Tech giants Apple, Microsoft, Alphabet and Nvidia are the other four companies to have previously recorded a market value of $3trn.

Amazon became the fifth company ever to see its market value surpass the $3trn mark when its shares rose by about 5pc yesterday (3 August).

Tech giants Apple, Microsoft, Alphabet and Nvidia are the other companies to have previously recorded a market value of $3trn.

Amazon hit the milestone having first reached the $2trn market capitalisation mark in June 2024. It first reached the $1trn value benchmark in late 2018.

Last Thursday (30 July), the e-commerce and cloud computing giant published its results for the second fiscal quarter of 2026, with net sales increasing 20pc year-on-year to more than $200bn.

Its operating income was up 43pc to $27.5bn, while sales at its AWS cloud infrastructure division increased 37pc year-over-year to $42.2bn, its fastest growth in 18 quarters.

Amazon president and CEO Andy Jassy said that “AWS is booming”, adding that “our AI and chips businesses each eclipsed run rates of more than $25bn”, which the company said amounts to “triple-digit percentages” growth year-on-year.

AI-driven demand for cloud and chips solutions contributed to AWS surpassing analysts’ forecasts for the quarter ending 30 June, causing the recent jump in Amazon share prices.

Jassy told investors on a conference call following the release of the earnings report that the company now projected capital expenditures for 2026 hit $220bn, up from the $200bn estimate it made in February and reaffirmed in April.

“But even at that amount, we will still not have enough capacity to meet all the demand we have in 2026, and I believe this dynamic will also be true in 2027 too,” Jassy said. “In fact, the demand we already have for 2028 is striking.”

Infrastructure spending and “investments in artificial intelligence” caused Amazon’s free cash status to see an outflow of $7.6bn for the preceding 12 months, the company said. This compares to a free cash flow inflow of $18.2bn for the 12-month period ended 30 June 2025.

Amazon has a 2GW deal with OpenAI for its Trainium capacity through AWS, as well as a 5GW deal for the same with Anthropic. The company is also investing $25bn into Anthropic.

AWS also has collaborations with up-and-coming Nvidia rival Cerebras; with Uber, for its Graviton and Trainium chips; and with Meta, to deploy tens of millions of AWS Graviton cores for its AI workflows.

Amazon is offsetting some of its AI expenses with large-scale layoffs at the company. In January, it cut about 16,000 jobs, which followed about 14,000 job cuts last October. Around 450 Irish jobs are understood to have been affected in this move.

Meanwhile, AWS could be set for a ‘gatekeeper’ designation in the EU due to the strength and prevalence of its market position as the largest cloud computing services provider in Europe.

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.

Elon Musk spent SpaceX’s first earnings call making some out-of-this-world claims about the company’s business and future prospects, while his fellow executives kept trying to bring his ideas closer to Earth — providing a hint of what’s to come now that his rocket-launching, compute-leasing, satellite-based telecom is public.

The conference call, held Tuesday, was the latest in a years-long succession of Musk making outrageous promises that his executives then have to make more digestible for the investing public — like at Tesla, where a recent analysis by TechCrunch showed that the world’s richest man is increasingly focused on futuristic topics while his colleagues spend their time talking about the actual business of selling cars.

Let’s start with one of the biggest ideas Musk floated on the call: that he expects SpaceX’s Starlink service to “deliver a majority of the world’s internet” in “less than 10 years.” Musk made the comment in the context of SpaceX preparing to launch the first “V3” versions of its Starlink satellites, which have much higher bandwidth than previous versions.

Here’s what he said:

It’s kind of hard for people to wrap their minds around this, but like, it’s not out of the question that at some point, Starlink will deliver a majority of the world’s internet, at least in countries where we’re allowed to operate, which is the vast majority of countries. So this is, you know, important to bear in mind, and it’s not in like the infinity future. It’s, you know, less than 10 years.

Contrast that with what chief operating officer Gwynne Shotwell said just a few minutes later, emphasis mine:

The significant amount of capacity we’re able to add to the Starlink constellation from the V3 satellites will enable us to continue providing even better service — and it’s pretty great already — but to do so while serving more and more customers over the world. In fact, in the years ahead, we expect Starlink will represent a significant portion of global internet traffic, which Elon also talked about.

It’s a far more lawyered-up claim to make, even though it’s still obviously ambitious.

But that was not the only instance where Musk optimistically diverged. At one point, SpaceX chief financial officer Bret Johnsen offered investors one of the few new financial targets discussed on the call. Johnsen was highlighting SpaceX’s relatively new business of renting out compute power to other AI players, which has helped the company generate billions in fresh, fast cash.

