Technology

The UK's Online Safety Act comes into force today, marking a shift in internet culture that is being felt around the world.

• goo.gl links will stop working on August 25, 2025
• Some users were being blocked from creating new goo.gl links years ago
• Google began warning users the links would stop working in 2024

A little over a year the company confirmed Google URL Shortener links will no longer be available, Google is still sticking to its guns, meaning we're just weeks away from the end of goo.gl links.

From August 25, 2025, all goo.gl links will stop working, returning an 404 error. This comes around a year after interstitial warning pages started appearing for some goo.gl links, stating that they would stop working soon.

Google had already closed down the goo.gl URL shortener in 2019 due to changes in how people find content online, with 99% of goo.gl links reporting no activity in June 2024.

goo.gl links will stop working from August 25

URL shorteners have generally become less relevant, however developers could still see some impact from their deprecation. For example, goo.gl links embedded in 302 redirects or with social metadata may fail to function correctly.

Interestingly, links from Google apps like Maps will continue to work, even after the shutdown.

Although users were able to bypass the interstitial page by adding the query param “si=1” to existing goo.gl links, the impending deprecation means that users and developers will now need to transition their links to another URL shortener or risk disruptions.

The Google URL Shortener lived for a relatively short nine years, from 2009 to 2018. When the company first confirmed anonymous and new users would no longer be able to create new goo.gl links starting April 13, 2018, it pointed users in the direction of bit.ly and ow.ly.

Although tech companies often get slated for enacting pretty major changes with insufficient notice, goo.gl users have had around seven years to get ready for the change, and with fewer than 1% of goo.gl links reporting activity a year ago, the impacts are likely to be minimal.

Anyone looking to re-situate or re-build their online presence should check out our list of the best website builders around, as well as advice on how to choose a domain name for your website.

Via The Verge

You might also like

• Build yourself an online presence with the best free website builders
• Google is closing its URL shortening service
• We've listed the best web hosting services

I love the Oral-B Pro 1000 so much that I don't miss my former $200 smart electric toothbrush. It's also a CNET Editors' Choice Award winner.

• GitHub wants the EU to create a Sovereign Tech Fund for OSS maintenance
• Microsoft hasn't committed to any contributions as yet
• GitHub says OSS is crucial infrastructure, just like roads

Microsoft-owned developer platform GitHub is urging the European Union to establish a publicly funded Sovereign Tech Fund to support open source software (OSS) maintenance.

GitHub Director of Developer Policy Felix Reda explained in a blog post that open source continues to be underfunded, and that the public sector could get involved to help financially support development.

The proposal is based on a study commissioned by GitHub and conducted by Open Forum Europe, Fraunhofer ISI and the European University Institute, and describes open source software a critical digital infrastructure that economies and societies rely on.

GitHub wants the EU to fund open-source development

Reda, together with the research, noted open source remains underfunded compared with traditional, physical infrastructure like roads, despite contributing €65-95 billion annually to the EU economy, and up to $8.8 trillion globally.

The survey showed one in three open source software maintainers are unpaid, with another one in three unable to make a living solely from their open source work.

As such, GitHub is proposing that the EU adds €350 million to its budget to fund open-source software maintenance. National governments and industries should also contribute to funding, GitHub believes, although the Microsoft-owned company has not volunteered any contributions itself.

GitHub identified five key investment areas, including: identifying critical EU OSS dependencies; investing in OSS maintenance; funding OSS security improvements; supporting project enhancements; and strengthening the overall OSS ecosystem.

Supporting GitHub's argument, Mercedes-Benz Chief Software Officer Magnus Östberg wrote: "Without sustainable funding and support, it is entirely foreseeable that ever more open source software projects will not receive the diligence and scrutiny appropriate for software of such criticality."

With the first legislative proposals for the EU budget "hit[ting] the desks" of the European Parliament and the national governments in the Council of Ministers, GitHub is urging the community to voice their support for the Sovereign Tech Fund.

You might also like

• We've listed the best productivity software
• Save some cash by installing the best free office software
• Open source software is now a multi-billion dollar industry

• Assassin's Creed Shadows is getting a bunch of updates throughout the year
• It starts with a New Game Plus update next week
• Further changes and an expansion are set to arrive in September

Ubisoft has provided a huge update for Assassin's Creed Shadows' upcoming content roadmap.

Revealed in a post to the official series website, the 'summer roadmap' begins with a progression update, arriving next week on July 29. Headlining this update is the arrival of New Game Plus, which lets players carry over character level, gear, skill points, Hideout progression and knowledge rank into a new game after they've rolled credits.

The progression update also brings a level cap increase from 60 to 80, new knowledge ranks, a new forge level and the addition of more achievements and trophies (depending on your platform of choice).

At the start of September (date TBC), Ubisoft will bring a 'quality-of-life improvement update' to Assassin's Creed Shadows. This free update will finally allow players to fast forward the time of day, 'unfog' the world map, and bring more Hideout and general content updates.

This update will also uncap the frame rate in cutscenes, though this is a feature that will only be found in the PC version of the game.

Lastly, on September 16, Assassin's Creed Shadows will receive its first story expansion. Known as Claws of Awaji, the expansion adds a new region to explore, as well as all-new weapons, abilities, gear, and enemies.

Ubisoft claims the expansion will add more than ten hours of content, and will unlock only after you've beaten the base game story. Likely, this means the expansion will carry on the story from the original ending.

You might also like...

• Battlefield 6 gets an action-packed first look along with a promise of a multiplayer showcase next week
• The new Secretlab OTT Adjustable Legrest is a versatile accessory that will go perfectly with any Secret gaming chair
• Splitgate 2 is being 'pushed back to beta' as 1047 Games shuts down the original Splitgate and undergoes more layoffs

• The Samsung Galaxy S26 Ultra could have a wider aperture for two of its cameras
• It might also be thinner and lighter than the S25 Ultra, and have faster charging
• However, the main sensor might not change, and there are disagreements over whether the battery will increase in size

There has been a flurry of Samsung Galaxy S26 Ultra rumors in the last day or so, and they’re largely promising – pointing to improved cameras, higher charging speeds, and more, but there are some elements to the tips that sound less positive.

