Connect with us

Tech

OpenAI inks huge lease in Bellevue, doubling down on Seattle region near Microsoft and Amazon HQs

Published

on

(GeekWire File Photo / Todd Bishop)

OpenAI is placing a bigger bet on the Seattle region, signing a massive new lease in Bellevue as the ChatGPT-maker expands near the headquarters of two key corporate cloud partners.

The company is taking an additional ten floors at City Center Plaza in downtown Bellevue, boosting its footprint to 296,000 square feet, according to sources familiar with the matter. OpenAI previously occupied two floors in the building. It’s one of the largest AI company leases in the region.

The San Francisco-based company now has room for more than 1,000 employees at the office, based on typical commercial real estate standards. OpenAI, which arrived in Bellevue in 2024, currently employs more than 300 people in the Seattle area, according to LinkedIn data — up from around 169 in September.

The office gives OpenAI a large hub just a short drive from Microsoft’s Redmond headquarters campus and within a few blocks of Amazon’s expanding Bellevue towers, tightening its ties with both cloud giants.

Microsoft has invested billions of dollars in OpenAI and serves as its primary strategic partner, providing the Azure cloud infrastructure that underpins many of OpenAI’s models and products.

Advertisement

At the same time, OpenAI has deepened its relationship with Amazon, inking a $38 billion cloud deal in November. Amazon is reportedly in talks to invest up to $50 billion in OpenAI as part of a new investment round.

Microsoft previously occupied a majority of the 26-floor City Center Plaza building but said in 2023 that it would not renew its lease. The building is adjacent to a light rail station that will offer transit connection to Seattle starting in March.

CoStar reported on OpenAI’s expansion earlier this week.

OpenAI recently acquired Seattle startup Statsig for $1 billion. The company is also reportedly gearing up for an IPO.

Advertisement

The Wall Street Journal reported that OpenAI is paying employees “more than any tech startup in recent history,” with the average stock-based compensation set around $1.5 million per person at the company, which has around 4,000 employees.

OpenAI CEO Sam Altman and Microsoft CTO Kevin Scott at Microsoft Build in 2024. (GeekWire File Photo / Todd Bishop)

OpenAI now has one of the largest offices among out-of-town tech companies with satellite engineering centers across the Seattle region. Meta, Google, Apple, and other Silicon Valley giants have substantial footprints in the area, which boasts one of the world’s top technical talent pools. OpenAI rival Anthropic opened an office in Seattle two years ago and is hiring.

Seattle has the most AI engineers in the U.S. behind Silicon Valley, according to a 2024 report from venture capital firm SignalFire.

OpenAI’s new lease also reflects a growing role for the Eastside in the AI boom. Many technology companies have signed new or expanded leases in and around Bellevue recently, including Snap, Anduril, Shopify, Snowflake, Walmart, and Chewy. Uber and Databricks are filling the city’s newest office tower, according to the Puget Sound Business Journal.

Vacancy rates still remain high in downtown Bellevue, reaching 25.4% at the end of last year, according to Broderick Group.

Advertisement

That’s still not as high as downtown Seattle, where vacancy rates hit a record high in Q4, up to 34.7%, according to CBRE.

“Notably, a growing number of new-to-market entrants … are choosing the Eastside over Seattle, drawn by Bellevue’s modern office inventory, business friendly climate and skilled technology workforce,” Broderick noted in its Q4 report.

Source link

Advertisement
Continue Reading
Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

Tech

The best wireless headphones of 2026

Published

on

The Bowers & Wilkins Px7 S3 replace the excellent Px7 S2e, raising the bar for wireless headphone audio.

Despite looking relatively the same as the older models, B&W have given it a significant overhaul. The headband has been redesigned to fit a wider range of heads, the controls reshaped to be easier to find and use, while the headphones are slimmer for a more attractive profile.

The only issue we have is with the controls, which we didn’t feel as if they needed to be changed but they work fine enough.

