An anonymous reader quotes a report from the BBC: New research coordinated by the European Broadcasting Union (EBU) and led by the BBC has found that AI assistants — already a daily information gateway for millions of people — routinely misrepresent news content no matter which language, territory, or AI platform is tested. The intensive international study of unprecedented scope and scale was launched at the EBU News Assembly, in Naples. Involving 22 public service media (PSM) organizations in 18 countries working in 14 languages, it identified multiple systemic issues across four leading AI tools. Professional journalists from participating PSM evaluated more than 3,000 responses from ChatGPT, Copilot, Gemini, and Perplexity against key criteria, including accuracy, sourcing, distinguishing opinion from fact, and providing context.

Key findings: – 45% of all AI answers had at least one significant issue. – 31% of responses showed serious sourcing problems – missing, misleading, or incorrect attributions. – 20% contained major accuracy issues, including hallucinated details and outdated information. – Gemini performed worst with significant issues in 76% of responses, more than double the other assistants, largely due to its poor sourcing performance. – Comparison between the BBC’s results earlier this year and this study show some improvements but still high levels of errors. The team has released a News Integrity in AI Assistants Toolkit to help develop solutions to these problems and boost users’ media literacy. They’re also urging regulators to enforce laws on information integrity and continue independent monitoring of AI assistants.

Read more of this story at Slashdot.

The makers of BIND, the Internet’s most widely used software for resolving domain names, are warning of two vulnerabilities that allow attackers to poison entire caches of results and send users to malicious destinations that are indistinguishable from the real ones.

The vulnerabilities, tracked as CVE-2025-40778 and CVE-2025-40780, stem from a logic error and a weakness in generating pseudo-random numbers, respectively. They each carry a severity rating of 8.6. Separately, makers of the Domain Name System resolver software Unbound warned of similar vulnerabilities that were reported by the same researchers. The unbound vulnerability severity score is 5.6

Revisiting Kaminsky’s cache poisoning attack

The vulnerabilities can be exploited to cause DNS resolvers located inside thousands of organizations to replace valid results for domain lookups with corrupted ones. The corrupted results would replace the IP addresses controlled by the domain name operator (for instance, 3.15.119.63 for arstechnica.com) with malicious ones controlled by the attacker. Patches for all three vulnerabilities became available on Wednesday.

Read full article

Comments

Tesla reported its financial results for the third quarter of 2025 this afternoon. Earlier this month, we learned that the electric vehicle manufacturer had a pretty good Q3 in terms of sales, which grew by 7.3 percent year over year and cleared out tens of thousands of cars from inventory in the process. However, that hasn’t translated into greater profitability.

Even though revenues grew by 12 percent to $28 billion compared to the same period last year, Tesla’s operating expenses grew by 50 percent. As a result, its operating margin halved to just 5.8 percent. And so its profit for the quarter fell by 37 percent to $1.4 billion.

Some growth in revenue came from its battery and solar division; this increased by 44 percent to $3.4 billion compared to Q3 2024. Services—including the Supercharger network, which is now open to an increasing number of other makes of EV—also grew, increasing by 25 percent to $3.4 billion. EV deliveries increased by 7 percent to 497,099, most of which were the Model 3 sedan and Model Y crossover. Automotive revenues grew slightly less, increasing 6 percent year over year to $21.2 billion.

Read full article

Comments

This is a video of a robotic spy octopus built by John Downer Productions for the BBC/PBS series Spy In The Wild capturing the daily activities of a coconut octopus. And what exciting daily activities they are! The coconut octopus is unique in that it’s one of only two species known for walking bipedal on two tentacles (seen above and at 2:24). They also seek protection in coconut husks (hence the clever name!) and collect other shells to use as protection, and, as seen at 3:33, launching a shrapnel barrage at an enemy. There is also some exciting shrimp-hunting action and color changing in the video. Plus ocean pollution in the form of a paint can lid the octopus uses as a shield! I’m telling you, this video has it all. Except topless mermaids and pirate treasure. And, with a full one point deduction for each, it’s an 8/10 for me.

