Category: 4. Technology

Fitness tracking company Polar has launched Loop, a $200 screenless wearable that it says will have no subscription fees. Preorders opened on Sept. 3, and the Polar Loop will start shipping on Sept. 10.

Like other fitness trackers, the Polar Loop will log steps, sleep patterns and daily activity patterns but Polar is touting the lack of a screen as “unobtrusive” and “discreet.” The Loop, which is a wearable band for your wrist, has eight days of battery life with continuous use and stores four weeks’ worth of data. It syncs with the Polar Flow app to view stats and analyze sleep and training data, among other information.

Because it has no buttons, activities can be started in the app or passively with what the company calls “automatic training detection.”

It’s available in the colors Greige Sand, Night Black and Brown Copper. Additional band colors are offered for $20 each.

There’s already a market of no-screen wearables, including the Whoop 5.0 wristband and smart rings such as the Oura Ring 3.

Will the Loop measure up?

Whether the Polar Loop’s attempt at simplifying a fitness wearable works out will largely depend on how well it runs and what it offers compared to other devices.

“The company is clearly tapping into the growing demand for screen-free wearables,” says CNET’s lead writer for wearables, Vanessa Hand Orellana. “It feels like a direct answer to the athlete-favorite Whoop band and even the Oura Ring, both of which collect similar health metrics to display and analyze in their respective apps.”

Hand Orellana says Polar has a good reputation, with its signature heart-rate chest straps, and may win over fans by eschewing the subscription fee that the Oura and Whoop require. 

“That said, as with most devices in this space, the real differentiator often comes down to execution… specifically, how well the data translates into clear, actionable insights. Personally, I’m curious to see how the Loop integrates with Polar’s app, which, at least in my experience with their HR straps, hasn’t always been the most intuitive to navigate,” she said.

Apple’s Siri overhaul may include an AI-powered web search tool with technology powered by Google’s Gemini, according to a new report from Bloomberg’s Mark Gurman. The iPhone maker, which has been criticized for falling behind in the AI race, delayed its long-awaited Siri update until 2026. In the meantime, the company has been scrambling to determine whether its own AI models alone will work well enough to make its upgraded Siri competitive with the AI answer engines available today from tech companies like OpenAI, Perplexity, and Google.

Per Bloomberg, Apple could be turning to Google for a solution to its problems. The report claims that Apple and Google reached a formal agreement this week that will see Apple testing a Google AI model in Siri. If successful, the technology could also be used in other areas of iPhone software, including the Safari browser and Spotlight search, which is available on the Home Screen.

In previous years, Spotlight seemed to be ramping up to become a rival of sorts to Google, as it allowed iPhone users to bypass web searches to get basic answers about popular topics, like information about actors, musicians, TV shows, and movies, among other things. With AI chatbots, however, consumers can now source quick answers about a wide range of topics beyond those that could be found on Wikipedia.

The report suggests that the upgraded search experience’s interface will use a combination of text, photos, videos, and local points of interest, as well as an AI-powered summarization feature. It will also be able to tap into users’ personal data and let them navigate their devices via voice.

I have a dead cockroach in my ceiling, and I only noticed it thanks to a Switch 2 screen protector that, ironically, is supposed to tame reflections. The protector in question is a pre-production sample of the Jsaux Obsidian AR screen protector, which is coming in October for $21.99.

The company claims it’s better than standard screen protectors because it features a similar physical vapor deposition (PVD) coating found in the Apple Studio Display. It’s supposed to reduce glare while retaining a glossy appearance — an ideal combination for gamers, since you can’t always choose where you’ll play games.

In reality, this pre-production screen protector boasts minor improvements that are easy to miss. Obsidian AR slightly reduces the intensity of reflections hitting the screen that could otherwise distract me during gameplay. However, I noticed these differences in photos more than I did while actively using the Switch 2.

One thing I did have the misfortune to notice as I played my Switch 2 with the screen protector equipped was the roach in my ceiling. That little sucker has probably been dead for a while and I only just noticed. In defense of Jsaux’s claims, it makes the bold LED tube illuminating the trapped roach a little dimmer and tougher to pick out, but not nearly enough to wipe this encounter from my memory.

