OpenAI president Greg Brockman called Moonshot AI’s Kimi K3 “pretty good” in a Bloomberg interview on Tuesday, a rare public acknowledgement from a US frontier lab that a Chinese rival has produced a competitive model. Brockman said it is “too early” to determine whether Moonshot trained K3 by distilling outputs from OpenAI’s own systems, a practice the company has repeatedly accused Chinese labs of using. His measured tone stood in sharp contrast to the White House, which hours earlier had levelled its most specific accusations yet against the Chinese startup.

OSTP Director Michael Kratsios accused Moonshot of accessing banned Nvidia chips through intermediaries in Thailand and of conducting large-scale distillation against American AI models, according to Bloomberg. The accusations landed the same day that Treasury Secretary Scott Bessent said the administration would scrutinise Chinese open-source models for stolen intellectual property and could impose sanctions within weeks. Nvidia CEO Jensen Huang pushed back against the alarm, calling Chinese open models “excellent” and urging the government to let American firms use them.

The distillation charge has been building for months. In February, Anthropic accused Moonshot, DeepSeek, and MiniMax of running roughly 16 million exchanges through about 24,000 fraudulent accounts to extract knowledge from its Claude models. OpenAI has made similar complaints, and the White House now appears to be treating the allegations as a matter of national security rather than a commercial dispute between rival labs.

Brockman framed OpenAI’s position around the company’s infrastructure investments, arguing that the US has a “huge advantage” because of the billions it has poured into compute. He said open-weight models like K3 are not truly free because running them at scale still requires expensive hardware, a point that undercuts the narrative that Chinese labs can simply give away frontier-level AI. He estimated that China remains roughly four months behind the US in overall model development.

K3 is a nearly three-trillion-parameter open-weight model that Moonshot released on July 16, and it has rattled Washington and Wall Street by matching or beating the best American systems on several benchmarks. The model topped Arena’s frontend coding leaderboard within 24 hours of launch, placing ahead of both Anthropic and OpenAI’s flagship offerings. Moonshot is pricing K3 at a fraction of what US labs charge, accelerating the commoditisation of the model layer that investors have been warning about since DeepSeek’s price cuts earlier this year.

Brockman, who donated to a Trump-aligned super PAC, declined to wade into the debate over whether Chinese models should be banned. He instead stressed that OpenAI believes open models are important for “democratizing access to AI,” a position that aligns him more closely with Huang than with the White House officials pressing for sanctions.

from the everything-is-more-nuanced dept

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

Concerns surrounding young people’s screen time are widespread.

Australia became the first country to ban social media for users under 16 in December 2025, and Denmark, France and the U.K. have since announced similar restrictions to begin this year.

In the U.S., as of mid-2026, more than 30 states have passed laws banning or restricting cellphones in K–12 classrooms; in 2023, the U.S. surgeon general issued a formal advisory on social media and children’s and adolescents’ mental health; and bestselling books tell parents that smartphones are “rewiring” their children’s brains.

These concerns and policies are part of a quickly changing national and international conversation around how young people spend time on screens and its relationship to their overall health and development. My reading of the mounting research on this issue across disciplines is that the popular narrative blaming screens and smartphones for an adolescent mental health crisis runs well ahead of the current evidence.

I study adolescent digital media use and its influence on social, emotional and academic outcomes. A growing body of research suggests that one-size-fits-all solutions are not the answer and that managing appropriate use of digital media needs to take into account a child’s developmental milestones, how parents and adults around them use media, and the ways kids use it to connect and learn with friends and family.

Screen time: From monolith to multifaceted

Wide adoption of digital media and the internet broadened the range of experiences young people could have online. At the same time, the digital age introduced newfound uncertainties. As with the advent of radio, comic books and arcades, adults worried about how children might interact with or be affected by internet use.

In response, the American Academy of Pediatrics first recommended in 1999 that parents and caregivers keep children under 2 away from screens. In the decades since, professional guidance largely treated children’s media use as a behavior to be mitigated.