Starting with the cameras, reputable tipster @UniverseIce has claimed that the Samsung Galaxy S26 Ultra will have a new lens with a wider aperture for its main 200MP camera. This should improve low light photos in particular, but they suggest all kinds of photos taken with this lens will benefit, with this apparently being “the most obvious upgrade since the Samsung Galaxy S20 Ultra.”

So that certainly sounds promising, though what’s less promising is that this new lens apparently won’t be paired with a new sensor, with the ISOCELL HP2 supposedly being used again.

BREAKING！Galaxy S26 Ultra :HP2+ large aperture ！will greatly increase the light input, improve the dim light quality, and improve all existing problems. The actual effect it brings will be the most obvious upgrade since Samsung Galaxy S20 Ultra.July 25, 2025

The same source has made numerous other claims about the Samsung Galaxy S26 Ultra over on Chinese social media site Weibo, including that its 50MP 5x telephoto camera will also apparently have a wider aperture than on the Samsung Galaxy S25 Ultra.

On the same site they’ve also said that the Galaxy S26 Ultra will be under 8mm thick, which is down from 8.2mm for the Galaxy S25 Ultra. It will also apparently be a few grams lighter than the 218g Galaxy S25 Ultra.

Faster charging and a debated capacity

Sadly though, in slimming the phone down Samsung might not be leaving space for a larger battery, with this source also claiming that will once again come in at 5,000mAh. But it will apparently at least charge at a higher 60W – up from 45W on the Galaxy S25 Ultra.

And there is some disagreement on the battery side of things, with leaker @chunvn8888 claiming the Samsung Galaxy S26 Ultra will actually have a larger 5,500mAh battery and 65W charging.

They also have a solid track record, though given Samsung’s seeming reluctance to go beyond 5,000mAh batteries, we suspect @UniverseIce is right in this instance.

We’ll probably find out for sure in early 2026, as that’s when the entire Samsung Galaxy S26 line is likely to launch.

You might also like

• There could be big changes coming to the Samsung Galaxy S26 line – but not to their screens
• Two Samsung Galaxy S26 models could be in for camera upgrades – and their codenames have leaked too
• Samsung explains why the Galaxy Z Fold 7 doesn't have a bigger battery, but its reason isn't entirely convincing

• Most firms fail high-severity compliance checks, report claims
• Many fall short at critical levels, putting them at risk
• Misconfigured firewalls could lead to downtime and cyberattacks

Firewalls are a staple of corporate cybersecurity and, next to multi-factor authentication (MFA) and endpoint protection solutions - essentials in every security tech stack.

However, how many companies run a misconfigured firewall which gets in the way instead of helping?

A new report from FireMon found 60% of enterprise firewalls fail high-severity compliance checks “immediately upon evaluation”, with another third (34%) “failing short at critical levels”.

How to stay safe?

For the researchers, this is a sign of deeper governance issues that could result in audit failures, operational downtime, or increased threat exposure.

The problems are not contained to a single environment - on-prem, cloud, and hybrid, all suffer from the same woes - misconfigurations, outdated rules, and bloated policies, leading to reduced performance, compliance risks, and more.

FireMon found that 95% of application objects and 82% of service objects show zero usage, which means they are unnecessary overhead and are just expanding the attack surface.

A third (30%) of firewall rules are completely unused, too , with 62.6% lacking any owner or documentation, leading to audit gaps and operational blind spots.

Finally, more than 10% of rules are either redundant or shadowed, reducing performance and hiding dangerous misconfigurations.

“Firewall complexity isn’t just a configuration issue, it’s a threat to resilience and trust,” said Jody Brazil, CEO and founder at FireMon. “Security teams are buried under policies they can’t explain, map to business objectives, or manage at scale.

You might also like

• VPN vs Firewall
• Take a look at our guide to the best authenticator app
• We've rounded up the best password managers

• Intel plans to cut even more workers by the end of the year
• Its most recent quarterly revenue was better than anticipated
• Other cost-cutting measures are being used

Intel is reportedly planning to reduce its workforce by around 15% by the end of 2025, bringing total headcount to the 75,000 mark.

Thousands of Intel workers have already lost their jobs in 2025, including a staggering 22,000 in April and a further 5,000 in July.

Intel has described the changes as cost-cutting efforts, through which it hopes to reduce middle management layers that cause operational efficiencies.`

Intel continues to cut jobs

Despite the economic challenges it faces, Intel managed to post a flat year-over-year quarterly revenue, at $12.9 billion, however the period saw the company rack up an estimated $1.9 billion in restructuring charges.

Speaking about Intel's second quarter, new CEO Lip-Bu Tan praised the company's performance: "We delivered revenue above the high end of our guidance, reflecting solid demand and execution across the business."

"The changes we are making to reduce our operating costs, improve our capital efficiency and monetize non-core assets are having a positive impact as we work to strengthen our balance sheet and position the business for the future," CFO David Zinsner added.

Besides tackling inefficiencies by removing thousands of workers, Tan is also an office-working advocate, citing enhanced productivity as a key reason behind the company's recently adjusted return-to-office policy which he says is "on track" for a September 2025 implementation.

In his letter to company workers, Tan explained Intel had previously invested "too much, too soon" in its foundry businesses, and that some changes would be coming that way.

Previously planned projects in Germany and Poland have been pulled, while assembly and test operations in Costa Rica will be consolidated to Vietnam and Malaysia.

Tan stressed that Intel remains committed to its x86 chips, like Panther Lake and Nova Lake consumer and enterprise models and the data centre-destined Granite Rapids.

"We must continue acting with urgency, discipline and focus in everything we do," he concluded.

You might also like

• Worried about the news? These are the best job sites and best recruitment platforms
• Intel set for huge factory job cuts as it makes a major policy shift
• Check out our roundup of the best processors, including Intel models

Too often, organizations still delay serious conversations about compliance until they’ve reached a certain size or revenue milestone. But this reactive approach is inherently risky; recent SurveyMonkey data shows that 52% of UK businesses have experienced data-related issues since GDPR came into effect seven years ago. Addressing the issue only after it becomes a legal or reputational crisis is, quite simply, too late.