I’ll once again emphasize the big promise Johnsen made during his prepared remarks, and note how carefully he phrased it. It’s a very hedged and quite specific claim that he’s making. It’s clearly meant to excite investors while also leaving room for the company to avoid legal exposure if it misses the projection:

Looking ahead, we continue to see robust demand in all three of our business segments, but in particular in our cloud services arrangements. We see increasingly favorable economics with each agreement we sign, and as Elon mentioned, we expect the supply-demand imbalance in the compute market to continue. The current economics have translated into a less than one-year payback on our new capital deployments for compute. For example, in the first few weeks of the third quarter, we’ve already contracted an additional $6.7 billion of cloud services revenue over a six-month period that begins ramping starting in October of this year. We believe this puts us on a trajectory, including contribution from Cursor, to reach $100 billion of ARR, or annualized revenue run rate by the end of this year, based on our expected revenue in the month of December of this year.

Musk, 20 minutes later, bulldozed that carefully constructed statement before immediately inflating it:

To be clear, the $100 billion ARR in December is not a question mark. That’s… that’s what we would achieve if we basically did nothing. So like, you know, I think it may be higher than that. It probably will be higher than that.

Musk also riffed on another major prediction about overall revenue on the call, pumping up a goal that SpaceX laid out just two months ago in its IPO documents:

It’s probably also worth mentioning that our internal projections for reaching a trillion dollars in revenue, not ARR, but revenue, have moved up from 2031 to 2030. So prior to the IPO, the financial projections we had were reaching a trillion dollars in revenue in 2031. We now expect that to be in 2030. And there’s a non-zero chance of that being in 2029.

The pattern kept repeating throughout the call. A shareholder question about progress on the “human landing system” that SpaceX is developing for NASA’s Artemis moon missions using Starship prompted Musk to all but claim that the prototype rocket will be ready to fly people by the end of next year. He later said SpaceX would be flying Starship rockets once a day, or “possibly more,” by this time next year.

Shotwell immediately followed Musk’s comments about human flight to clarify that SpaceX is still focused on NASA-mandated milestones, and offered a more vague (but still ambitious) goal (again, emphasis mine) that “we want to put boots on the ground, boots on the moon, in 2028.”

None of that will happen unless SpaceX can prove that Starship can fly without failing and, crucially, become fully reusable. A huge part of making it reusable is the heat shield that keeps the Starship upper stage from exploding when it re-enters Earth’s atmosphere. The company saw the best results from its improved heat shield on the most recent Starship test flight, which splashed down in the Indian Ocean last month and is still intact. But before the rocket stage had even been recovered, Musk was willing to claim on Tuesday that he’d “consider the heat shield problem solved at this point.”

Musk has made many wild promises about SpaceX that never came true, such as when he said in 2016 that he’d put humans on Mars in six years. The difference now is that SpaceX is a public company, and ostensibly subject to regulation and fines if the company and its executives make promises they know can’t be met.

Of course, the Securities and Exchange Commission has pulled way back on corporate enforcement, especially against public companies. The Department of Justice is doing the same. And if SpaceX can’t follow through on Musk’s wide-eyed claims, investors won’t even be able to do much in civil court — because the company has all but inoculated itself against those kinds of lawsuits by incorporating in Texas.

Media reports have placed the fund behind Germany’s The Exploration Company and the Paris-based Mistral AI.

The Europe Union is ready to back deep-tech businesses in the bloc with a new funding mechanism aimed at helping the region catch up with the US and China.

The European Commission said on Tuesday (4 August) that its $5bn Scaleup Europe Fund has passed final legal hurdles and is ready to start operating at full capacity.

The multibillion-euro fund was initially announced back in October 2025, and is designed to build on the ‘choose Europe to start and scale’ strategy launched earlier that year.

It’s the largest fund of its kind ever launched in Europe and is expected to direct growth capital at high-potential companies across a range of strategic sectors, including artificial intelligence, quantum computing, clean energy, space technology, biotech and medical innovation. With an initial goal of €5bn, the Commission hopes to eventually raise €25bn for the scale-up fund.

The bloc tapped Stockholm-headquartered EQT to manage the fund earlier this year. The EU’s investment into the fund is backed by Horizon Europe, which is the world’s largest research and development programme and headed by the EU with more than €90bn in allocated funding up to 2027.

Recent reports have already placed Scaleup Europe Fund behind the Bavarian space-tech The Exploration Company in an upcoming $300m funding round. It is also reportedly expected to back the Paris-based Mistral AI in its upcoming raise.