These headphones feature noise cancelling and a transparency mode and despite Bower’s claims of improving both areas, the noise cancelling isn’t as strong as the Bose QuietComfort Ultra Headphones or Sony WH-1000XM6. The transparency mode could be clearer too. ANC is not these headphones’ strongest point.

Advertisement

The Bowers & Wilkins Music app offers the means to customise bass and treble, as well as a custom EQ  option to create your own sound profile, a first for a pair of Bowers wireless headphones.

These headphones keep the feature set relatively simple, and aren’t as ‘smart’ or as feature-laded as the less expensive Sony WH-1000XM5 but the app does have built-in streaming support for services such as QobuzDeezer, and Tidal.

The battery life remains 30 hours of listening from one charge, though in our tests we found it could go longer with an Android smartphone.

Bluetooth support includes aptX Lossless, the one of the higher quality wireless codecs, and as usual the wireless connection is excellent.

Advertisement

These are also one of the best headphones for call quality, offering great clarity and detail while keeping background sounds to a minimum.

The sound quality here is the best it’s been for the Px7 range. It’s energetic, clear, expressive and natural in how it sounds, the headphones’ levels of detail, dynamism and sense of spaciousness make it one of the best-sounding models on the market.

Low frequencies have more depth and power, the midrange is detailed and the high frequencies clear. If you’re after a pair of wireless headphones for the sound, there’s none better at this price than the Px7 S3.

Advertisement

Source link

Continue Reading

Tech

Rethinking What We Choose to Measure in Schools

Published

on

This story was published by a Voices of Change fellow. Learn more about the fellowship here.

Sitting in a recent district administrator meeting, I found myself excited about a new student data platform my district is rolling out. This new tool, called by a catchy acronym and presented on a flashy dashboard, would collect a variety of information about student skills, mindsets and achievement. It would let us break down information by subgroup and assign overall scores to students, helping us identify who needs additional support.

Initially, I was enthusiastic about how it could empower teachers to better understand students and improve outcomes. But since then, after conversations with the teachers in my building and reflecting on my own experiences using data in the classroom, I’ve begun to wonder whether we are focusing on the wrong data or placing too much emphasis on data overall.

I love looking at data. I’m excited when data surprises me or shows me something more clearly. It’s motivating to see trend lines sloping upward and green arrows pointing toward the sky. Data can help us see the bigger picture when looking at larger systems. We can see which schools are suspending too many students of color and which districts are improving reading scores. As an administrator, I find this illuminating and helpful in guiding how schools make decisions.

Advertisement

But as data trickles down to classrooms and individual students, the usefulness and impact get murkier. In the Montessori school where I teach, where our focus is guiding the child according to their interests and readiness, the data we have to collect affects what we focus on, often in unexpected ways, and sometimes to the detriment of the system itself.

Teaching to the Test

My school is a successful one, and looking at our annual school report card should be a source of pride for the teachers. The report card is based primarily on our state test scores in math and reading, and various calculations are made from our students’ performance on it. But when we shared the most recent report card that showed our school once again exceeded expectations, the results were met with shrugs and muted applause. It isn’t that they aren’t proud of what our students can do; they just recognize the narrowness of the data and how indirectly it connects to what is happening in their Montessori classrooms.

When I pointed out that our report card showed math achievement was an area for improvement, the response was, “Are you saying we should teach to the test?” They know that we could game the system by focusing on test prep and the specific questions their students might encounter. Because we follow a Montessori curriculum with three grade levels in our classrooms, our sequence doesn’t always align with grade-level standards, which can show up on tests, with students scoring poorly on topics they haven’t been introduced to yet. We could align our curriculum with the test and focus our teaching on what the test assesses, but doing so goes against our philosophy of allowing students to make choices about their learning at their own pace.

With this tension in mind, I wonder if data distorts the focus of education? Our current focus on reading and math scores, based on standardized testing, is part of what we want our schools to do. But teachers know that students are capable of achieving much more than our report cards show. Is there some golden indicator that we just haven’t found yet — a measurement like happiness or flourishing — that would be more meaningful? And of course, if we find it, won’t it also become distorted?