Data protection isn’t merely a big business issue, and it should not be only a legal or IT concern; it is an organization-wide responsibility for every business. Yet, many companies focus exclusively on appointing a team or employee with special responsibility for privacy, or on purchasing automation tooling and frameworks.

In doing so, they overlook the most critical component: without a strong culture, even the most robust tools and policies fall flat. Culture, not just compliance, is what truly transforms policy into sustained practice.

Data protection is everyone’s responsibility

One of the most common misconceptions about data protection is that it belongs solely to legal, compliance, or IT teams. In reality, nearly every department interacts with sensitive data daily. Marketing teams manage customer preferences, procurement assesses third-party vendors' system access, and product teams collect user insights. Even frontline employees, increasingly using AI tools, can inadvertently expose data. It’s not about how many employees you have, but the volume and sensitivity of the data you manage, and how exposed that makes your organization.

That’s why data protection must be treated as a shared, organization-wide responsibility. The goal isn’t to turn everyone into a compliance expert, but to foster shared understanding and accountability. This empowerment encourages teams to proactively flag issues, without fear of blame, when something doesn’t feel right, and to educate on the pathways for escalation. Such vigilance significantly bolsters an organization's defenses and reactions to both malicious cyberattacks and human error.

Crucially, businesses shouldn’t wait until they’ve scaled to embed this mindset. The earlier organizations integrate accountability into their everyday thinking, the easier it becomes to maintain trust and mitigate risk as it grows. However, even with shared responsibility in place, policies remain ineffective unless people truly understand why they matter.

Asking ‘why’ should sit at the heart of every data policy

A common challenge is that many employees don’t fully understand the fundamental importance of data protection. Everyone across an organization needs to understand the types of data they’re working with, the potential risks involved, and the guardrails in place to prevent misuse. Without a clear sense of purpose, data policies risk becoming an afterthought, or worse, a box-ticking exercise.

For data protection to be truly effective, leadership needs to define and communicate the ‘why’ of the organization’s data protection program and policies, and that should be embedded into the cultural fabric at every level—not just during formal training or audits, but in everyday discussions and decisions. Leaders play a crucial role in setting this tone by openly questioning the rationale behind data decisions and discussing their broader impact. Beyond leadership, businesses can actively create space for employees to ask questions by making the ‘why’ visible in practical ways.

This could mean gathering feedback from employees through a survey tool to understand how much employees understand privacy policies and decisions, incorporating short, plain-language explanations into every new policy rollout or tool to address knowledge gaps, detailing not just what the rule is, but why it exists. It could also involve having privacy champions or stewards who introduce regular ‘privacy moments’ in team meetings, where staff are encouraged to bring questions or explore hypothetical scenarios related to data use.

When employees grasp the ‘why’, understanding the potential risks to the business, customer trust, and legal standing becomes second nature, rather than a chore. Embedding this understanding into daily behavior starts with how we continuously train and support teams.

Continuous, role-specific training and awareness

Ongoing, tailored training is essential for effective data protection. By offering targeted training paths and regularly reinforcing policies, organizations can help employees stay ahead of evolving risks and technologies. This training should directly reflect the specific data challenges of each role: from marketing teams managing customer consent, to procurement staff assessing vendor risk, to developers handling user data securely, and beyond.

At a minimum, organizations should provide formal training every year, with shorter refreshers or updates quarterly for high-risk or data-heavy roles and privacy champions. Scenario-based exercises, such as simulated phishing attacks or breach response tabletops, help reinforce learning and build practical resilience. Regular touchpoints, whether through internal updates, team discussions, or quick tips, keep awareness high and help embed data protection into day-to-day thinking.

Prioritizing this continuous awareness not only ensures organizations are compliant today but also prepares them for the challenges of tomorrow. With 92% of UK business leaders saying strong data privacy gives them a competitive edge, this commitment to awareness reduces risk and also builds trust.

Ultimately, getting data protection right isn’t just about avoiding fines or ticking regulatory boxes; it’s about earning and maintaining trust. That means recognising it's a company-wide responsibility, clearly communicating the ‘why’ behind every policy, and equipping teams with role-specific training to stay ahead of evolving risks.

It also means ensuring accountability flows from both the top down and the bottom up. Whether you're a start-up or a scaling business, building this foundation early creates a safer, smarter, and more trusted organization.

Now is the time to invest in your data culture, not just the controls, because true protection starts with people.

We list the best privacy tool and anonymous browser.

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

“It turned out that getting fired from Apple was the best thing that could have ever happened to me.”

You wouldn’t expect to hear that from Steve Jobs, a man famous for his short temper and strong opinions. Yet as it turns out, being removed from his own company transformed Jobs, helping him grow into the creative force that, years later, would spawn world-changing products like the iPhone, the iPod, and the best Macs and MacBooks.

His ouster went down 40 years ago in 1985, and it’s worth looking back on what happened at that time and how it changed Jobs – and Apple, and perhaps even the wider computing world – for the better.

Far from being the end of the road for Jobs, this was just the beginning. After leaving Apple, he founded another computer company, NeXT, and later took the helm at Pixar. Under his watch, it became the foremost animation studio on the planet. In his absence, Apple lost its way, churning out uninspired products and coming within a hair’s breadth of going bankrupt.

Twelve years later, Jobs returned to Apple and led it to one of the most dazzlingly potent periods in its history. Yet that likely would never have happened had Jobs not been very publicly fired from his own company. Here, I want to show you why.

Showdown at Apple

(Image credit: Apple)

By the mid-1980s, Steve Jobs was running the Macintosh division at Apple, taking the lead on a product that would go on to change the world of computing forever.

At the same time, Apple was searching for a new CEO, and Jobs thought he had found one in the form of Pepsi marketing chief John Sculley. After luring Sculley in with his famous powers of persuasion (“Do you want to spend the rest of your life selling sugared water, or do you want a chance to change the world?” Jobs asked him), the new hire was on board.

