBM4 is built on Samsung’s sixth generation 10nm-class DRAM process and uses a 4nm logic base die, which reportedly helped the South Korean memory giant reach stable yields without redesigns as production ramped up.

Gillian Keating and Caroline O’Driscoll explore the opportunities for women and girls in STEM – as well as the many challenges.

Caroline O’Driscoll and Gillian Keating are the co-founders of I Wish, a national initiative promoting gender equity in STEM that aims to inspire, inform and empower teenage girls. A major event on the organisation’s annual calendar is the I Wish Festival, which this month welcomed nearly 4,000 girls to Dublin’s RDS.

There, they had the opportunity to engage with industry leaders and role models, hear about real career journeys, and see STEM careers as visible, achievable, and relevant. According to O’Driscoll, the festival directly impacts gaps in STEM education for young women by connecting them with positive role models and providing guidance. 

“Over the past 12 years, I Wish has built a sustainable, inclusive and gender-equitable STEM ecosystem in Ireland, supporting girls from primary school through higher education,” O’Driscoll tells SiliconRepublic.com. 

“Many alumni return as volunteers, mentors and speakers, highlighting the lasting impact of the programme in creating confident, skilled young women who believe they belong in STEM.”

The marginalisation that impacts girls and women often starts early and accumulates over time, notes Keating, who explains that a lack of visible role models in leadership positions is key, especially as confidence gaps shaped by social conditioning and structural barriers prevent access to STEM subjects. 

“This marginalisation is rarely about talent – it’s about access, encouragement and belonging,” she says. 

This is why, for both O’Driscoll and Keating, progress in STEM equality doesn’t come from a single event or initiative but from consistent commitment, collaboration and belief in potential. 

Keating says: “This is on all of us. The I Wish Festival matters because it doesn’t just ask girls to adapt to STEM, it asks STEM to adapt to girls. That shift in thinking is where real, lasting change begins.”

Ireland’s future

Recent research from the Economic and Social Research Institute found that Ireland currently has Europe’s largest gender gap in advanced digital skills usage in the workplace. It found that 44pc of men in Ireland use advanced digital skills in their jobs, compared to just 18pc of women, a 26pc difference that is close to double the European average.

Prof Joyce O’Connor, the co-founder and chair of Block W and a research contributor, made note of the fact that in an economy like Ireland’s, which is close to full employment, “failing to fully utilise the advanced digital capability of women already in the labour market is inefficient and unsustainable”. 

This point is echoed by O’Driscoll, who says: “Ireland’s future economy depends heavily on STEM skills, innovation and problem-solving. Excluding or under-supporting half the population is not just inequitable, it’s unsustainable.

“By enabling girls and women to fully contribute their skills and expertise, Ireland benefits from a larger, more diverse talent pool, stronger innovation and creativity, better decision-making and problem-solving, and a workforce that reflects the society it serves. Gender inclusion in STEM is both a social imperative and an economic necessity.”

First impressions

While the present is always a good place to start, when attempting to undo bias or break down barriers, often the work begins from birth and never stops. O’Driscoll explains how educational and workplace institutions often bridge the gap for girls and women in STEM. 

She says: “At primary level, gender inclusive STEM education is key. At secondary level, they need to expose the girls to initiatives like I Wish to connect learning to real world careers.

“At third level, they need to provide mentorship and targeted support. At industry level, they need to support flexible pathways that account for different life stages and responsibilities.”

But to truly advance the system, for all women, Keating emphasises the importance of offering continuous support and – when you have made it – showing others how it can be done. 

Speakers at the I Wish Festival included Olympic paracyclist Richael Timothy, International Rose of Tralee Katelyn Cummins and Olympians Louise Shanahan and Gráinne Walsh, all of whom shared their authentic career journeys, highlighting both the challenges and the successes they have experienced.  

“Our goal is simple but powerful,” says Keating. “If even one girl in our audience leaves I Wish feeling that she truly belongs in this space and is inspired to pursue her ambitions, then we consider it a real win. Through these stories, we hope to spark belief, courage and a sense of possibility in every participant.”

O’Driscoll confirms the importance of a strong role model, noting that when young girls are looking to their futures and the careers they may want, seeing women succeed in STEM can help them visualise themselves in similar roles, breaking down stereotypes about who belongs in these fields. 

She adds: “Our I Wish surveys repeatedly show that girls want more information about STEM and say that access to female role models positively influences their choices. One of the central aims of I Wish is to connect girls with inspiring women in STEM, not just to inform them, but to support them. 