I won’t go as far as to say that the Obsidian AR isn’t worth your money, having only tried an early version. Everyone’s tolerance for glare is different, but I’m skeptical that the differences are worth paying for compared to cheaper alternatives. To reiterate, just one of these protectors will cost $21.99 while you can get a three-pack of good-enough tempered glass protectors for under $10.

Photography by Cameron Faulkner / The Verge

A microRNA (miRNA)-based liquid biopsy known as the Destinex assay demonstrated high accuracy sensitivity in identifying gastric cancer (GC), including early-stage cases, according to research published in JAMA Surgery. Silei Sui, M.D., Ph.D., from the Department of Molecular Diagnostics and Experimental Therapeutics, Beckman Research Institute of City of Hope, Biomedical Research Center, Monrovia, California, and team sought to address a significant unmet clinical need for cost-effective, minimally invasive approaches to early GC detection in low-GC prevalence countries.

GC remains the third leading cause of cancer-related deaths worldwide, largely due to late-stage diagnosis. Most patients diagnosed at an advanced stage face a poor prognosis and are ineligible for curative treatment. In high-incidence countries like Japan and South Korea, mass endoscopic screening has improved outcomes, but such programs are invasive, costly, and impractical in low-prevalence regions.

Current blood-based tumor markers (e.g., carcinoembryonic antigen, cancer antigen (CA) 19‑9, CA125) lack the sensitivity and specificity needed for early-stage detection, leaving a critical gap in noninvasive, accurate, scalable screening tools.

Tissue biopsy remains the gold standard for cancer diagnosis, providing histopathologic confirmation and molecular profiling. However, it is invasive, carries procedural risks, and may be contraindicated in frail patients. Sampling errors can occur due to tumor heterogeneity.

Repeating biopsies for monitoring are often impractical. Liquid biopsies, including circulating tumor DNA (ctDNA), circulating tumor cells, and exosomal RNA, offer a minimally invasive alternative. While ctDNA is highly specific, it may lack sensitivity in early disease.

Exosomal miRNAs, encapsulated in vesicles secreted abundantly by tumor cells, combine stability with tissue specificity, making them promising for early detection.

The DESTINEX multicenter case-control study sought to address this gap by developing a comprehensive microRNA (miRNA)-based liquid biopsy that combines cell-free and exosomal miRNA signatures. Conducted across major referral centers in Japan and South Korea, the study analyzed 809 GC specimens from 480 patients between 2016 and 2020, progressing through discovery, training, validation, and evaluation phases.

Genome-wide transcriptomic profiling led to the development of a panel of 8 cell-free miRNAs and 10 exosomal miRNAs. Leveraging machine learning algorithms, Sui and team created a 17-miRNA GC signature or a molecular fingerprint, with the ability to distinguish between people with and without gastric cancer.

This combined signature demonstrated superior performance compared to individual panels, achieving robust Area Under the Curve (AUC) values of 96.3% (95% CI, 94.3%-98.4%) in the training cohort and 95.3% (95% CI, 92.8%-97.9%) in the independent validation cohort for GC detection. Building on this, researchers sought to create a more clinically feasible and inexpensive assay. They identified 5 overlapping miRNAs (miR-21-3p, miR-21-5p, miR-215-5p, miR-27a-3p, and miR-95-3p) common to both cell-free and exosomal panels. This led to the development of the final 10-miRNA signature, named Destinex.

The Destinex assay demonstrated comparable diagnostic performance to the 17-miRNA signature, achieving an AUC of 95.8% in the training cohort and 94.8% in the validation cohort. Crucially, Destinex proved highly effective in identifying early-stage (pT1) GC with an impressive AUC of 96.8% (95% CI, 93.5%-100%). The tumor-specific origin of these miRNA markers was confirmed by a significant decrease in their expression levels in post-surgery serum specimens.

Furthermore, Destinex exhibited high specificity for GC by outperforming its diagnostic value in other gastrointestinal cancers such as colorectal cancer, pancreatic ductal adenocarcinoma, esophageal squamous cell carcinoma, intrahepatic cholangiocarcinoma and hepatocellular carcinoma.