Policies introduced by the academy in 2013 and 2016 continued to advise that school-age kids and adolescents – those ages 5 to 18 – be restricted to no more than two hours of “entertainment” screen time a day. The goal was to curb risks associated with heavy media use, among them disrupted sleep, online safety, cyberbullying and physical inactivity.

Originally created for young people’s engagement with stationary media that tend to be confined to one room or context – for example, watching television – these hourly limits became outdated with the integration of smartphones and other digital devices into everyday life. Compared with watching television, online media was far more difficult to track and define, and more nuanced in its use.

Developmentally beneficial activities such as education, socializing and leisure have come to rely on the internet to extend and maintain face-to-face connections. Remote schooling and social distancing during the COVID-19 pandemic only accelerated this digitization of daily life.

In my view, adopting strict time limits and restrictions could pose risks to children’s well-being, autonomy and development, for example, by harming adolescent self-esteem.

The latest guidelines

In January 2026, the American Academy of Pediatrics retired its decade-old framework that had largely organized its advice around hourly screen limits. The new policy statement on children, adolescents and digital media diverges from this blanket approach. Instead, it suggests parents consider the larger picture in which this media use exists rather than lumping all screen use together.

Similar to the World Health Organization’s 2019 guidance for children under 5, the American Academy of Pediatrics still advises that parents avoid screen media for children younger than 18 months. This recommendation is largely because extended use by children by themselves can be problematic for many young children, crowding out important developmental milestones.

Both the World Health Organization and the American Academy of Pediatrics also recommend that when children under 24 months use screens, they should be limited to content and devices that encourage children and caregivers to interact. For ages 2 to 5, screen time – including TV and interactive apps on devices – may be extended to more solo use, provided it’s high-quality digital media designed around learning goals in mathematics and reading. But recreational use should be limited to roughly an hour per day.

For school-age children and teens, the newest guidance has begun to step away from fixed screen time limits and asks families to weigh online activity in the context of everyday life.

Doing so recognizes that a child’s digital experiences are shaped by diverse factors rather than the hours spent online. Current guidelines call on caregivers to distinguish among types of media, from television and social media to video games and interacting with artificial intelligence chatbots. They also call for taking into account a child’s individual characteristics, such as their interests and personality, family members’ own use of screens, and the type of content children are spending time on.

Rethinking screen time

Moving beyond strict screen time limits includes questioning the kind of digital activities kids and adolescents participate in. Do the activities encourage time spent interacting with others online, which can help young people develop important skills and competencies?

Scrolling an algorithm-based, auto-playing video feed likely does not equate to the same opportunities as video-chatting with friends, creating digital art or working with teammates in a multiplayer game. Research suggests these different uses relate to development in different ways and can help kids develop varying skill sets pertaining to everyday life and schooling.

Indeed, a large review of current research found that young people who take part in a range of digital activities, such as browsing the web, online gaming or interacting on social media, show positive associations with social connection, identity exploration, civic participation and learning.

Using these guidelines at home

The current evidence suggests parents and caregivers are best positioned to be digital instructors. Cutting children off altogether can carry its own risks for social and emotional development. Caregiver mediation of children’s screen time can produce widely different outcomes and effects, depending on whether the guidance is supportive or controlling.

Considering your own digital media use is the first step: Are family members engaging in problematic or heavy media use that children in the household might emulate? What applications and uses are most common in the family, and what positive or negative effects might they have, depending on the child’s age? How could these digital activities be safely integrated with other everyday experiences to increase their benefit for children? Conversely, what online time might be better spent on face-to-face experiences?

The American Academy of Pediatrics’ Family Media Plan tool turns these ideas into concrete questions. For example, it recommends working out what each child needs from digital technology, what activities screens might be crowding out, and where their family or household can build in screen-free time. The recommendation is to talk with each child about why they are drawn to particular apps or online activities, what they encounter while browsing, and what might be lost when kids bring phones to gatherings such as mealtimes.

The debate over young people’s screen time is not going away. But the most up-to-date guidelines, and the growing body of research behind them, make a strong case for a more holistic approach. The guidelines treat digital media as a complex, diverse and evolving environment that children need to learn to navigate in the digital age. The risks and rewards depend, as with any developmental setting, on the child, the content and what online time might be crowding out.

