Google AI Overviews Put People at Risk of Harm With Misleading Health Advice Slashdotby msmash on google at January 1, 1970, 1:00 am (cached at January 2, 2026, 10:35 pm)

A Guardian investigation published Friday found that Google's AI Overviews -- the generative AI summaries that appear at the top of search results -- are serving up inaccurate health information that experts say puts people at risk of harm. The investigation, which came after health groups, charities and professionals raised concerns, uncovered several cases of misleading medical advice despite Google's claims that the feature is "helpful" and "reliable." In one case described by experts as "really dangerous," Google advised people with pancreatic cancer to avoid high-fat foods, which is the exact opposite of what should be recommended and could jeopardize a patient's chances of tolerating chemotherapy or surgery. A search for liver blood test normal ranges produced masses of numbers without accounting for nationality, sex, ethnicity or age of patients, potentially leaving people with serious liver disease thinking they are healthy. The company also incorrectly listed a pap test as a test for vaginal cancer. The Eve Appeal cancer charity noted that the AI summaries changed when running the exact same search, pulling from different sources each time. Mental health charity Mind said some summaries for conditions such as psychosis and eating disorders offered "very dangerous advice." Google said the vast majority of its AI Overviews were factual and that many examples shared were "incomplete screenshots," adding that the accuracy rate was on par with featured snippets.

Read more of this story at Slashdot.

Trump Signs Defense Bill Prohibiting China-Based Engineers in Pentagon IT Work Slashdotby msmash on usa at January 1, 1970, 1:00 am (cached at January 2, 2026, 9:05 pm)

President Donald Trump signed into law this month a measure that prohibits anyone based in China and other adversarial countries from accessing the Pentagon's cloud computing systems. From a report: The ban, which is tucked inside the $900 billion defense policy law, was enacted in response to a ProPublica investigation this year that exposed how Microsoft used China-based engineers to service the Defense Department's computer systems for nearly a decade -- a practice that left some of the country's most sensitive data vulnerable to hacking from its leading cyber adversary. U.S.-based supervisors, known as "digital escorts," were supposed to serve as a check on these foreign employees, but we found they often lacked the expertise needed to effectively supervise engineers with far more advanced technical skills. In the wake of the reporting, leading members of Congress called on the Defense Department to strengthen its security requirements while blasting Microsoft for what some Republicans called "a national betrayal." Cybersecurity and intelligence experts have told ProPublica that the arrangement posed major risks to national security, given that laws in China grant the country's officials broad authority to collect data.

Read more of this story at Slashdot.

AMD Closes in on Intel in Latest Steam Hardware Survey Slashdotby msmash on amd at January 1, 1970, 1:00 am (cached at January 2, 2026, 8:06 pm)

AMD's share of processors among Steam users climbed to 47.27% in December 2025, a 4.66% jump in a single month that continues the company's steady encroachment on Intel's once-dominant position in the gaming CPU market. Intel held roughly 77% of the Steam Hardware Survey five years ago, and that lead has eroded considerably as AMD broke the 40% threshold in the third quarter of 2025 and kept climbing. The gains came despite an ongoing memory shortage that has pushed DDR5 prices to record highs -- AMD's AM5 platform requires DDR5 exclusively, while Intel's Raptor Lake Refresh chips support both DDR4 and DDR5. Many gamers are turning to older AMD Zen 3 processors like the Ryzen 5 5800X, which topped Amazon's bestseller lists during the holiday period and work on DDR4-compatible platforms. Meanwhile, the proportion of Steam users running 32GB of RAM rose to 39.07%, nearly matching the 40.14% still on 16GB, as gamers likely rushed to upgrade before prices climbed further amid AI's demand for memory.

Read more of this story at Slashdot.