Advertisement

Information Overload

There is also a heavy focus in our district on using data to determine which students qualify for additional support through differentiation, interventions and individualized instruction. Administration requires us to hold monthly meetings to review student data and determine who is progressing and who might need more support. On one level, this seems like a great practice for identifying who needs help, but in reality, the system’s capacity to act on that information is overstretched, leading to distortion and ultimately to burnout.

I remember my frustrations as a teacher in these meetings. The data was interesting and could help you to confirm or question ideas you had about students based on your classroom observations. But it didn’t often provide helpful information for supporting students. The time spent in these meetings outweighed the benefit I got from them, and took away from the little time I had to prepare and plan for my students.

Teachers I work with have regularly expressed feeling overwhelmed by the amount of information they need to consider and the testing required to gather it. In our early grades, due to a new state law mandating early literacy assessments, students are tested monthly on letter-sound identification and oral reading fluency. This generates an unending stream of data to grapple with and a constant feeling of needing to do more to address it, all of which adds to stress on teachers, students and the system. I’ve seen amazing teachers, skilled at connecting with kids and providing rich learning experiences, brought to tears because there was too much red on a data spreadsheet.

Teachers don’t have the time to assess and examine all the data they’re now expected to, and monthly checks of early reading indicators take time away from actually teaching those skills. Being responsive to the data you gather means stopping what you’re doing and finding new ways to help kids learn what the data says they need. Teachers are expected to find new resources and determine when and how to work with small groups that need similar support, while also providing meaningful learning opportunities for other students. And, of course, different kids need different things, so you’d need to do this for multiple groups, which is unrealistic to expect all teachers to have the capacity to do.

Advertisement

Meaningful Measurement

Schools, as they are currently designed, weren’t supposed to be responsive to the amount of data we’re collecting. They were designed to teach a group of students a set of information in a specific sequence each year, and then grade them on how well they learned what they were expected to learn. They were designed to tell us which students could meet the standards, and who couldn’t, not to ensure that each child could learn and flourish.

When I was a classroom teacher, I kept track of how many books my students read each month. It wasn’t research-backed or scientifically valid, but I found the data helpful for identifying who was and wasn’t reading, and thinking about how I could support them. In some cases, it helped me direct kids to books that they might get excited about; in other cases, it just let me know that a particular kid wasn’t that into reading, and that that might have to be OK for now. The data wasn’t complicated, but it let me quantify what I was observing in my classroom in a way that was meaningful to me and, most importantly, helped me connect with my students as whole people.

A key component of Montessori philosophy is the teacher as observer — watching and documenting what students choose and do to understand and assess what they are ready for. Every teacher should have the time and space to measure and track what feels meaningful and helpful to them.

This may look different for every teacher, but the important factor is that it has meaning to them and is connected to their students and their practice. Likewise, we need to remember that standardizing the expectations for students goes against what we know about how people develop. There’s always going to be variation in a dataset — there’s no metric on which we are all the same.

Advertisement

As an administrator, my responsibility is to understand and use data in ways that are helpful, while also protecting teachers and students from distractions and distortions that undermine the larger goals of creating opportunities for growth and learning for all students.

Ultimately, data should serve as a guide rather than a governor, informing our decisions without eclipsing the human elements of teaching and learning. If we can strike that balance, we can create systems that honor both the complexity of children and the professional wisdom of the educators who know them best.

Source link

Advertisement
Continue Reading

Tech

Memory Prices Have Nearly Doubled Since Last Quarter

Published

on

Memory prices across DRAM, NAND and HBM have surged 80 to 90% quarter-over-quarter in Q1 2026, according to Counterpoint Research’s latest Memory Price Tracker. The price of a 64GB RDIMM has jumped from a Q4 2025 contract price of $450 to over $900, and Counterpoint expects it to cross $1,000 in Q2.