“By inviting women to share their stories and engage directly with students, I Wish creates a network of support that helps girls feel they belong and that their pathways are real and possible.”

Don’t miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic’s digest of need-to-know sci-tech news.

A new variation of the fake recruiter campaign from North Korean threat actors is targeting JavaScript and Python developers with cryptocurrency-related tasks.

The activity has been ongoing since at least May 2025 and is characterized by modularity, which allows the threat actor to quickly resume it in case of partial compromise.

The bad actor relies on packages published on the npm and PyPi registries that act as downloaders for a remote access trojan (RAT). In total, researchers found 192 malicious packages related to this campaign, which they dubbed ‘Graphalgo’.

Researchers at software supply-chain security company ReversingLabs say that the threat actor creates fake companies in the blockchain and crypto-trading sectors and publishes job offerings on various platforms, like LinkedIn, Facebook, and Reddit.

Developers applying for the job are required to show their skills by running, debugging, and improving a given project. However, the attacker’s purpose is to make the applicant run the code.

This action would cause a malicious dependency from a legitimate repository to be installed and executed.

“It is easy to create such job task repositories. Threat actors simply need to take a legitimate bare-bone project and fix it up with a malicious dependency and it is ready to be served to targets,” the researchers say.

To hide the malicious nature of the dependencies, the hackers host the dependencies on legitimate platforms, like npm and PyPi.

In one case highlighted in the ReversingLabs report, a package named ‘bigmathutils,’ with 10,000 downloads, was benign until it reached version 1.1.0, which introduced malicious payloads. Shortly after, the threat actor removed the package, marking it as deprecated, likely to conceal the activity.

The Graphalgo name of the campaign is derived from packages that have “graph” in their name. They typically impersonate legitimate, popular libraries like graphlib, the researchers say.

However, from December 2025 onward, the North Korean actor shifted to packages with “big” in their name. However, ReversingLabs has not discovered the recruiting part, or the campaign frontend, related to them.

According to the researchers, the actor uses Github Organizations, which are shared accounts for collaboration across multiple projects. They say that the GitHub repositories are clean, and malicious code is introduced indirectly via dependencies hosted on npm and PyPI, which are the Graphalgo packages.

Victims running the project as instructed in the interview infect their systems with these packages, which install a RAT payload on their machines.

It is worth noting that ReversingLabs researchers identified several developers that fell for the trick and contacted them for more details about the recruiting process.

The RAT can list the running processes on the host, execute arbitrary commands per instructions from the command-and-control (C2) server, and exfiltrate files or drop additional payloads.

The RAT checks whether the MetaMask cryptocurrency extension is installed on the victim’s browser, a clear indication of its money-stealing goals.

Its C2 communication is token-protected to lock out unauthorized observers, a common tactic for North Korean hackers.

ReversingLabs has found multiple variants written in JavaScript, Python, and VBS, showing an intention to cover all possible targets.

The researchers’ attribute the Graphalgo fake recruiter campaign to the Lazarus group with medium-to-high confidence. The conclusion is based on the approach, the use of coding tests as an infection vector, and the cryptocurrency-focused targeting, all of which aligning with previous activity associated with the North Korean threat actor.

Also, the researchers note the delayed activation of malicious code in the packages, consistent with Lazarus’ patience displayed in other attacks. Finally, the Git commits show the GMT +9 time zone, matching North Korea time.

The complete indicators of compromise (IoCs) are available in the original report. Developers who installed the malicious packages at any point should rotate all tokens and account passwords and reinstall their OS.

Modern IT infrastructure moves faster than manual workflows can handle.

In this new Tines guide, learn how your team can reduce hidden manual delays, improve reliability through automated response, and build and scale intelligent workflows on top of tools you already use.

Get the guide

Sony’s recent State of Play event revealed that the Unbreakable X-Men are joining Marvel Tōkon: Fighting Souls. Arc System Works, the company behind the recent Guilty Gear successes, has stated that their next major item will be released on PS5 and PC on August 6, 2026. Pre-orders begin on February 19th on the PlayStation Store and certain retailers.

The new X-Men squad taking center stage is a line-up put together by Charles Xavier, who reminds the team to go retrieve some of the rest of the original crew, notably Magik, Wolverine, and Danger. They want to save the planet from a catastrophic threat, but the majority of humanity is still frightened of these mutant-types. Producer Takeshi Yamanaka suggests that things won’t be easy: “We’re really hoping you’ll be fascinated about the harsh choices the crew will have to make, and how their story pans out.”