From a managed care perspective, Destinex could enable earlier detection in asymptomatic or high-risk populations and potentially shift diagnosis toward curable stages. As a potential cost-effective screening tool, its implementation could reduce reliance on invasive endoscopy in low-prevalence settings, lowering procedural costs and patient burden.

Early detection can reduce downstream costs associated with advanced disease management, hospitalizations, and palliative care. Moreover, this assay may support precision oncology workflows by integrating with guideline-concordant biomarker testing. There may also be an opportunity to improve care equity by offering a scalable, blood-based test that can be deployed in community settings, where access to advanced endoscopy is limited.

However, as with all biomarker innovations, real-world implementation will require validation in multinational, diverse populations. Health economic modeling must demonstrate cost-effectiveness to achieve easier integration into clinical pathways and payer coverage policies.

Elissa Welle is a New York-based reporter at The Verge covering all things AI, including the technology itself and the companies and policies that make and shape it. She joined the newsroom in 2025 and is currently supported by the Tarbell Center for AI Journalism. Previously, she wrote about science and health trends at Reuters, Stat News, IEEE Spectrum, among other outlets. Prior to being a journalist, she researched brain-computer interfaces for her biomedical engineering Ph.D. When she’s not writing about AI, she’s probably doing human things like eating food and touching grass. Contact her on Signal for tips: elissawelle.39.

Google has spent the last few weeks hyping up its new Pixel 10 phones, which are very nice devices. They’re just not a big leap over last year’s phones. If you’ve decided to hang onto your Android phone a bit longer, there are some new goodies headed your way. If you’ve got a Pixel, Google’s revamped Material 3 Expressive interface is rolling out.

Google’s Pixel Drop updates, which arrive quarterly, are not quite a new version of Android, but they include more than the bug fixes and security patches you get in other monthly updates. The September Pixel Drop (previously tested as Android 16 QPR1) is beginning its rollout today with one big change and a few little ones.

At the top of the list is the new Material 3 Expressive interface. Google revealed Material 3 in May, promising it would come to Android 16 phones later, but it didn’t intend to include it with the initial OS release. Indeed, Android 16 launched without Material 3 earlier this summer. Google’s new Pixel 10 series shipped with the new UI, and now more of the Pixel lineup is following.

Material 3 Expressive comes to more phones.

Google says it’s adding Material 3 Expressive to all of its phones from the Pixel 6 onward. This interface makes better use of theme colors, which are automatically extracted from your wallpaper. Google also included more bouncy animations, resizable quick settings, custom call screens, and more. While Google didn’t mention it, the new update should also have the new desktop mode experience.

Pixel phones are also getting a few more features for connected accessories. Later this month, Google will enable Adaptive Audio on the Pixel Buds Pro 2. This will allow the buds to adjust volume based on ambient noise levels. If you’ve got an Android smartwatch paired with your Pixel, Google Maps will open automatically when you begin navigation on your phone.

Revitalizing American manufacturing has been at the center of political and economic discourse — and presidential campaigns — for decades. David Siegel thinks he has an answer: Start paying more attention to our physical world and less on the virtual one.

Northeastern University President Joseph E. Aoun recently sat down with Siegel — a computer scientist, philanthropist and co-founder of tech-focused financial services company Two Sigma — to discuss everything from revitalizing American industry and infrastructure to artificial intelligence and next steps for higher education. 

The world of higher education will play a vital role in the revitalization, Siegel says. Partnerships between private industry and academic institutions are more necessary than ever, not only to train the next generation of innovators but also to provide them with the resources and space to pursue entrepreneurial endeavors in high-need sectors in the U.S., such as advanced manufacturing.

“Higher ed, like everything else in society, has to evolve, and it has evolved. Northeastern is an example of innovation in higher education. There’s a lot of money to be made, actually, by refocusing on these challenges, but it has to start with the education because who’s going to do it if no one graduates with an itch and the right skillset with an understanding of what the industry problems are?” Siegel told Aoun during a fireside chat in Northeastern’s ISEC Auditorium.

The solutions to many of our most pressing problems — housing, food, transportation — won’t be solved by AI but by everyone pitching in, Siegel says.

“It really comes from a thousand shining lights,” he says.

Endeavors like jumpstarting American advanced manufacturing or things like semiconductors face deficits in workforce and know-how, both of which have gone abroad. 