But after a year-long honeymoon period, cracks began to appear. Jobs felt Sculley was not a 'product person' and simply didn't get his way of thinking. For Sculley, Jobs was too obsessive and obnoxious.

The showdown came in 1985, when disappointing sales of the Macintosh and Jobs’s chaotic management style led to questions over his leadership abilities. Although it was suggested that Jobs could run a small “skunkworks” team developing exciting new moonshot devices, Jobs rejected the idea, leading Sculley to conclude that he had to be forced out of the company.

Eventually, Sculley convinced the board to remove Jobs from his role as head of the Macintosh division, a plan that was executed over a few days in May 1985.

Jobs was given the symbolic role of Chairman of the Board, but it was a position with no actual power and influence inside Apple. He had been trapped inside the airlock, unable to influence the company or the direction it took. Jobs recognized this, and in September 1985, he left to form his own computing company called, fittingly enough, NeXT.

He was 30 years old and exiled from the successful company he had founded ten years earlier. For many people, it would be a chance to leave the industry and never look back. But for Jobs, it was the beginning of a learning process that brought him back stronger than ever.

Finding focus

(Image credit: Apple)

Jobs (unsurprisingly) took his firing very personally, feeling that he was both a personal failure and that he had let down those around him. Yet this didn't stop him from leading companies and getting hands-on with their operations.

At first, he exhibited some of his worst habits at NeXT. Released from the strictures of a board and CEO that didn’t look kindly on his antics, he indulged his taste for perfectionism at every turn, regardless of the cost or time implications.

As such, NeXT struggled to put out a competitive product on time and under budget. After a few years of little progress, it exited the hardware world altogether, instead focusing entirely on software. This would later prove to be a fortuitous decision.

But it was at Pixar that he truly turned things around. After becoming CEO, he took a more hands-off approach and focused on funding and dealing with Disney, with which Pixar had partnered to make films. Pixar was led by a team of artists and creatives – a class of people who Jobs truly admired – and he was happy to let them do their thing.

Pixar became the most important animation studio in the world, with a string of hits like Toy Story, Monsters Inc., and A Bug’s Life. Jobs could marry technology with art – the thing he believed he was doing at Apple, and an idea that would define how he worked later in his life. Jobs had found a way to be focused, creative, and productive without imploding.

The value of being fired

(Image credit: Getty Images)

In the late 1990s, Sculley was gone and Apple was nearly bankrupt, but it had enough cash to buy NeXT and bring Jobs back on board, first as interim CEO, then as CEO in a full-time capacity. NeXT’s operating system – now part of Apple’s stable – went on to become the foundation of what we now know as macOS.

When Jobs returned to Apple, he was a changed man. Under his renewed leadership, the company went on an incredible creative tear: the iPhone, iMac, iPod, iTunes, and much more were produced under his watch. He not only brought Apple back from the dead, but made it one of the strongest – and hippest – tech companies in the business.

Yet Jobs believed that none of this could have happened if he had not been fired from Apple. As he said in the commencement address he gave at Stanford University in 2005, years after being ousted from his own company, “The heaviness of being successful was replaced by the lightness of being a beginner again, less sure about everything. It freed me to enter one of the most creative periods of my life.”

Perhaps the most important aspect was that Jobs realized he still loved what he was doing. “Your work is going to fill a large part of your life,” he said, “and the only way to be truly satisfied is to do what you believe is great work. And the only way to do great work is to love what you do.”

For Jobs, it took being fired from Apple to realize that. But things could have been very different if it had never happened at all.

You might also like

• 24 years ago, Steve Jobs made tech sexy again
• If you want to be like Steve Jobs, this is the one thing you must do
• The original Apple Computer cost $666.66, but you can get Steve Jobs' Polaroids of it for $2148

Pepsi will launch its very own prebiotic soda this year, but are these "healthy sodas" actually healthy? We asked experts.

Full spoilers follow for The Fantastic Four: First Steps.

The Fantastic Four: First Steps has achieved lift off in theaters – and I imagine you have plenty of questions about the Marvel movie's ending and its post-credits scenes.

Below, I'll do my best to answer your biggest questions about the first Marvel Phase 6 project, including how it sets up the next Avengers film, aka Avengers: Doomsday. So, here's your final warning: major spoilers immediately follow for First Steps. Turn back now if you haven't seen it yet.

What happens to Galactus and The Silver Surfer in The Fantastic Four: First Steps?

What becomes of Galactus and his Herald during First Steps' climax? (Image credit: Marvel Studios)

They're transported to the furthest reaches of the alternate universe – designated Earth-828 – that The Fantastic Four reside in. Without his planet-devouring space vessel, it'll take the now-stranded Galactus millions of years to make his way back to the quartet's home world, and that's even if he's capable of doing so.

So, how will the eponymous team prevent Galactus from destroying their planet? After all, they refused to hand over Reed Richards and Sue Storm's son Franklin, which would've spared their world. Galactus sees Franklin as his successor because, according to the terrifying space god, Franklin possesses the Power Cosmic, a source of unlimited energy that turned the former mortal man known as Galan into the immortal, never-satiated entity known as Galactus.

Anyway, the gang aims to defeat Galactus by using a scaled-up version of the teleportation device Reed was working on earlier in the film. You know, the one that he used to move an egg from one transmat machine to another in his lab.

Galactus comes to Earth-828 to kidnap Franklin Richards (Image credit: Marvel Studios)

With the help of every nation on Earth, Marvel's First Family initially builds a series of giant teleporters with the intention of moving their world to a different universe. That way, Galactus won't be able to reach them. However, that plan fails when The Silver Surfer destroys all but one of the devices.

Reluctantly, Reed, Sue, and their Fantastic Four cohorts decide to use Franklin as bait to ensnare Galactus. They plan to place the baby in a carrier under the last remaining teleporter. Once Galactus steps inside its field range, the group will distract him by attacking, while one of them replaces the baby carrier with an empty one. Then, they'd activate the machine and flee the scene as a trapped Galactus is transported to whereabouts unknown.