The fund emerges as EU-based companies face “significant challenges”, its operators said, in accessing the necessary capital for growth from within the bloc, often leading them to search for funding elsewhere or move bases away from Europe altogether.

In July 2025, an Irish Government report found that scale-up enterprises in the country face a €1.1bn gap in equity financing over the next three to five years.

“When Europe invests in its innovators, Europe invests in its future,” said European Commission president Ursula von der Leyen.

“This is the goal of our Scaleup Europe Fund. From today, it will ensure our scale-ups can find what they need right here in Europe to grow into world-leading companies, to turn European innovation into our competitive edge.”

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.

ai and ml

New plugins for K-12 teachers, university instructors, and college kids? Read the room, guys

In the face of an epidemic of AI-enabled cheating and research suggesting its products hamper learning, OpenAI is doing the sensible thing and pushing more AI on students and teachers. 

Wait, did I say sensible? My mistake. 

The House of Altman announced a trio of new education-focused offerings on Tuesday: one for K-12 teachers, another for college educators, and a third for college students. The new plugins, the company explained, will help students and educators make more use of ChatGPT’s agentic capabilities for both studying and teaching. The new features are available through ChatGPT Edu, an institutionally licensed suite for higher education, and ChatGPT for Teachers, a free resource available to verified US K–12 educators and school districts.

The new offerings, says OpenAI, build on its educational AI philosophy that “AI should support learning, not shortcut it, and the best learning experiences keep educators and students in control.”

Plenty of educators might disagree. Cheating with AI has become a sad norm in schools around the world, and the US is no exception. Many young people admit to using AI to cheat on school assignments, and college students have been caught doing it, too.

Mexico’s largest university, the National Autonomous University of Mexico, last week suspended enrollment for incoming students after suspected widespread cheating in its admissions exams. The university also decided this week to require about 58,000 applicants who qualified in the remote entrance exam to take an in-person proctored control exam after scores surged during its first remotely administered admissions test.

As for how teachers relate to AI in the classroom, a 2025 study from the Center for Democracy and Technology suggests it’s going to take more than some new OpenAI software to help teachers feel more comfortable. 

CDT said last year that K-12 teachers widely complained of not being given the necessary resources to understand how to integrate AI into schools, or how to tackle the potential harms AI may be causing to their students’ ability to learn and personal development alike.

Rather than helping teachers and professors regain control of the academic process or better understand how AI is affecting their students, however, OpenAI wants to simply let them foist off more of their work to ChatGPT, too. 

“The K–12 Educator plugin is designed to help teachers plan and create for their classrooms,” OpenAI explained. The tool can use materials teachers already have on hand “to create differentiated resources, design interactive visuals, and surface actionable insights.” 

Integration with educational AI infrastructure outfit Learning Commons allows ChatGPT to automatically consider local educational standards for anything it creates for teachers, the company explained, while educators get to remain “in control of pedagogical decisions, grading, and agentic actions.” 

The plugin for college professors is largely the same.

“The College Educator plugin enables course design, teaching, and academic planning,” OpenAI said in its announcement. “Faculty can update syllabi, create interactive websites or multimedia assessments, adapt materials for diverse learners, or package content for their Learning Management System.” 

For college students, the plugin is designed to bring ChatGPT further into students’ study routines, the latest expansion of AI tools that have already complicated essay writing and take-home testing.

“The College Student plugin helps students turn what they are already studying into more personalized learning experiences,” said the ChatGPT maker. AI tutors are available (sorry, grad students), as are AI-generated quizzes, study guides, flashcards, and AI-generated visual explanations. 

“The plugin draws on learning science to prioritize deeper understanding and build stronger study habits,” OpenAI claims. You’d be right to doubt that. Aside from being used to cheat on exams, AI also has a demonstrable effect on harming learning outcomes. 

MIT researchers last year hooked a group of students up to EEGs to study their brain activity while writing essays. Some of them were allowed to use the internet to conduct research, while others were told to use AI. Not only did the AI cohort have far less ability to recall what they wrote when asked about it later, but they also showed far less brain activity, suggesting that relying on AI simply doesn’t lead to better learning. 

It’s not much of a leap to go from AI lowering brain activity while doing essay research to figuring that relying on AI tools to help one study, create flashcards, or receive tutoring will likely lead to similar outcomes. Real learning still takes legwork, and other studies suggest overreliance on tools like ChatGPT can undermine deeper learning.