Reading is a Vice Slashdotby msmash on books at January 1, 1970, 1:00 am (cached at January 2, 2026, 7:35 pm)

The International Publishers Association spent the past year promoting the slogan "Democracy depends on reading," but Atlantic senior editor Adam Kirsch argues that this utilitarian pitch fundamentally misunderstands why people become readers in the first place. The most recent Survey of Public Participation in the Arts found that less than half of Americans read a single book in 2022, and only 38% read a novel or short story. A University of Florida and University College London study found daily reading for pleasure fell 3% annually from 2003 to 2023. Among 13-year-olds, just 14% read for fun almost every day in 2023, down from 27% a decade earlier. Kirsch says to stop treating reading as civic medicine. "It would be better to describe reading not as a public duty but as a private pleasure, sometimes even a vice," he writes. When literature was considered transgressive, moralists couldn't stop people from buying dangerous books. Now that books are deemed virtuous, nobody picks them up. He points to Don Quixote and Madame Bovary -- novels whose protagonists are ruined by their reading habits. Great writers, he notes, never idealized literature the way educators do. The pitch to young readers should emphasize staying up late reading under the covers by flashlight, hoping nobody finds out.

Read more of this story at Slashdot.

A Decade of BBC Question Time Data Reveals Imbalance in Journalist Guests Slashdotby msmash on themedia at January 1, 1970, 1:00 am (cached at January 2, 2026, 6:35 pm)

A new study [PDF] from Cardiff University analyzing a decade of the popular topical debate programme BBC Question Time found that the broadcaster's flagship political debate show relies disproportionately on journalists and pundits from right-wing media outlets, particularly those connected to The Spectator magazine. Researcher Matt Walsh examined 391 editions and 1,885 panellist appearances between 2014 and 2024. Journalists from right-leaning publications accounted for 59.59% of media guest slots, compared to just 16.86% for left-leaning outlets. The Spectator, a conservative magazine with a circulation of roughly 65,000, had an outsized presence among the most frequently booked guests. The study's list of top non-politician appearances reads like a roster of right-wing media figures. Isabel Oakeshott appeared 14 times, Julia Hartley-Brewer 13, Kate Andrews (formerly of the Institute for Economic Affairs and now at The Spectator) 13, and Tim Stanley of The Telegraph and Spectator also 13. No equivalent frequency existed for left-wing journalists; Novara Media's Ash Sarkar and podcaster Alastair Campbell each appeared six times. Walsh said that the programme's need to be entertaining may explain some of these choices, as columnists unconstrained by party talking points tend to generate livelier debate. The BBC maintains that Question Time aims to present a "breadth of viewpoints," but the data suggests the programme's construction of impartiality tilts notably in one direction.

Read more of this story at Slashdot.

I’m never too old to make a young developer’s mistake. Scripting News(cached at January 2, 2026, 6:04 pm)

This is one of those times when you pay the price for working past your timeout, like a pitcher pitching too many pitches in one game. There's an item on your checklist, and you'd love to take it off before you start the next session. You copy paste something, get it to work, and close out your session for the day.

While you're preparing to get up you decide to share a linkblog item, and when it boots up there's something severely broken. It has nothing to do with anything on your list! You roll it up for the day anyway, figuring it'll be easy to find when you start the next session. No sir. You have to work through complicated code you haven't touched in months! Where the fuck is this, and how the fuck did it happen. Drink more coffee. Keep stepping through the code to identify when things went bad. It happens in the most impossible place. You keep digging in. Even more impossible. Unfathomable. But you keep going in, making sure every bit of code does something you think it should and then boom! You left a copy in some DOM code in something you cribbed so now there are two of this thing you hadn't touched or even looked at in months. Delete the mistake, rebuild, click off all the breakpoints. And yup that was it.

I started work at approx 9AM and it's 11:41AM — wasted the best part of the day, all to learn the lesson that no more "one more feature" before I get up bullshit. If you're tired you make mistakes, and you can't not pay for them.

No doubt I will continue to learn this and other lessons all through what remains of my career. Some things you never learn, like "I can always roll back" — but where should I roll back to? When did I break this mofo? Oy. I'm never willing to go back that way.