It's a plan that almost works, but before he steps into the teleporter's area of effect, Galactus realizes Franklin has been whisked away to the Baxter Building, aka the group's headquarters, and makes a beeline for him. Long story short: despite their best efforts, The Fantastic Four can't stop Galactus from taking Franklin.

All hope appears to be lost until Sue Storm regains consciousness, after being knocked out when Galactus swats the Fantasticar into a building, and sees the immortal cosmic entity leaving with her only child.

Sue Storm is the most valuable player of First Steps' third act (Image credit: Marvel Studios)

Summoning all her strength, which is heightened by her maternal instinct to protect her offspring, Sue uses her force field powers to slowly force Galactus toward the teleporter. The rest of the team, who have come to their senses after also being taken out by Galactus, rush to help. Ben Grimm uses his strength to knock down a building that Galactus is steadying himself on, Johnny Storm activates the device, and Reed uses his stretch powers to climb Galactus and rescue Franklin.

With one final push, an over-exerted Sue shoves Galactus through the portal. Exhausted, she collapses to the ground.

But, it doesn't seem to have been enough as Galactus starts to re-emerge through the portal. Johnny, who tried to negotiate with The Silver Surfer earlier and offer his services to Galactus in exchange for sparing Franklin and his planet, decides to make one last Hail Mary move.

Aiming at Galactus, he flies towards the portal in a self-sacrificial move. However, he's knocked out of the way by The Silver Surfer. Clearly, the pair's earlier exchange appealed to her humanity, and The Silver Surfer makes the sacrifice to force Galactus back through the portal before it closes, thus stranding them both in the far reaches of this universe.

Reed and Sue's baby boy brings the latter back to life with his own superpowers (Image credit: Marvel Studios)

With Galactus gone, Reed, Johnny, and Ben race to Sue's side, but try as Reed might to revive his wife, it seems he's too late. Sue has died.

Or so it seems. A distraught Franklin wriggles out of Ben's and then Reed's arms to be with his mom. Lying on her chest, Franklin smiles and quietly places his hands on her. Miraculously, the fingers on Sue's right hand twitch, she gasps for air, and is resuscitated. Smiles, hugs, and happy tears all around.

Does The Fantastic Four: First Steps have a mid-credits scene?

Johnny Storm and Ben Grimm don't appear in First Steps' mid-credits stinger, but someone else does (Image credit: Marvel Studios)

The joy of defeating Galactus, plus the relief of Sue's revival, are somewhat short-lived, though, because of what happens in the first of two end credits stingers.

In it, we learn that four years have passed since Galactus was banished. Sue and Franklin are sitting in the Baxter Building's living room and have just finished reading 'The Very Hungry Caterpillar'. Considering Galactus wanted Franklin to replace him as this universe's planet-consuming deity, it's quite an on-the-nose book for Sue and Franklin to have *ahem* devoured.

Anyway, Sue walks away to find 'A Fly Went By,' another of Franklin's favorite books. Locating it, she returns to her son, but hears a strange noise as she does so.

Activating her powers, Sue (and, by proxy, the camera) peers around the living room's central column to see that Franklin is not alone. A mysterious figure is not only squatting down in front of Franklin and holding his right hand to Franklin's face, but he's wearing a dark green hooded cloak and holding a metal mask in his left hand. Say hello to Doctor Victor von Doom, everyone!

So, is that Robert Downey Jr playing Doctor Doom in The Fantastic Four: First Steps?

First Steps answers the biggest question on everyone's minds: is Doctor Doom in it? (Image credit: Marvel Studios)

It's not confirmed, but I think it is.

Let's break down what we know. At San Diego Comic-Con 2024, Downey Jr, the superstar actor who played Tony Stark and his heroic alter-ego Iron Man in the Marvel Cinematic Universe (MCU) between 2008 and 2019, was confirmed to be playing fan-favorite Marvel supervillain Doctor Victor von Doom. It was an announcement few people expected, so there was shock, delight, and the odd eyebrow raised when the reveal was made.

Downey Jr's return to the MCU came after Marvel fired Jonathan Majors in December 2023. The disgraced actor, who was found guilty of numerous crimes, including sexual assault, had been hired to play Kang the Conqueror, aka the Marvel Multiverse Saga's Big Bad. He even appeared in projects like Disney+ series Loki and Marvel Phase 5 film Ant-Man 3, so Marvel was all-in on Kang being this saga's primary nemesis.

However, after Majors was found guilty of the allegations made against him, Marvel quickly moved to replace Kang with Doom. Latveria's most famous son, then, is the Multiverse Saga's new overarching antagonist, and will square off against our heroes in Avengers: Doomsday and its sequel, Avengers: Secret Wars.

With First Steps being the precursor to that pair, fans began to question whether Doom would appear in this Phase 6 flick. However, director Matt Shakman consistently denied that would be the case. Indeed, he said it wasn't part of his purview and later told me that he never considered installing Doom as First Steps' primary villain.

Downey Jr is expected to have top billing as Doctor Doom in Doomsday and Secret Wars (Image credit: Robert Downey Jr's Instagram)

Ahead of First Steps' release, though, Shakman's tone has changed. Speaking to ComicBook.com, he confirmed that Joe and Anthony Russo, who are directing the next two Avengers films, shot First Steps' mid-credits scene. The siblings also filmed Thunderbolts* post-credits stinger, which has big ties to The Fantastic Four's latest big-screen adventure and Doomsday.

So, is Downey Jr playing Doom in First Steps' mid-credits scene? That's unclear. We only see Doom from the back, and his hood is raised, so I can't say for sure if it's Downey Jr. The A-lister isn't credited with an appearance in the film's end credits crawl, either. For someone so synonymous with the MCU, you'd expect Downey Jr's name to pop up if he actually shows up in First Steps.

That said, considering the Russos shot this sequence, I'm more inclined to believe Downey Jr is playing Doom in this scene. First Steps underwent some minor reshoots and pickups around the time that principal photography began on Doomsday. It's plausible, then, that the Russos filmed this scene prior to commencing work on Avengers 5, which would make it more likely that Downey Jr appears (albeit uncredited) in this footage.