NAND, relatively stable last quarter, is tracking a parallel increase. Device makers are cutting DRAM content per device, swapping TLC SSDs for cheaper QLC alternatives, and shifting orders from the now-scarce LPDDR4 to LPDDR5 as new entry-level chipsets support the newer standard. DRAM operating margins hit the 60% range in Q4 2025 — the first time conventional DRAM margins surpassed HBM — and Q1 2026 is on track to set all-time highs.

Source link

Continue Reading

Tech

Kembara Fund 1 targets funding cliff for Europe’s deep tech, climate start-ups

Published

on

With an ultimate target of €1bn, Spain’s Mundi Ventures closed on €750m this week for its Kembara Fund for deep tech and climate start-ups.

Deep tech targets the world’s biggest problems, from climate and energy to defence and healthcare. Europe has the talent and the start-ups, but has struggled on the capital side for scaling start-ups. This is what Mundi Ventures’ Kembara Fund aims to address with its focus on Series B and C funding of €15m-€40m, and beyond, for companies based in EU member states.

According to the Kembara team, Europe produces 28pc of global deep tech innovation, but only 3pc of European deep tech companies successfully raise Series B or C rounds. It is that very gap that the Kembara Fund is hoping to bridge using “€1bn dedicated to backing Europe’s deep tech champions at the exact moment when technology is proven and global scale becomes possible”.

In his own words, Kembara partner Yann de Vries says the funding cliff is “deeply personal” to him after his experience with Lilium, which declared insolvency in 2025 having failed to raise adequate investment. Its patents now belong to Archer Aviation.

Advertisement

“A decade ago at Lilium – a leading electric aviation company – we went from a sci-fi idea in a hangar to a NASDAQ-listed company in five years,” de Vries said in a LinkedIn post yesterday.

“I saw first-hand how brutally hard it is for European deep tech teams to raise €50m-€100m rounds and scale globally. That journey is why we built Kembara.”

The European Investment Fund (EIF) is a lead backer of Kembara, announcing in July last year that it would invest €350m in Kembara Fund 1. At the time, the EIC said it was the experience of the Kembara management team and its “differentiated strategy” that were key to receiving the support of the EIF.

“Companies that achieve strategic autonomy in critical technologies – from AI and quantum computing to space systems and clean energy – have the potential to become trillion-dollar global champions,” said de Vries in his post.

Advertisement

“Our ambition is clear: fix Europe’s growth-stage funding gap, where only 3pc of deep tech companies make it to Series B/C today.”

Other Kembara partners include Javier Santiso, Robert Trezona, Pierre Festal and Siraj Khaliq, who de Vries says have a combined experience of 100 years in deep tech, in companies like SpaceX, Palantir, PsiQuantum, OpenAI, Lilium, Ceres Power, Anduril and The Exploration Company.

“We are entrepreneurs united by a shared mission: to build Europe’s leading deep tech platform – one that keeps Europe competitive in the global technology race, tackles the world’s most pressing challenges and delivers outsized returns,” said de Vries. “This is only the beginning…”

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.

Advertisement

Source link

Continue Reading

Tech

Hollywood’s AI Bet Isn’t Paying Off

Published

on

Hollywood’s recent attempts to build entertainment around AI have consistently underperformed or outright flopped, whether the AI in question is a plot device or a production tool. The horror sequel M3GAN 2.0, Mission: Impossible — The Final Reckoning, and Disney’s Tron: Ares all disappointed at the box office in 2025 despite centering their narratives on AI.

The latest casualty is Mercy, a January 2026 crime thriller in which Chris Pratt faces an AI judge bot played by Rebecca Ferguson; one reviewer has already called it “the worst movie of 2026,” and its ticket sales have been mediocre. AI-generated content hasn’t fared any better. Darren Aronofsky executive-produced On This Day…1776, a YouTube web series that uses Google DeepMind video generation alongside real voice actors to dramatize the American Revolution. Viewer response has been brutal — commenters mocked the uncanny faces and the fact that DeepMind rendered “America” as “Aamereedd.”