'Results Were Fudged': Departing Meta AI Chief Confirms Llama 4 Benchmark Manipulati Slashdotby msmash on facebook at January 1, 1970, 1:00 am (cached at January 2, 2026, 5:06 pm)

Yann LeCun, Meta's outgoing chief AI scientist and one of the pioneers credited with laying the groundwork for modern AI, has acknowledged that the company's Llama 4 language model had its benchmark results manipulated before its April 2025 release. In an interview with the Financial Times, LeCun said the "results were fudged a little bit" and that the team "used different models for different benchmarks to give better results." Llama 4 was widely criticized as a flop at launch, and the company faced accusations of gaming benchmarks to make the model appear more capable than it was. LeCun said CEO Mark Zuckerberg was "really upset and basically lost confidence in everyone who was involved" in the release. Zuckerberg subsequently "sidelined the entire GenAI organisation," according to LeCun. "A lot of people have left, a lot of people who haven't yet left will leave." LeCun himself is departing Meta after more than a decade to start a new AI research venture called Advanced Machine Intelligence Labs. He described the new hires brought in for Meta's superintelligence efforts as "completely LLM-pilled" -- a technology LeCun has repeatedly called "a dead end when it comes to superintelligence."

Read more of this story at Slashdot.

Ghana Tries To Regulate Online Prophecies Slashdotby msmash on social at January 1, 1970, 1:00 am (cached at January 2, 2026, 4:35 pm)

Ghana has decided to deal with the viral spread of prophetic content on social media by setting up an official reporting mechanism for sensitive predictions, a move triggered by the August 2025 helicopter crash that killed the country's defence and environment ministers along with six others. After the accident, TikTok clips circulated showing pastors who claimed to have foreseen the disaster before it happened. Elvis Ankrah, the presidential envoy for inter-faith and ecumenical relations, now asks prophets to submit their predictions for review. Charismatic preacher-prophets have been a fixture of Ghanaian public life since Pentecostalism arrived in the 1980s, but social media has amplified their reach and made their claims increasingly outlandish. Police have threatened to arrest prophets who cannot prove their predictions eventually came true. Some two-thirds of Ghanaians favor giving divine intervention a role in politics. Ankrah recently declared that most prophecies submitted to him are "total bunk."

Read more of this story at Slashdot.

[no title] Scripting News(cached at January 2, 2026, 4:33 pm)

BTW, I used to have a tradition in the early days of this blog to write new stuff about an important idea on January 1 each year. At some point I stopped doing that. Now I realized that unintentionally I have just written such a piece, below. There's a lot of good stuff in that piece and in the places it links to. See the web is still useful. You won't hear these ideas on CNN or MS.NOW or in the NYT, WP, or from any billionaires either. I'm not saying I'm right, I've definitely been wrong before. But I think I'm mostly right. ;-)
Good Old Medicare Scripting News(cached at January 2, 2026, 4:03 pm)

This year I switched to Original Medicare for 2026 after being on super duper insurance industry enhanced Medicare for the first five years. I should've known that it's super duper for them but not so great for the insured person or the taxpayer, that is -- me. They have set it up so they can scam it 18 different ways, Some of the "services" they offer are just ways for them to charge the government more for your care, in return for doing absolutely nothing.

So today I had my first Regular Medicare experience. I went to renew my prescriptions, and the vendor already knew my insurance info which was impressive use of computers. I was prepared to give it the new info. And in case you're confused, you still have to pay the insurance industry for Regular Medicare, even though that's actually a deal between me and my government. No way to deal them out, though I'm sure it would make total sense for us to do exactly that. Don't believe that when it comes to health care the US has been anything like a democracy.

Where the old plan each prescription cost $0, this time I actually have to pay out of pocket approx $100 for a 90 day supply of five different medications. On the other hand, I'm paying much less for this insurance. I haven't done the math yet, but it feels like I'm not getting ripped off. See how many qualifiers there are.

Oh the great old US of A where you can give your tax money and retirement funds to insurance companies who buy politicians etc and so on. Maybe the only thing to appreciate is that we get some health care for all our tax payments over our careers?

And one good thing is now I get to keep most of my Social Security payments instead of it all going to United Healthcare. ;-)

The web after AI Scripting News(cached at January 2, 2026, 4:03 pm)

I still use Google. And I like the AI response they put at the top of the page, even though I get that it's bad for the web. Maybe there's a position for a web-only search engine. One where my writing has a chance of being seen by interested parties, rather than being converted to slurry to be fed to consumers as one would get a tasty hot dog from a NYC street vendor. Another level of processing. If I recall correctly one of the browsers is saying they're not using AI at all. But this would be different, it would be for search, not the browser.

But Google, if you want to be fair about it, there should be a permalink for these requests. This is something none of the AI vendors seem to be able to do. If so, I wonder why?