Gabriel E. Hales is a Research Fellow in Media and Information at Michigan State University

Filed Under: kids, screentime

Juzear might not be a familiar name to most American listeners, but the Chinese manufacturer is gaining attention in the increasingly crowded wired IEM market. That category is enjoying a rather healthy moment as more listeners rediscover that better sound, replaceable cables, and freedom from batteries still have value in a world drowning in disposable wireless earbuds.

For the new Juzear Nebula, the company has partnered with Japanese reviewer KOTO on a five-driver tribrid IEM featuring one dynamic driver, three balanced-armature drivers, and one micro-planar driver in each earpiece. Juzear says the Nebula has been tuned around forward, expressive vocals and an expansive three-dimensional soundstage, although those claims will ultimately require independent listening rather than blind faith in a specification sheet.

The collaboration also follows a growing trend of YouTube personalities and online reviewers lending their names and tuning preferences to audio products. Some of those partnerships clearly move units, but they also muddy the waters when the same people helping to develop products are part of an ecosystem built around reviewing them. A subscriber count is not proof of objectivity or a hearing test, but recognizable names apparently move the needle for enough readers and viewers to keep the collaborations coming.

The Juzear Nebula utilizes a 1 dynamic driver, 3 balanced-armature drivers, and 1 planar driver configuration, combined with a high-precision four-way physical crossover to deliver a transparent sound profile. The goal is to reduce congestion and produce a more spacious, high-fidelity listening experience.

Juzear’s four-way acoustic crossover system incorporates four independent sound tubes, with specific frequency ranges assigned to the dynamic, balanced-armature, and micro-planar drivers. According to the company, this helps the Nebula deliver a clear, transparent, and highly detailed presentation across the entire frequency range.

The Nebula’s MACCD dynamic driver, which uses a Multi-Layer Amorphous Carbon Composite Diaphragm, is designed to produce fast, deep, and tightly controlled bass.

For the midrange, a custom three-driver balanced-armature configuration is intended to deliver natural, clear vocals, while a customized micro-planar driver handles the treble and provides extended high-frequency response and greater detail.

Related Reviews:

Making a Strong First Impression

Beyond its internal architecture, the Nebula also showcases Juzear’s attention to craftsmanship. 

The Juzear Nebula draws its visual inspiration from the flowing light trails of deep-space nebulae. Its intricate Phantom Color Artistic faceplates are created using a combination of laser layering, micro-jet printing, and color-refraction processes.

The result is a visually striking finish that appears to glow and shift depending on the viewing angle. Lightweight, ergonomically shaped resin shells complete the design, with each earpiece weighing just 6.1 grams for a comfortable fit during longer listening sessions.

Cable & Connectivity 

Juzear includes a high-purity cable with the Nebula that appears more substantial than its price might suggest. It features 6N OCC silver-plated conductors with a 400-core configuration, protected by a durable, double-layer PVC jacket.

The cable also incorporates an interchangeable termination system, allowing users to switch easily between the included 3.5 mm single-ended and 4.4 mm balanced plugs to match their source equipment.

Juzear + Koto Nebula IEM Specifications

The Bottom Line 

Wired IEMs continue to gain popularity among listeners who want reliable sound quality without batteries, Bluetooth codecs, or software complications. At $129.99, the Juzear Nebula stands out with its five-driver design, four-way crossover, lightweight resin shells, and interchangeable 3.5 mm and 4.4 mm plugs.

It is aimed at listeners who prioritize vocals, spacious presentation, and distinctive styling. What remains unknown is how well the five drivers integrate, how effective the passive isolation will be, and whether the fit works across a wide range of listeners. The specifications look competitive, but the tuning will determine whether the Nebula is genuinely special.

Price & Availability

The Juzear x KOTO Nebula IEMs are priced at $129.99 at HiFiGo.

Related Reading

More and more, we want to know: Is it human, or is it AI? Is that image real? That song? That video? The publishing platform Substack is trying to add some clarity by partnering with the AI detector Pangram to help readers understand just how much AI is in all the content they consume on the publishing platform.