A Taika Waititi-directed Xfinity commercial set to air during this weekend’s Super Bowl, which de-ages Jurassic Park stars Sam Neill, Laura Dern and Jeff Goldblum, has already been mocked for producing what one viewer called “melting wax figures.”

Source link

Advertisement
Continue Reading

Tech

Capitol Hill probe spotlights claims Apple and Google were pushed to block ICE-monitoring apps

Published

on

The removal of an ICE-monitoring app almost a year ago has triggered new questions about whether the US Department of Justice crossed a constitutional line in its dealings with Apple and Google.

Phone screen displaying an app store page for ICEBlock with an ice cube icon, a 3.9-star rating, and age recommendation of 9+.
How ICEBlock appeared in the App Store before being pulled

On Friday, House Judiciary Committee ranking member Jamie Raskin vowed to investigate the Department of Justice over allegations that it pressured tech giants into removing ICE tracking apps.
In a letter addressed to Attorney General Pam Bondi, Raskin asks, “Why is the Department of Justice (DOJ) violating the First Amendment by coercing big tech to block access to lawful apps that the American people use to record, report, and monitor the actions of our own government officers?”
Continue Reading on AppleInsider | Discuss on our Forums

Source link

Continue Reading

Tech

The “micro” build: why your next PC should fit in a shoebox

Published

on

For decades, PC gaming meant owning a monolith: a massive, flashing tower that dominated your floor space. But in 2026, the era of the giant box is over. Components have become efficient enough that you no longer need 60 liters of air to cool them. The “Small Form Factor” (SFF) movement’s gone mainstream, proving that you can fit an RTX 5080 and a top-tier CPU into a case the size of a shoebox.

It’s minimal, it’s sophisticated, and it looks a lot better on a desk than a plastic tower. If you’re ready to downsize without downgrading performance, this is where you should start.

The quick list

The cases

Fractal Design — Terra Jade

The Terra changed the game by proving a PC could look like mid-century furniture. Featuring a genuine walnut wood front panel and anodized aluminum sheets, it’s designed to be seen. The “sandwich” layout puts the GPU on one side and the CPU on the other, allowing it to stay incredibly small (10.4 liters) while still fitting full-sized graphics cards.

Fractal Design — Era 2 Silver

Advertisement

While the Terra is rustic, the Era 2 is pure modern elegance. The sculpted silver aluminum exterior feels like high-end audio equipment. It’s optimized for airflow with a unique chimney design, pulling cool air from the bottom and exhausting it out the top. It’s the perfect housing for a professional creative workstation.

Lian Li — A4-H2O

This is the reference standard for water-cooled SFF builds. Collaborating with DAN Cases, Lian Li created a sub-11 liter case that somehow fits a 240mm AIO liquid cooler. It’s an industrial, no-nonsense aluminum box that maximizes every millimeter of internal space. If you want the smallest possible footprint with liquid cooling, this is it.

Cooler Master — NR200P MAX V2

Building in a small case can be intimidating. Cooler Master solves this with the MAX V2. It comes pre-installed with a custom 280mm AIO cooler and an 850W Gold Power supply, with the cables already routed and managed. You just drop in your motherboard and GPU, and you are done. It is designed to handle next-gen power, officially ready for cards like the RTX 5080. It is also 12% off right now.

Power & cooling

ASUS — ROG Loki SFX-L 850W Platinum

Small builds used to mean low power, but not anymore. The Loki pushes 850W of Platinum-rated efficiency, enough to drive top-tier silicon. It uses the slightly longer “SFX-L” standard to fit a larger 120mm fan, making it quieter than standard small power supplies. Plus, it includes an RGB fan if you want a subtle glow.