“What’s needed is everything, and everyone can do their part,” Siegel says. “Northeastern and other institutions can partner with industry to improve the workforce. The government can do certain kinds of things. Hopefully, the youth of America will want to devote their careers to it. They’ll hopefully do the thing that people do with software: They’ll create startups.”

However, universities have to be more open to their faculty and students starting businesses and entrepreneurial ventures in the first place, Aoun adds.

“Many of our faculty have launched various companies, and also some of these have been incubated here in the labs, and we really foster that,” Aoun explains.

All of that work will be a generation-spanning project but one that Siegel views as vital if the U.S. wants to solve some of its most challenging problems. For a generation, Siegel says, the U.S. “has ignored physical world problems from shipbuilding to building infrastructure of all kinds, any kind of manufacturing problem” in favor of pursuing innovation in tech that remains rooted in the virtual world.

Although Siegel expresses skepticism about AI — “it’s in the virtual world, our problems are in the physical world,” he says — he admits that AI tools will become turbochargers for academic research in a way that could produce physical world solutions. As a result, Siegel founded the Open Athena Project last year to help provide academic researchers with private-industry-caliber AI software engineers to fill a gap that exists in higher education.

Siegel and Aoun both agreed that, despite some anxieties about AI replacing entry-level jobs, these tools could instead raise the standard for what an entry-level job is in the first place. That, in turn, could raise the caliber of students who are coming out of universities and provide an even more skilled workforce downstream.

“Essentially, entry-level jobs are no longer going to be entry-level jobs,” Aoun says. “They’re going to be second level, which is good.”

To address the challenges Siegel discussed, the solutions have to start with education. That means not only evolving higher education but learning from global leaders across industries, Aoun says.

The conversation was the final piece of a daylong leadership retreat focused on Northeastern’s priorities across teaching and research, with a particular emphasis on the university’s artificial intelligence strategy.

Follow ZDNET: Add us as a preferred source on Google.

---

ZDNET’s key takeaways

• OpenAI for Science was announced in a Tuesday X post.
• Its goal is to accelerate scientific discovery through AI.
• The post suggests GPT-5 will play a key role in the effort.

---

Artificial intelligence researchers have long dreamed of automating the process of scientific discovery. Now OpenAI is setting out to turn that vision into reality.

The company is launching an initiative called OpenAI for Science, aimed at building “the next great scientific instrument: an AI-powered platform that accelerates scientific discovery,” according to a Tuesday X post from company Chief Product Officer Kevin Weil. The timeline for the project is not yet known; Weil added in his post that more information would arrive in the coming months.

Also: Every AI model is flunking medicine – and LMArena proposes a fix

(Disclosure: Ziff Davis, ZDNET’s parent company, filed an April 2025 lawsuit against OpenAI, alleging it infringed Ziff Davis copyrights in training and operating its AI systems.)

Weil, who will lead the effort, wrote that OpenAI will hire a team of “world-class” academics who are “completely AI-pilled” and “great science communicators” to join a small group of researchers already employed by the company.

What can we expect from the project?

Not much is known at this point about the “platform” that Weil and his fledgling team plans to build. But his X post hints at some aspects of the scientific process that OpenAI for Science might try to more effectively automate. 

Weil wrote that GPT-5, the newest model from OpenAI, which was released last month, “is clearly a new threshold” in the ability for AI to help advance scientific progress. For example, he cited a recent theoretical physics paper in which the model was used to suggest ideas for proofs, suggesting that OpenAI for Science could eventually aim, in part, to help researchers formulate hypotheses and research methods, thereby speeding up the pace of discovery.

Also: Patients trust AI’s medical advice over doctors – even when it’s wrong, study finds

The emphasis on GPT-5 in Weil’s X post smacks of a sales tactic: the model has received mixed reviews since it was launched, with many users complaining that it was inferior to its predecessor, GPT-4o. By wedding GPT-5 with the company’s new scientific research program, OpenAI could be attempting to repair the model’s damaged credibility. After all, if the company can prove that GPT-5 is able to meaningfully contribute to scientific discovery — a rigorous process which requires abstract, multistep reasoning, and which could hypothetically result in practical benefits for society at large — individual users and businesses could be more inclined to trust the model with their own sensitive tasks.