Why teachers and professors would want to embrace an expansion of AI into their workflow when they’ve seen what it can do to their students is a question for OpenAI. The AI giant didn’t initially respond to our questions for this story, but we’ll update if that changes. ® 

What just happened? Texas has put a hold on new data center approvals that require access to the state’s grid, a move that could slow a long list of projects already waiting in line. Governor Greg Abbott has ordered the Public Utility Commission of Texas and ERCOT to pause approvals until the agencies complete an audit of applicants, a move that could stall a large backlog of projects seeking to connect.

The state wants more information before granting access. Abbott said applicants must provide tax break details, power use and generation data, water use and cooling plans, steps to reduce community impact, and facility ownership information. Any project missing that information will not be allowed onto the grid.

“Our top priority is to protect Texans’ safety and quality of life,” Abbott wrote. “Any project that fails to comply with the requirements set forth by the PUCT (Public Utility Commission of Texas) and ERCOT (Electric Reliability Council of Texas), and by state law, must be denied connection to the Texas grid. Simply put, Texans must come first.”

The review appears to follow a PUCT survey that most companies ignored. The agency said it asked 377 data center firms for information on water and power use, but only 28 responded.

The order isn’t a statewide ban, but it still affects a major chunk of the projects in ERCOT’s queue. More than 1,800 projects are waiting for approval to connect, and about 90% of them are data centers. Together, those applications could require 474 gigawatts, which is far above ERCOT’s current peak demand.

Even so, the directive does not hit every project in Texas. Data centers outside ERCOT’s territory, including those in El Paso, are not covered. Projects that bring their own power are exempt too, since they don’t need to connect to the grid.

The move drew criticism from opponents pushing for the legislature to step in with stricter rules. Texas Agriculture Commissioner Sid Miller called the directive “all hat and no cattle” on the Texas Department of Agriculture’s official Facebook page, saying, “Texans need laws with teeth, not another press release designed to make it look like something is being done while hyperscale projects continue moving forward.”

State Rep. Gina Hinojosa, who is challenging Abbott in the 2026 governor’s race, dismissed the move as too little, too late. She said Abbott has spent months promoting Texas as a friendly place for data centers, and argued his new pause wouldn’t convince anyone the state is suddenly cracking down.

The Texas move is narrower than New York’s statewide data center moratorium, but it still adds a major hurdle for developers. For an industry built around fast buildouts and heavy power demand, the pause could have immediate effects.

As strong and light as carbon fiber-epoxy composites are, the same can’t always be said of carbon-fiber reinforced 3D printer filaments. Of those that do improve over stock filament, the best performance comes from long, continuous strands, but the printers that can embed these are quite expensive. [MagicLAG], looking for a cheaper method, made something even stronger: prints reinforced with subsurface carbon-fiber cloth.

They tried a few other methods first, including pausing the print and manually embedding carbon fiber strands, ironing strands into the finished part, and ironing carbon fiber cloth into the bottom layer. For the main method, though, he printed the test part in three pieces: a core part, and two outer shell layers. Between the core and the shell is a small gap, into which carbon-fiber cloth can be epoxied. Under good conditions (not using quick-setting epoxy), this mostly preserves the outer surface and dimensional accuracy.

To test the various strengthening methods, [MagicLAG] printed hooks and tensioned them on a load cell until failure. None of the methods using single-stranded fiber showed any improvement; the fiber simply bent and let the surrounding plastic break. As a control for the epidermal cloth parts, they printed shells and cores and epoxied them together. These controls performed better than the standard parts, but not nearly as well as the carbon-fiber cloth composites. With only a few layers of cloth, these more than tripled the yield strength of the basic hook.

If you’d rather use a carbon-fiber filament, the type of plastic matters; carbon fiber makes PLA, at least, weaker. Regardless of form, some caution is called for whenever handling carbon fiber, since it seems to show some asbestos-like effects.

Google has finally addressed one of the biggest frustrations for Fitbit users on iPhone.

The latest Google Health 5.05 update allows health and fitness data collected by Fitbit devices to sync directly with Apple’s Health app. This eliminates a long-standing limitation that previously only allowed information to flow in one direction.

Until now, iPhone users could import data from Apple Health, HealthKit and Apple Watch into Google Health. However, they couldn’t send Fitbit-recorded information back to Apple’s platform without relying on third-party apps, a restriction that dates back more than a decade. This happened after Fitbit chose not to support Apple’s HealthKit framework when it launched in 2014.