The tool is available to all Substack subscribers and works on posts, notes, comments and replies published on or after Tuesday that are longer than 100 words. Creators will also be able to scan their drafts with Pangram to see how much AI is detected and can also provide a statement to explain their writing process, Substack said.

Web and iOS users can try the Pangram features now, and Android users will get access later.

In a blog post announcing the new feature, Substack co-founder and CEO Chris Best said his platform aims to avoid “Claudefishing,” meaning when a reader consumes content without realizing it was generated by AI.

“When readers have to wonder if what they’re reading is real, it undermines trust in authorship and threatens the livelihood of writers — including those who use AI tools thoughtfully to produce work they believe in,” Best wrote.

From CNET: AI Slop Is Destroying the Internet. These Are the People Fighting to Save It

Pangram estimates that roughly 40% of text on some social media platforms is AI-generated, with a stunning two-thirds of that on the networking site LinkedIn. And AI growth agency Graphite recently published a white paper saying that AI is now writing as many online articles as humans. A Gartner survey found that 68% of people wonder if the content they see is real, and that half of consumers prefer brands that aren’t using AI in their marketing.

Testing it for myself

I’ve only recently joined Substack, but I wanted to check out the new Pangram AI detection tool. The tool works on posts, notes, comments and replies, so I decided to use posts as my makeshift testing lab. First, I logged in to my Substack account and published a new post. I copied and pasted text from an article I had written — by myself, no AI — into the Substack post and published it.

After a few minutes, I clicked on the live post. On the three-dot menu at the upper right of the page, I saw a “Scan for AI text” option. I clicked on it, and a small window popped up accurately declaring that my article was “Fully Human-Written.” A line appears below this that indicates the percentage of the text that is AI, AI-assisted, or human.

I next took the same post but added a long, fully AI-generated paragraph that I pasted from ChatGPT. When I performed the “Scan for AI text,” the tool reported that 10% of the article was written by AI, a roughly accurate estimate.

I next published a fully AI-generated post, and the scan correctly identified it as 100% AI-generated.

For another test, I added a couple of AI-generated sentences to a post that was otherwise purely human content, and the scan flagged it as 100% human-generated. Close, but not quite right. So, the Pangram tool worked pretty well for my crude experimentation.

Although I was fairly satisfied with how well Substack’s AI detection system worked on my own content, I was disappointed to learn that one of my favorite podcaster-bloggers apparently went full AI on his latest post. Other writers I checked out, however, used fully human-generated text.

What about false positives?

In his announcement, Best admitted that the AI detection tool is “not perfect,” but cited independent research that found Pangram’s software was 97.5% accurate in detecting fully generated AI text. The University of Maryland found that Pangram was 99.3% accurate in identifying “humanized AI-generated text,” which basically means a human has rewritten or paraphrased AI content.

Substack writers who do not use AI are understandably concerned about being incorrectly labeled as AI-assisted authors. As a result, the risk of false positives is a significant issue. Substack said creators can scan their content to see if it will be labeled as containing AI while it’s still in draft form. If Pangram incorrectly identifies AI text, the creator can report the error to Substack and have the false scan removed.

To promote greater transparency, publishers can now include a “How I make this” statement if they want readers to know how much, if any, AI they have used in their content. Best said Pangram can’t detect if AI was used to gather information or as a source for written pieces.

“We’re not against people using AI to assist their work, and we think people should be free to choose which tools they use to express themselves,” Best said. “But people should know what they’re getting.”

Ever since being admittedly fascinated by the Cambridge coffee webcam from the 1990s, I’ve written about VPNs, the NFL, smartphones, living wages, over/unders and everything in between.

See full bio

Advertisement

You can now gift games without a Steam account and sort your wishlist into categories.

AI AND ML

Yet the Chocolate Factory is also buying more third-party compute capacity to handle demand for the G-Cloud

Google’s parent company Alphabet is managing its fleet of AI accelerators to prioritize research on artificial general intelligence, rather than renting them all to customers.