Wolverine arrives with huge claws to bash his way through the competition, while Magik is granted a Soulsword that allows her to tap into otherworldly powers such as magic gateways and summoned demons. Danger, the AI formed during the X-Men’s training sessions, has gotten herself inside a quite advanced-looking mech and is ready to work. Then there’s Storm, who has a combination of wind and lightning at her disposal, which should give her a slight advantage, but as team captain, the true beauty of this squad is that each member contributes something unique to the table.

When the game is released on August 6th, you can expect 20 fighters right away, including teams like the X-Men unit we just mentioned, the Avengers with Captain America at the helm, the villain crew led by some pretty powerful foes like Doctor Doom, and even the odd couple of Ghost Rider and Spider-Man, who will both take charge at some point. Of course, there are additional teams that have yet to be announced. The 4v4 tag-team mode allows you to swap in friends mid-fight, and the combination of Arc’s tight controls with Marvel’s explosive powers promises to be a crazy ride. And for those that enjoy multiplayer, you’ll be able to join forces with dozens of other players online, resulting in unending rivalries.

When it comes to the single-player experience, Episode Mode introduces a completely new storyline. Arc System has created a series of motion comics with voice acting in ten different languages, drawing on a blend of American comic books and Japanese manga. The PS5’s spatial audio is incredible, and the DualSense controller makes every impact feel like it’s hitting directly in your lap. According to producer Yamanaka, this will be a new way to experience the Marvel universe, unlike anything seen in books or films and only feasible due of the games.

Finally, there are several special editions, and they provide some great alternatives for collectors. The standard edition costs $59.99 and includes the base game. If you want to go all in, the Digital Deluxe edition costs $84.99 and includes the game as well as a slew of post-launch bonuses such as four new characters, a brand new stage, a Howard the Duck skin, and even a Cosmo pet. The top-tier Ultimate Edition costs $99.99 and includes some very sweet-looking costumes for several of the characters, including Spider-Man’s beautiful new suit.

Gaming laptops aren’t getting any cheaper, and HP seems to have decided that fighting rising hardware costs head-on isn’t the answer. Instead, the company is reportedly exploring something different: letting gamers rent their laptops instead of buying them outright. While the service has actually been available for a couple of months, it’s been brought into the notice again owing to the shift in the market.

HP seems to be experimenting with subscription-style access to gaming machines, where players pay a monthly fee to use high-end laptops rather than dropping a large upfront sum. The idea is simple on paper. Instead of spending thousands on a new gaming rig, you spread the cost out like a Netflix plan, with HP handling upgrades, servicing, or replacements behind the scenes.

For some players, that could lower the barrier to entry. A powerful gaming laptop becomes a smaller monthly commitment rather than a major one-time purchase. It also means access to newer hardware more frequently, which is appealing in a space where GPUs and CPUs age fast. At a time when memory prices and component shortages are pushing system costs higher across the industry, the rental pitch might feel practical.

Renting your rig is convenient, but is it the future you want?

But there’s a bigger shift happening here that’s worth pausing on. Renting hardware fits neatly into a broader tech trend where ownership slowly gives way to subscriptions. First it was movies and music, then software, and now even games through cloud services. With platforms like NVIDIA GeForce Now and Xbox Cloud Gaming, players are already streaming titles they don’t locally own. HP’s approach pushes that one step further: you might not even own the device running them.

On the one hand, it’s flexible and potentially cheaper in the short term. On the other hand, it means you’re effectively paying forever. Stop the subscription, and both the laptop and access disappear. No resale value, no long-term asset, and no tinkering or upgrading on your own terms. For budget-conscious gamers, renting could make sense as a stopgap. But if this model becomes the norm, the industry might quietly move from “buy and own” to “subscribe and borrow.” That’s convenient, sure, but it also changes what gaming hardware really means.

So while HP’s rental idea may solve today’s pricing crunch, it also raises a bigger question: do you want your next gaming rig to be yours, or just temporarily checked out?

Apple TV has a slimmer-than-usual lineup of new movies on the way this year, and while that’s disappointing in itself for us film buffs, what’s even more notable is the lack of theatrical releases that have been announced.

If you saw my roundup of new Apple TV movies and shows coming in 2026 last week, you’ll already know that the studio doesn’t have a blockbuster-worthy film slate either. For instance, how do you release a movie about one of Mattel’s most popular kids’ toys and not expect even a sliver of the same reception that Barbie had?