CORSAIR — SF750 (2024)

Advertisement

Ask any SFF builder what PSU to buy, and they’ll say “Corsair SF750.” The 2024 refresh brings ATX 3.1 compliance and native PCIe 5.1 cables for modern GPUs. It is incredibly dense, reliable, and features a zero-RPM mode so the fan doesn’t even spin during light work. You can grab it now for 20% off.

SCYTHE — Big Shuriken 4

In cases like the Fractal Terra, you can’t fit a big liquid cooler. You need high-performance air cooling that stays low. The Big Shuriken 4 is designed exactly for this. At just 67mm tall, it fits where standard coolers can’t, yet it can handle up to 200W of heat thanks to its dense fin stack and high-static pressure fan.

The bottom line

If you want a PC that doubles as home decor, the Fractal Design Terra is the clear winner. For first-time builders who don’t want to stress about cable management or part compatibility, the Cooler Master NR200P MAX V2 is a cheat code that saves hours of frustration. But if you need absolute maximum cooling for high-end components in the smallest possible footprint, the Lian Li A4-H2O remains the gold standard.

Source link

Advertisement
Continue Reading

Tech

How to watch the Opening Ceremony at the 2026 Milan Cortina Winter Olympics rebroadcast tonight

Published

on

The 2026 Milan Cortina Winter Olympics Opening Ceremony has concluded. The festivities featured performances from Mariah Carey and Andrea Bocelli, 3,000 athletes walking in the Parade of Nations, and not one but two Olympic cauldrons being lit. (One at Milan’s Arco della Pace, since Milan is serving as the main hub for this year’s Games, and the other in the Alpine city of Cortina d’Ampezzo, where events like skiing are taking place.) If you missed out on watching live, the Opening Ceremony will re-air in primetime tonight on NBC. Here’s what you need to know.

How to watch the Opening Ceremony at the 2026 Winter Olympics

Image for the mini product module
Image for the mini product module

Date: Friday, Feb. 6

Time: primetime re-air from 8-11 PM ET

Location: San Siro Stadium, Milan

TV channels: NBC

Advertisement

Streaming: Peacock, DirecTV, NBC.com, and more

Where can I stream the Opening Ceremony at the 2026 Winter Olympics?

Image for the small product module

For $11/month, an ad-supported Peacock subscription lets you stream live sports and events airing on NBC, including the 2026 Winter Olympics, Super Bowl LX, and more. Plus, you’ll get access to thousands of hours of shows and movies, including beloved sitcoms such as Parks and Recreation and The Office, every Bravo show and much more.

For $17 monthly you can upgrade to an ad-free subscription which includes live access to your local NBC affiliate (not just during designated sports and events) and the ability to download select titles to watch offline.

How to watch the 2026 Opening Ceremony on TV:

There will be two broadcasts of the 2026 Milan Cortina Winter Olympics Opening Ceremony this Friday. You can tune in live from 2PM – 5PM ET on Friday afternoon, or catch the encore broadcast from 8PM – 11PM ET that night. Both broadcasts will air on NBC, which is available with DirecTV, Hulu + Live TV, and more.

Advertisement

With a live TV streaming service subscription or cable package, you can also catch all of NBC and Peacock’s Olympics coverage on NBC.com and via NBCOlympics.com or the NBC App, just by logging in with your provider.

Image for the small product module

DirecTV’s Entertainment tier gets you access to loads of channels where you can tune in to college and pro sports, the Winter Olympics, and more. Channels include ESPN, TNT, ACC Network, Big Ten Network, CBS Sports Network, and, depending on where you live, local affiliates for ABC, CBS, Fox and NBC.

Whichever package you choose, you’ll get unlimited Cloud DVR storage and access to ESPN+’s new streaming tier, ESPN Unlimited. 

DirecTV’s Entertainment tier package is $49.99 for your first month. But you can currently try all this out for free for 5 days. If you’re interested in trying out a live-TV streaming service for football season but aren’t ready to commit, we recommend starting with DirecTV. 