CEO Sundar Pichai revealed the company’s priorities during its second quarter earnings call, during which the company confirmed it has delivered on its plan to sell its tensor processing units (TPUs) to some customers. In response to news of those sales, Goldman Sachs analyst Eric Sheridan asked how Alphabet balances demand from customers who want to buy its TPUs and the web giant’s own need for processing power.

“In terms of allocating our TPUs … our first priority is making sure we are allocating what we need to compete at the frontier in terms of AGI development,” Pichai replied, referring to Artificial General Intelligence –AIs that possesses human-like intelligence.

“That is the foundation for everything we do,” Pichai added.

Another analyst, Mark Shmulik of Bernstein, revisited the matter by asking how Google allocates processing capacity among its search business, cloud operation, and model training efforts.

“On allocation, I think the baseline with which we start is what it takes to continue AGI development at the frontier,” Pichai responded. The CEO said Alphabet is also “prioritizing our core product areas like Search, YouTube, et cetera, as well as Cloud.” And in the G-Cloud, Google is “prioritizing the compute to make sure we can serve our models in the context of Vertex and Gemini Enterprise, and our core solutions, be it data analytics and cybersecurity.”

“Our core services for our core products across consumers and enterprises is where the compute is primarily going, and that’s how we think about it,” the CEO added.

Google’s core services are going gangbusters.

Google Cloud revenue leapt 82 percent year over year, to $24.75 billion for the quarter, and delivered an $8.8 billion profit which represented 214 percent growth. The Big G said that growth came from “strong demand for AI infrastructure and AI solutions.” There’s probably more to come as the G-Cloud now has $514 billion of cloudy backlog on the books, meaning customers have signed up for services they’re yet to consume.

Search revenue grew 17 percent and YouTube ads grew 13 percent.

When generative AI came along some pundits suggested it could threaten Google’s search ads biz. That was not a good take because Pichai said the company’s AI Mode for search is “driving an incremental increase in Search queries overall.”

AI search needs specialist hardware that is expensive to buy and run. Pichai said Google is on top of that. “Thanks to our engineering and hardware optimizations, this quarter we reduced the cost of AI Mode responses to its lowest level since launch, even as we’ve brought more advanced AI capabilities.”

Google continues to spend megabucks on AI infrastructure – CFO Anat Ashkenazi said the company now plans to spend between $195 billion and $205 billion this financial year, up from a previous estimate of $180 billion to 190 billion.

Ashkenazi said Google can’t get all the kit it needs due to what she described as “the supply-constrained environment.” Google therefore plans to “expand the use of third-party capacity in Q3 as a bridging strategy while we build out more internal capacity.”

Pichai said buying bridging capacity will help Google to land monster cloud clients.

“There are very, very large customers of ours on Cloud who we are trying to support them through this extraordinary moment,” the CEO said. “The incremental opportunities they are bringing to us, while a short-term cost over a few months may be very high, in the lifetime of the deal, as we bring more capacity on, is highly ROI positive.”

“Those are factors we are taking into account. Are you willing to take upfront six-month deal to be able to serve that customer in what is a multi-year opportunity, where the margins and the returns are very, very attractive over that multi-year horizon?”

Alphabet’s quarterly revenue landed at $119.8 billion, up 24 percent year over year. Operating income hit $40.8 billion, up 34 percent.

Yet even those torrents of money couldn’t stop Google’s free cash flow landing at -$5.9 billion –  the first time the company hasn’t had spare cash to splash since 2004. Investors seem not to like that and sent the price of the company’s shares down by four percent in after hours trading. ®

from the good-deals-on-cool-stuff dept

StreamSkill has combined seven of their best Office 2021/365 courses to create this awesome Ultimate Microsoft Office 2021/365 Bundle. Whether you’re a student, an office professional, a business owner, a freelancer, or just someone who wants to get started with the most widely used Microsoft programs, this course bundle is for you. Join now and gain a thorough understanding of Microsoft Office 2021/365. It’s on sale for $40.

Note: The Techdirt Deals Store is powered and curated by StackSocial. A portion of all sales from Techdirt Deals helps support Techdirt. The products featured do not reflect endorsements by our editorial team.

Filed Under: daily deal

Adding new dimensions to the device.