Is there a post-credits scene in The Fantastic Four: First Steps?

First Steps' post-credits scene shows the main title sequence of The Fantastic Four's in-universe cartoon (not pictured above) (Image credit: Marvel)

Yes, but truth be told, it's a nothing burger (from an MCU impact, anyway) of a post-credits scene.

Indeed, the main title sequence to The Fantastic Four cartoon series, which exists in-universe, is this movie's post-credits scene. It's a fun little sequence to include, and it's nice to see those who created it recognized for their work. As I said, though, it doesn't drop any hints about the future of Marvel's cinematic juggernaut, so it's ultimately not worth sticking around for.

How does The Fantastic Four: First Steps set up Avengers: Doomsday?

Avengers: Doomsday will be released in cinemas worldwide in December 2026 (Image credit: Marvel Studios)

Potentially big spoilers follow for Avengers: Doomsday. Proceed at your own risk.

Given what we learn in First Steps and Thunderbolts* post-credits scene – read my Thunderbolts* ending explained piece for more on the latter – some puzzle pieces appear to be falling into place.

For one, we now know Franklin has superpowers. If we believe what Galactus said about Franklin, i.e., he's a being of unlimited power, he'll be one of the most powerful beings to ever exist. Franklin is immensely powerful in Marvel Comics, so that checks out.

Thanks to Earth-828's various news agencies, it'll be public knowledge that Franklin possesses superhuman abilities, too. Rumors might also spread about what powers he has and/or how he resurrected Sue – well, if the general public finds out that he did. If they do, Doom will become aware of Franklin's powerset.

Furthermore, if you believe online speculation about Doomsday's plot – read my Avengers: Doomsday hub for more information on its confirmed details – Doom may actually try to prevent the Multiverse's collapse (via those pesky events known as Incursions) by merging all known universes into a single reality. That's the basis of Marvel's two 'Secret Wars' comic book runs, which the Russo brothers told me they're taking inspiration from for Avengers 5 and its follow-up.

Franklin Richards could hold the key to creating Battleworld, the planet created out of fragments from various universes (Image credit: Marvel Comics)

If that's Doom's aim in Marvel's cinematic franchise, he'll need someone like Franklin to achieve his goal. In Marvel literature, Franklin has the ability to warp reality, manipulate matter, and create pocket dimensions. All of those powers would come in handy for an egotistical individual like Doom, who'd view himself as the savior of the multiverse and, as he does in 2015's 'Secret Wars' comic book storyline, install himself as God Emperor of the MCU's take on Battleworld.

Okay, how does this tie into First Steps' mid-credits scene and Thunderbolts* post-credits stinger?

The former makes it clear that Doom has found Franklin. The fact that he has means he'll have some grasp of the unlimited power Franklin possesses. Franklin's mom won't be pleased (to put it lightly!) that another supervillain has come calling to kidnap her child, either.

My best guess, then, is that The Fantastic Four will collectively fight off Doom and, using the Excelsior and Reed's likely upgraded teleportation device, flee to another universe not unlike their own – i.e., the MCU. That's why we see the Excelsior emerging from a wormhole in the final Thunderbolts* end credits stinger.

The Thunderbolts – or, rather, The New Avengers – spotted the Excelsior entering Earth-616's orbit in their film's post-credits scene (Image credit: Marvel Studios)

Of course, that doesn't mean they're safe. The MCU's take on Earth's Mightiest Heroes will want to know why a group of superpowered beings has suddenly shown up from another reality. They're used to extraterrestrial threats emerging out of the blue, and if they don't get an answer they like, you can bet a fight will ensue between the MCU's Avengers and The Fantastic Four.

Considering Doom's own genius-level intellect, plus the plan of action he's put in place, it won't be long until he tracks down The Fantastic Four to Earth-616. Franklin may be temporarily safe from Doom's clutches, but it won't stay that way.

Whether I'm proved correct or not, I'm convinced Franklin – and Loki, who currently sits at the center of the multiverse in an attempt to hold it all together – is central to Doomsday's story. Don't be surprised, then, if he becomes the human McGuffin that Avengers 5 revolves around and is the individual who contributes most to the soft reboot that Marvel President Kevin Feige confirmed to be coming post-Secret Wars.

You might also like

• I watched The Fantastic Four: First Steps, and it heralds a much-needed new dawn for Marvel’s First Family on the big screen
• When might Marvel release The Fantastic Four: First Steps on Disney+?
• Find out how to watch the Marvel movies in order

There are over one billion smart meters already deployed across the world - 38 million in the UK alone. These are embedded IoT devices, designed to have an ultra-small footprint and fitted with lightweight software that is constantly communicating with energy providers and the national grid.

The cybersecurity of embedded devices like these is weighed against their size and performance requirements - while we want smart meters to be secure, we also want them to operate smoothly, use up very little energy and space, and send the right data at the right time. Additional layers of encryption would increase the size of this data, potentially impacting the performance and cost of smart meter infrastructure.

But this encryption is something the energy sector, and the technology supply chain as a whole, now has to rethink. Last year, the US National Institute of Standards and Technology revealed its final standards for post-quantum cryptography (PQC) - a new form of encryption designed to protect data from potential quantum computer attacks.

Now, agencies across the US and Europe, including the NSA and NCSC, are recommending that the governments and businesses coordinate their migration to PQC so that every device is quantum-safe by 2035.

This the biggest cybersecurity transition in a generation, and a real challenge for the world’s 1B+ smart meters.

Why do smart meters need to be upgraded?

There are three main reasons for upgrading smart meters to PQC - risk, compliance, and market forces.

Risk is the word at the heart of every conversation around cryptography and cybersecurity. Every new iteration of an encryption algorithm or cybersecurity application is designed to stay one step ahead of the attackers and mitigate the chance of a breach. PQC is designed to mitigate the risk of an attack from future quantum computers, which experts anticipate will easily be able to crack current encryption standards.