I just tried looking up "Regular Medicare" on Google, I was looking for an official page for it and ended up using the Medicare.gov one. But I could have decided the Google synopsis was the best. No way that I can see to use that as a link.

Maybe we should all start thinking about how we would like the new post-AI web to work?

I am trying to reach the well-know political pundits to tell them we don't have to accept the web the billionaires give us. The costs are very low, their money doesn't help them as much as you might think, because they have different concerns from the ones you (writer) and I (developer) have. I don't care about revenue. This is my way of giving back, the area I have the greatest leverage. I always argued with Bill Gates who kept saying he was going to be a scrappy competitor now and give away his wealth when he retired. He did do that btw, but I thought man you're never going to have the leverage you have now. Your money as a way to end famine or fight the climate crisis is nothing compared to the power you have when you have the dominant operating system in a time of great transition. You could, I argued, make the web a breakthrough for self-government. That possibility really was within reach, I believed then and still do now.

I made the choice Bill didn't and it worked. We were able to build something amazing, only to see it broken down into that slurry I mentioned at the top.

I love AI, I don't want to go back. But I also have big plans for human thinkers and developers -- writers, using the technology we never got a chance to develop while we were busy being dominated by Zuckerberg, the Twitter founders, et al. As I said on Bluesky today, we're going to insist this time that the political punditry get involved, assuming they are serious about saving what's left of American democracy.

The Atlanta Journal-Constitution Prints Final Newspaper, Shifts To All-Digital Forma Slashdotby msmash on themedia at January 1, 1970, 1:00 am (cached at January 2, 2026, 3:35 pm)

CBS News: The Atlanta Journal-Constitution has printed its final newspaper, marking the end of a 157-year chapter in Georgia history and officially transitioning the longtime publication into a fully digital news outlet. The front-page story of the final print edition asks a fitting question: "What is the future of local media in Atlanta?" The historic last issue is also being sold for $8, a significant increase from the typical $2.00 price. Wednesday, Dec. 31, marks the last day The AJC will be delivered to driveways across metro Atlanta. Starting Jan. 1, 2026, the newspaper will exist exclusively online, a move its leadership says reflects how readers now consume news and ensures the organization's future. AJC President and Publisher Andrew Morse said the decision was not made lightly, especially given how deeply the paper is woven into daily life for generations of readers. The move makes Atlanta the only major U.S. city without a daily printed newspaper.

Read more of this story at Slashdot.

How Nokia Went From iPhone Victim To $1 Billion Nvidia Deal Slashdotby msmash on business at January 1, 1970, 1:00 am (cached at January 2, 2026, 1:35 pm)

Nokia, the Finnish company whose iconic ringtone was played an estimated 1.8 billion times daily at the height of its mobile phone dominance and whose 3310 "brick" sold 126 million units, has reinvented itself again -- this time as a key piece of AI infrastructure. In October, Nvidia announced a $1 billion investment in Nokia and a strategic partnership to incorporate AI into telecommunications networks. The company that was once worth $335 billion and controlled more than a quarter of the global handset market seemed destined for irrelevance after the iPhone's 2007 arrival. A last-ditch bet on Microsoft's Windows phone system in 2011 failed, and Nokia sold its devices division to Microsoft for $6.34 billion in 2014. Revenues had fallen from $44.27 billion in 2007 to $12.56 billion. Nokia rebuilt around its $2 billion acquisition of Siemens' networks stake in 2013, then added French network provider Alcatel-Lucent for $18.32 billion in 2015. Current CEO Justin Hotard, who took over in April, has pushed the company further into cloud services, data centers and optical networks. Nokia acquired optical specialist Infinera for $2.3 billion in February. The company's optical technology enables information to pass between data centers, and it produces routers for cloud-based services.

Read more of this story at Slashdot.

Record year for wind and solar electricity in Great Britain in 2025 BBC News | Science/Nature | UK Edition(cached at January 2, 2026, 1:01 pm)

Wind, solar and other renewables hit a new high last year but the government is still some way off its clean power target.
Controversial post-Brexit farm subsidy scheme 'landmark moment for Wales' BBC News | Science/Nature | UK Edition(cached at January 2, 2026, 12:00 pm)

The post-Brexit subsidy scheme for farmers begins, almost a decade after the vote to leave the EU.