With version 5.05, supported Fitbit data including workouts, step counts, sleep tracking and other health metrics can now be shared with Apple Health automatically. The feature is available through the Partner apps section within the Google Health app. Here users can connect Apple Health and decide exactly which categories of data they want to sync.

Rent the Runway’s Mark Walsh discusses the tech ecosystem in Galway and how he found himself in the role of director of engineering.

It was “half accidental and half strategic, with the accidental being the bigger half, as my kids would say”, joked Mark Walsh, when asked what led him to become the director of engineering at Rent the Runway’s Galway location.  

“On the accidental side, I began my tech career in a start-up in Spiddal in Galway, and spent a decade working as a software engineer,” he said. “Then came the advent of DevOps and a wave of emerging technologies beyond Java, which pulled me into different areas, cloud, solution architecture and even product management-type roles. 

“People management arrived completely by accident about six years ago, not as any great career strategy, and it has evolved from there into what I do today.”

Prior to joining the organisation he was commuting to a different role, four hours there and back, which he explained took a toll on his family life. Moreover, he had always planned on returning to Galway.

“I wanted to return to Galway. Crucially though, I was focused on a destination, not just another job. Rent the Runway was somewhere I already knew through former colleagues, it had the kind of opportunities that genuinely motivated me and it felt like somewhere I could add something meaningful rather than just fill a seat.”

What is the current engineering landscape like in Ireland?

I have to be honest and say I can’t speak adequately for the tech landscape much beyond Galway, as that would be overstepping what I actually see day to day. What I will say is that from a Galway perspective, the picture is more positive than the broader headlines might suggest.

There is a real depth of engineering talent here, and a sense that you don’t need to be in the capital city to work on interesting, complex problems at scale. The mix of homegrown companies maturing alongside international organisations that have put down genuine roots has created a stronger local ecosystem than people sometimes give it credit for. It is not without its challenges, but the opportunity feels real.

What does a typical day look like, if there is such a thing?

There genuinely isn’t one, which I think is part of what keeps it interesting. As a US company, our natural overlap with our US team members falls in the afternoon Irish time, so that tends to be where the collaborative work lands: cross-team meetings, planning sessions, working through anything that needs real-time conversation across time zones. Mornings are generally freer, which I try to protect for focused work, Galway-based one-to-ones, or anything that benefits from uninterrupted thinking time.

The nature of the work itself shifts depending on where we are in the cycle. We run quarterly planning and within that there are shorter-term operational necessities running alongside medium-term delivery work and longer horizon strategic efforts. As a platform team with an SRE mindset, that means BAU maintenance such as managing upgrades across our existing portfolio, future-looking initiatives addressing our strategic vision such as platform migrations and responding to emergent support or security issues as they arise.

What is the current focus of your work and its potential long-term impact?

The intent is always the long-term health of Rent the Runway as a platform and a business. In practical terms that means a few things running in parallel: reducing operational overhead and complexity where it has built up over time, managing risk and compliance thoughtfully and continuing to advance our existing capabilities while staying open to where new opportunities might emerge.

What I find genuinely motivating about engineering at this level is that the decisions you make about architecture, tooling and how teams are structured have a compounding effect. Good calls made now create headroom for the business to move faster later. Poor ones create drag that someone, usually a future version of your team, has to unpick.

So while the immediate focus is always on what needs to happen this quarter, the framing I try to bring to it is, what are we making easier or harder for ourselves two or three years from now?

How does the field of research impact your role at Rent the Runway?

When I hear the word research I tend to map it back to engineering practice as a whole. My teams are engaged in continuous research, whether we label it that way or not, because the technology landscape never stops moving.

The old saying that today’s problem was yesterday’s solution is as true now as it has ever been, arguably more so. My teams are always engaged in some form of discovery work: reducing technical uncertainty, evaluating emerging approaches, making sure that what we build is robust, secure and fit for purpose beyond the immediate delivery. That process of deliberate inquiry is baked into how our engineering teams operate, even when it doesn’t carry the formal title of research.

Have you any advice for professionals considering a research-oriented role in the engineering space?

My advice is, even if it sounds old-fashioned, given the pace of change right now, don’t abdicate your thinking completely to your tooling.

The engineers and knowledge workers I see navigating uncertainty well are those grounded in the fundamentals, reasoning through complex problems, communicating clearly about risk and sitting with ambiguity rather than rushing to the first solution. Those qualities are always in demand, at least with me.

Staying current matters and I am not suggesting otherwise. But understanding why you are reaching for a particular approach, rather than just something that appears to work, has always been important. Value your judgement and decision-making rigour first, the tools will follow.

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.