When this cryptographically-relevant quantum computer emerges, critical national infrastructure (like the energy grid) will be a prime target for disruption. Therefore, energy networks need to act now to protect themselves and their data from this future risk. As vulnerable endpoints in the energy network - with the technical capability to cut off power supplies to households - smart meters need to be secured to ensure that infrastructure is protected from attack.

Providers should also have one eye on whether their smart meters comply with new regulations. Government guidelines are all recommending that hardware and software align with NIST’s PQC standards by 2035 at the latest - much sooner if your customer is the government itself. Simply put, the transition must take place, and is in fact already underway.

Finally, market forces will soon compel decision-makers still on the fence to upgrade smart meters to PQC. As migration deadlines approach, energy suppliers and hardware manufacturers who can promise PQC-enabled devices will be preferred for government and corporate contracts over those that have delayed their transition.

The challenge of upgrading smart meters

There are two parts to the smart meter PQC challenge - upgrading the millions of “brownfield” devices that are already deployed, and ensuring that the millions of “greenfield” devices currently on the production line are prepared for the upcoming PQC deadlines.

In most cases, already-deployed devices will require an over-the-air firmware update to become PQC-secure. This could be a major challenge for older memory-constrained models, and replacing this legacy hardware is likely to be the most costly part of the transition.

Where these upgrades are possible, there are physical challenges as well. Smart meters are small, embedded devices with minimal amounts of RAM and computing capacity. They are also limited on bandwidth - transmitting small amounts of data with every network communication they make. PQC implementations will need to work within these constraints, but some may run into issues.

For example, post quantum encryption keys are larger than RSA/ECC keys, meaning that a quantum-safe message is larger than those currently being sent by a smart meter.

Many smart meters rely on fixed-function hardware cryptography that is unchangeable, and cannot be upgraded in the field - this means that, on these devices, it’s not possible to update secure boot processes and maintain cryptographic agility (the ability to rapidly adapt the cryptography on a device).

Manufacturers don’t need to worry about over-the-air upgrades for “greenfield” smart meters that are still being designed, as they have a chance to protect devices before production. They will still face issues with memory and CPU, and will need to ensure that PQC is factored into their design process to ensure devices are compliant beyond 2035.

The next steps for smart meters

The first and most important step for the energy sector is to plan ahead thoroughly. 2035 is sooner than it seems - especially for large-scale digital transformation change projects - and this is a process that many companies will be hoping to finalize well ahead of that deadline.

The goal of the transition is to maintain the highest standards of security without compromising performance and without racking up unsustainable costs. Inevitably, the oldest models of smart meters that cannot receive over-the-air updates will need to be replaced - the ten-year transition timeline means that this can be factored into annual budgeting for hardware upgrades in the field, rather than through an impractical all-at-once rollout.

For all other devices - deployed and in production - manufacturers and energy providers need to identify where the most critical data on their device is transmitted from and focus on securing this as a priority. For smart meters, this means communication modules and the process by which they could trigger an energy shutoff, as these are the vectors that attackers will target first.

To navigate the challenges of migrating embedded and memory-constrained systems to PQC, smart meters will need low-footprint implementations of PQC, which are designed to apply NIST’s standards without exerting excessive demand on CPU and RAM. It’s worth bringing on PQC expertise to ensure that the right implementation for the right device is found - as robust as the PQC algorithms published by NIST are, they are also only as secure as the way they are implemented.

Manufacturers will need to factor PQC into their product roadmap. This sounds daunting, but as much as 80% of this transition will be handled in the supply chain - meaning that the vendors who supply the communication modules, HSMs, and microprocessors used in smart meters will themselves be responsible for upgrading vulnerable cryptography. The remaining 20% is the manufacturer’s responsibility - communications channels and metering software that needs to be upgraded in-house.

The key message for energy providers and device manufacturers is that this process needs to start as soon as possible. Smart meters are designed to have a long shelf life, and the risk of deploying devices in 2030 that are obsolete in 2035 is one that should be avoided.

We list the best antivirus software.

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

Here are the answers for The New York Times Mini Crossword for July 25.

Starlink has confirmed it's experiencing an outage, but says it's now "mostly recovered."

• Patriot Memory MD330 Storage Hub delivers 100W charging and 4K output, all from one tiny USB-C port
• Storage goes up to 1TB, but transfer speeds and encryption are still a mystery
• Handles video output and file storage, but reliability under pressure is not guaranteed

Patriot Memory has introduced the MD330 Storage Hub, a device that combines a unique set of capabilities in a single unit.

The company says it offers high-speed charging, local storage, and 4K video output through a single USB-C port - features which are rarely found together, especially in a device weighing just 21 grams.

This makes the MD330 less a refinement of existing accessories and more an attempt to redefine what a USB hub or docking station can be.

Designed for tight workflows

The MD330 connects via USB-C and supports 4K display output, with both mirrored and extended screen modes, and also delivers up to 100W of power, which theoretically allows it to charge laptops or handheld consoles while connected to an external display.

Whether this works reliably under sustained use, particularly with power-hungry devices, is something Patriot has yet to fully demonstrate.

These capabilities are usually reserved for more bulky setups, so questions remain about thermal handling and power stability.

Beyond video and charging, the MD330 integrates flash storage, available in capacities from 128GB to 1TB, which turns the hub into a portable archive for media or project files.

However, transfer speeds are not specified, and there’s no mention of encryption or file system compatibility - and without those details, it’s difficult to determine whether the storage is suitable for anything beyond casual use.

The MD330 aims to simplify mobile setups by combining a power bank, USB hub, and dock into one device, but that doesn’t guarantee consistent performance.

Running display output and power delivery at the same time could strain the device, especially if file transfers are also in progress.

How it manages bandwidth and thermal load across these tasks will ultimately determine whether it’s genuinely useful or just an overpromised accessory.

At the time of writing, Patriot has not revealed the price of the MD330, a key detail that could heavily influence its appeal.