Advertisement
Image for the mini product module

How to watch the Opening Ceremony in Milan without cable:

You can watch the Opening Ceremony live or on-demand on Peacock. If you already subscribe to a live TV streaming service or cable package, you should also be able to catch all of Peacock’s Olympics coverage on NBC.com, NBCOlympics.com and the NBC app.

Image for the small product module

For $11/month, an ad-supported Peacock subscription lets you stream live sports and events airing on NBC, including the 2026 Winter Olympics, Super Bowl LX and more. Plus, you’ll get access to thousands of hours of shows and movies, including beloved sitcoms such as Parks and Recreation and The Office, every Bravo show and much more.

For $17 monthly you can upgrade to an ad-free subscription which includes live access to your local NBC affiliate (not just during designated sports and events) and the ability to download select titles to watch offline.

Who hosted the Winter Olympics Opening Ceremony?

Sportscaster Terry Gannon hosted coverage of the 2026 Winter Games Opening Ceremony. Former Olympic snowboarder Shaun White also made an appearance. NBC Olympics primetime host Mike Tirico also participated remotely from San Francisco, where he’s pulling double duty covering the Olympics and prepping to call Super Bowl LX.

Who performed at the 2026 Winter Olympics Opening Ceremony?

Mariah Carey, iconic Italian tenor Andrea Bocelli, actress Sabrina Impacciatore (The Paper, The White Lotus) and pianist Lang Lang performed at the Opening Ceremony.

Advertisement

Where is the 2026 Olympics Opening Ceremony being held?

The 2026 Winter Olympics Opening Ceremony was held at Milan’s San Siro Stadium, home to football clubs AC Milan and Inter Milan. The Opening Ceremony will actually be one of the final events held at San Siro Stadium, which is set to be demolished sometime after the Games end.

Source link

Continue Reading

Tech

Amazon’s Tax Bill Plunges 87% After Tax Cuts

Published

on

An anonymous reader shares a report: Republicans’ tax cuts shaved billions off Amazon’s tax bill, new government filings show. The company says it ran a $1.2 billion tax bill last year, down from $9 billion the previous year, and even as its profits jumped by 45% to nearly $90 billion.

That’s largely because of the generous new depreciation breaks GOP lawmakers included in their One Big Beautiful Bill, something that’s particularly important to Amazon which — in addition to maintaining a vast infrastructure for its ubiquitous delivery business — has been spending billions to build out artificial intelligence data centers.

Also helping, though less important: The law’s expanded breaks for businesses research and development expenses. The company has long been criticized by Democrats for paying little in tax, and it appeared to be bracing for criticism in the wake of the report to the Securities and Exchange Commission.

Source link

Advertisement
Continue Reading

Tech

Former Federal Judge: ICE’s Home Raiding Policy Violates A Basic Constitutional Right

Published

on

from the in-case-it-wasn’t-obvious dept

This article is republished from The Conversation under a Creative Commons license. Read the original article.

As Immigration and Customs Enforcement, or ICE, agents continued to use aggressive and sometimes violent methods to make arrests in its mass deportation campaign, including breaking down doors in Minneapolis homes, a bombshell report from the Associated Press on Jan. 21, 2026, said that an internal ICE memo – acquired via a whistleblower – asserted that immigration officers could enter a home without a judge’s warrant. That policy, the report said, constituted “a sharp reversal of longstanding guidance meant to respect constitutional limits on government searches.”

Those limits have long been found in the Fourth Amendment to the U.S. Constitution. The Conversation’s Politics editor Naomi Schalit interviewed Dickinson College President John E. Jones III, a former federal judge appointed by President George W. Bush and confirmed unanimously by the U.S. Senate in 2002, for a primer on the Fourth Amendment, and what the changes in the ICE memo mean.

Okay, I’m going to read the Fourth Amendment – and then you’re going to explain it to us, please! Here goes:

Advertisement

“The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized.” Can you help us understand what that means?