OpenAI did not immediately respond to ZDNET’s request for comment.

While Weil didn’t mention grant-writing in his blog post, this is another area in which generative AI tools like ChatGPT could fruitfully be applied: researchers currently spend close to half (45%) of their time writing grant proposals, according to the think tank the Institute for Progress.

Big strides

AI has yet to discover new physical laws, cure cancer, suggest a comprehensive solution to climate change, or make any of the other major scientific breakthroughs that many artificial general intelligence true-believers think could be around the corner. Perhaps one day an AI model will be able to completely automate the scientific process, from formulating entirely new hypotheses to conducting experiments to analyzing results. But for the time being, AI’s scientific prowess is rooted in its ability to identify intricate networks of patterns from existing data. 

Still, researchers have made some significant strides, and AI is rapidly becoming an integral tool in mainstream science.

Also: Open AI, Anthropic invite US scientists to experiment with frontier models

Google DeepMind CEO Demis Hassabis and Director John Jumper were awarded the Nobel Prize in Chemistry last year for their work on AlphaFold2, which uses AI to predict the structure of virtually all known proteins. The Nobel Prize in Physics last year was awarded to Geoffrey Hinton, one of the so-called “Godfathers of AI,” and physicist John Hopfield for their pioneering work on neural networks, which have become the technological framework for the current AI boom.

AI’s mathematical abilities are also continuing to rapidly evolve. In July, OpenAI reported that one of its experimental reasoning models achieved a gold medal-level performance on the International Math Olympiad, which is widely considered to be one of the world’s most elite math competitions. Google DeepMind reported the same level of performance from its own model, Gemini 2.5 Pro.

Want to follow my work? Add ZDNET as a trusted source on Google.

Apple Inc. is planning to launch its own artificial intelligence-powered web search tool next year, stepping up competition with OpenAI and Perplexity AI Inc.

The company is working on a new system — dubbed internally as World Knowledge Answers — that will be integrated into the Siri voice assistant, according to people with knowledge of the matter. Apple has discussed also eventually adding the technology to its Safari web browser and Spotlight, which is used to search from the iPhone home screen.

A short time into developing 007 First Light, something happened that meant parts of James Bond’s next adventure needed to be rewritten. No, it wasn’t anything going down internally at the studio that led to such a change, but something entirely more global: the death of Queen Elizabeth II of England.

The news was huge, but the impact on the script, not so much. It mainly revolved around instances of characters having to refer to protecting both King and country rather than Queen, and Bond being on His Majesty’s Secret Service rather than her.

“It was fairly small, it did feel significant,” revealed Martin Emborg, First Light’s narrative director, in an interview with IGN.

“It wasn’t like going in and going like, delete, delete, delete. It did feel, and we knew it obviously when we recorded it, that it was kind of a big deal that it’s like His Majesty’s Secret Service. It’s a new world.”

So new, in fact, that it’s the first time a 007 story has taken place during the reign of a male British monarch, with Ian Fleming’s original novels having been written mere months after Queen Elizabeth II’s 1952 coronation, and Daniel Craig’s final Bond film, No Time to Die, having been released a year before her passing. It’s a new world for Britain’s most famous spy, but not one that felt too daunting for IO Interactive to enter.

“It doesn’t feel weird”, replied Emborg. “I feel a great sense of kind of reverence for it. This is something that’s been going on for way beyond our lifetimes, and it’s inspired generations. So yeah, it feels like you get to play in a sandbox that is just way bigger than you.”

You can read about how Daniel Craig’s face was put into a Hitman map in order to secure the Bond rights here. You can also check out our story about the casting of IO’s new James Bond here, and the IO CEO’s thoughts on publishing MindsEye following its disastrous launch.

For much more on 007 First Light, make sure to read my extensive preview full of gameplay and story details.

Photo by Dominic Lipinski – WPA Pool/Getty Images.

Simon Cardy is a Senior Editor at IGN who can mainly be found skulking around open world games, indulging in Korean cinema, or despairing at the state of Tottenham Hotspur and the New York Jets. Follow him on Bluesky at @cardy.bsky.social.