Via Techpowerup

You might also like

• These are the best mobile workstations you can buy right now
• We've also listed the best mini PCs for every budget
• This is probably the best-looking mini PC with AMD's fastest AI CPU

• Google Photos is adding a feature that turns still images into six-second AI-generated videos
• The platform is also adding a tool called Remix that will remake images into different styles
• All generated content includes visible and invisible watermarks

Google Photos is rolling out new generative AI features that can transform still images into short video clips, briefly bringing anyone in the photo to life and including natural-looking motion. The Photo to Video tool employs Google’s Veo 2 AI video model, the same model deployed on YouTube, Gemini, and other parts of Google’s ecosystem. The feature doesn’t turn your snapshots into full movie trailers; it just creates six-second clips.

Once you see the option to make your images into videos, you just pick the image picture you want to animate, then choose either “Subtle movements” or “I’m feeling lucky” from the buttons below. As you can imagine, the subtle movement choice has the people in the picture move around a little bit. The model is designed to guess what might have happened in that frozen second. The other choice could do anything, perhaps even throw confetti in the air.

The update is rolling out in the U.S. on Android and iOS right now, but there are other AI tools coming later this summer to Google Photos. Most notable is the Remix feature coming in the next few weeks. Remix takes your existing photos and restyles them into looking like comic book panels, anime stills, 3D renderings, or pencil sketch art. It's an ability that Gemini and its many rivals already offer, but now it will be built directly into your photo gallery and won't need you to write a full prompt for it.

All of this comes together in a new section of the app called the Create tab, which will serve as a hub for these tools and any other AI features Google may release in the months ahead. In the near term, it will include the Photo to Video and Remix features alongside the existing collage and highlight video creators. But as Veo gets smarter and Google’s confidence grows, the possibilities could expand into any number of AI enhancements such as extended video clips, voiceovers, or multi-image stories.

The packaging is what is crucial here. This is the first time that photo-to-video generation has been embedded into a mainstream app like Google Photos, which the company claims has more than a billion users.

AI-powered video tools like Sora and Veo have generated headlines for their jaw-dropping realism and deepfake potential. But Google Photos isn’t pitching this update as a creative revolution. It’s presenting it as a memory enhancement. That said, Google doesn't want to accidentally trick anyone about where the new images and videos come from. That's why every AI-generated video or remix will carry a visible label showing that the content was created with AI. They will also each include an invisible SynthID watermark identifying the AI behind its production, the same as the one used by all of Gemini’s image and video generators.

AI photo inspiration

It's unlikely Google will simply drop these new features and move on. After all, the company has already deployed Veo 3, the latest iteration of the text-to-video model, to Gemini and YouTube for higher-quality short videos complete with synced dialogue and background audio. Tools that animate stills today may very well narrate them tomorrow.

This is more of a play for those not constantly trying the latest AI toy, but who do like to share photos and look at pictures taken by others. It's easy to poke fun at the idea of making your selfie move, but that's the sort of feature that attracts a lot of users who want to see just how animated AI can make them.

You might also like

• You can now edit images in Gemini directly
• I tried Gemini's new AI image generation tool - here are 5 ways to get the best art from Google's upcoming Flash 2.0 built-in image upgrade
• I used Google Gemini to analyze YouTube, and the results were seriously impressive - 4 ways you can use video integration to get the most from AI

This smart home addition might help you save on electricity, too.

• Photonic chips with quantum lasers are finally being built without redesigning the whole system
• These lasers work directly on silicon and still survive high heat for over six years
• University of California researchers filled the laser gap with polymers and nailed precision beam control on-chip

A new fabrication method could make photonic circuits cheaper and more practical by directly integrating quantum dot (QD) lasers onto silicon chips, a process that could influence how future smart home devices, fitness trackers, and even laptops are engineered.

The research team, led by Rosalyn Koscica at the University of California, achieved this by combining three key strategies.

They used a pocket laser configuration for direct integration, followed a two-step growth method involving metalorganic chemical vapor deposition and molecular beam epitaxy, and introduced a polymer gap-filling technique to reduce optical beam spread.

Closing the gap with careful engineering

This development addresses longstanding challenges involving material incompatibilities and coupling inefficiencies that have historically limited the performance and scalability of integrated photonic systems.

The combined efforts minimized the initial interface gap and made it possible for lasers to function reliably on silicon photonic chiplets.

As the researchers note, “Photonic integrated circuit (PIC) applications call for on-chip light sources with a small device footprint to permit denser component integration.”

The new approach enables stable single-mode lasing at the O-band frequency, which is well-suited for data communications in data centers and cloud storage systems.

By integrating the lasers directly with ring resonators made of silicon or using distributed Bragg reflectors from silicon nitride, the team has also addressed issues related to alignment and optical feedback.

One of the more surprising findings from the research is how well the lasers perform under heat.

“Our integrated QD lasers demonstrated a high temperature lasing up to 105 °C and a life span of 6.2 years while operating at a temperature of 35 °C,” says Ms. Koscica.

These performance metrics suggest a level of thermal stability previously difficult to achieve with monolithically integrated designs.

This thermal resilience opens the door to more durable applications in real-world environments, where temperature fluctuations can limit the reliability of photonic components.

It may also reduce the need for active cooling, which has traditionally added cost and complexity to past designs.

Beyond performance, the integration method appears well suited to large-scale manufacturing.

Because the technique can be executed in standard semiconductor foundries and does not require major changes to the underlying chip architecture, it holds promise for broader adoption.

The researchers argue that the method is “cost-effective” and “can work for a range of photonic integrated chip designs without needing extensive or complex modifications.”

That said, the approach will likely face scrutiny regarding consistency across large wafers and compatibility with commercial photonic systems.

Also, success in controlled lab environments does not guarantee seamless deployment in mass manufacturing settings.

Still, the combination of a compact laser design, compatibility with conventional processes, and integration of O-band functionality makes this development notable.

From data centers to advanced sensors, this silicon-compatible laser integration could bring photonic circuits closer to mass-market viability.

Via IEEE

You might also like

• These are the best cloud computing services available right now
• We've rounded up the best cloud hosting providers for your needs
• AMD ThreadRipper PRO 9995WX could sell for $13,000, over 2x the price of the 96-core EPYC 9655

Getting enough sleep can sometimes be difficult. This five-step sleep method can help.