Since the beginning of the republic, it has been uncontested that in order to invade someone’s home, you need to have a warrant that was considered, and signed off on, by a judicial officer. This mandate is right within the Fourth Amendment; it is a core protection.

In addition to that, through jurisprudence that has evolved since the adoption of the Fourth Amendment, it is settled law that it applies to everyone. That would include noncitizens as well.

What I see in this directive that ICE put out, apparently quite some time ago and somewhat secretly, is something that, to my mind, turns the Fourth Amendment on its head.

Advertisement

What does the Fourth Amendment aim to protect someone from?

In the context of the ICE search, it means that a person’s home, as they say, really is their castle. Historically, it was meant to remedy something that was true in England, where the colonists came from, which was that the king or those empowered by the king could invade people’s homes at will. The Fourth Amendment was meant to establish a sort of zone of privacy for people, so that their papers, their property, their persons would be safe from intrusion without cause.

So it’s essentially a protection against abuse of the government’s power.

That’s precisely what it is.

Advertisement

Has the accepted interpretation of the Fourth Amendment changed over the centuries?

It hasn’t. But Fourth Amendment law has evolved because the framers, for example, didn’t envision that there would be cellphones. They couldn’t understand or anticipate that there would be things like cellphones and electronic surveillance. All those modalities have come into the sphere of Fourth Amendment protection. The law has evolved in a way that actually has made Fourth Amendment protections greater and more wide-ranging, simply because of technology and other developments such as the use of automobiles and other means of transportation. So there are greater protected zones of privacy than just a person’s home.

ICE says it only needs an administrative warrant, not a judicial warrant, to enter a home and arrest someone. Can you briefly describe the difference and what it means in this situation?

It’s absolutely central to the question here. In this context, an administrative warrant is nothing more than the folks at ICE headquarters writing something up and directing their agents to go arrest somebody. That’s all. It’s a piece of paper that says ‘We want you arrested because we said so.’ At bottom that’s what an administrative warrant is, and of course it hasn’t been approved by a judge.

Advertisement

This authorized use of administrative warrants to circumvent the Fourth Amendment flies in the face of their limited use prior to the ICE directive.

A judicially approved warrant, on the other hand, has by definition been reviewed by a judge. In this case, it would be either a U.S. magistrate judge or U.S. district judge. That means that it would have to be supported by probable cause to enter someone’s residence to arrest them.

So the key distinction is that there’s a neutral arbiter. In this case, a federal judge who evaluates whether or not there’s sufficient cause to – as is stated clearly in the Fourth Amendment – be empowered to enter someone’s home. An administrative warrant has no such protection. It is not much more than a piece of paper generated in a self-serving way by ICE, free of review to substantiate what is stated in it.

Have there been other kinds of situations, historically, where the government has successfully proposed working around the Fourth Amendment?

Advertisement

There are a few, such as consent searches and exigent circumstances where someone is in danger or evidence is about to be destroyed. But generally it’s really the opposite and cases point to greater protections. For example, in the 1960s the Supreme Court had to confront warrantless wiretapping; it was very difficult for judges in that age who were not tech-savvy to apply the Fourth Amendment to this technology, and they struggled to find a remedy when there was no actual intrusion into a structure. In the end, the court found that intrusion was not necessary and that people’s expectation of privacy included their phone conversations. This of course has been extended to various other means of technology including GPS tracking and cellphone use generally.

What’s the direction this could go in at this point?

What I fear here – and I think ICE probably knows this – is that more often than not, a person who may not have legal standing to be in the country, notwithstanding the fact that there was a Fourth Amendment violation by ICE, may ultimately be out of luck. You could say that the arrest was illegal, and you go back to square one, but at the same time you’ve apprehended the person. So I’m struggling to figure out how you remedy this.

Filed Under: 4th amendment, administrative warrants, ice, privacy, warrants

Advertisement

Source link

Continue Reading

Trending

Copyright © 2025