An anonymous reader quotes a report from Reuters: The hack of Jaguar Land Rover, owned by India’s Tata Motors, cost the British economy an estimated $2.55 billion and affected over 5,000 organizations, an independent cybersecurity body said in a report published on Wednesday. The report was produced by the Cyber Monitoring Centre, an independent, not for profit organization made up of industry specialists, including the former head of Britain’s National Cyber Security Centre. It said losses could be higher if there were unexpected delays to the restoration of production at the vehicle manufacturer to levels before the hack took place in August.

“This incident appears to be the most economically damaging cyber event to hit the UK, with the vast majority of the financial impact being due to the loss of manufacturing output at JLR and its suppliers,” the report said. JLR will report its financial results in November, according to the company’s website. A spokesperson for JLR declined to comment on the report. […] JLR, which analysts estimated was losing around 50 million pounds per week from the shutdown, was provided with a 1.5 billion pound loan guarantee by the British government in late September to help it support suppliers.

Read more of this story at Slashdot.

OpenAI has officially entered the browser wars. On Tuesday, the company announced Atlas, a new web browser with ChatGPT integration. At the moment, it’s Mac-only, but I wouldn’t recommend even my Apple friends jumping on board immediately—at least not without understanding the underlying risks.

Atlas’ AI web browsing

If you’ve already used other AI browsers, like Perplexity’s Comet, Atlas is going to feel familiar. In fact, that’s also likely true if you’ve used any web browser before: Atlas is built on Chromium, the engine that powers browsers like Google Chrome, Microsoft Edge, and Opera. That means the core mechanics of Atlas are fairly standard; there’s nothing particularly revolutionary happening here when it comes to sorting tabs or the browsing experience itself.

The same is true when it comes to some ChatGPT interactions. As with other AI browsers, ChatGPT is assigned to the sidebar of the browser window. You can call it up by clicking on the “Ask ChatGPT” button, where you can ask it questions about the content you’re currently browsing. You can also ask ChatGPT for writing assistance any time you enter an open text field in the browser.

Like Comet, Atlas has an agent mode, but the latter’s is built off the existing ChatGPT agent. The idea is that you can task Atlas with performing functions on your behalf. So, rather than pulling up DoorDash’s website and ordering yourself dinner, you could ask Atlas to order dinner for you. You can even watch Atlas get to work, and see its thinking behind each decision. OpenAI has other ideas for how to use Atlas’ agent mode, including giving the browser a recipe to shop for, or asking the bot to run through team documents at work to generate a brief.

Deeper ChatGPT integration is what might set Atlas apart from the competition. If you’re a regular ChatGPT user, you’ll probably appreciate it having that contextual awareness of your past conversations—if you’ve already asked ChatGPT about a topic, and you’re currently researching it in a browser window, you can pick up the conversation and assume ChatGPT will remember what you’ve already talked about.

Credit: OpenAi

Similarly, Atlas will track your browsing and activity history and call upon it in future sessions. Perhaps you’ll open your browser to find personalized suggestions on which sites and topics to explore next. Does that sound creepy? Absolutely. But if you’re someone who doesn’t mind the privacy trade-off, there could be some benefits there. In OpenAI’s announcement, it suggested asking Atlas to pull up all the job postings you looked at last week, and produce a summary of industry trends to use in interview prep. If you find that these memories are a bit too much for you, you can disable them from the browser settings. (OpenAI says deleting your browsing history also deletes associated browser memories, and the browser’s incognito window logs you out of ChatGPT.)

The company includes a setting called “ChatGPT page visibility,” which lets you control whether ChatGPT can actually see the webpage you’re visiting. If you choose “Not Allowed,” you can block the bot from seeing what you’re doing, which is good. But then again, doing that defeats the purpose of Atlas a bit. If you don’t want ChatGPT seeing what you’re doing, you might as well use a browser that doesn’t have ChatGPT built right into it. (The company does promise it won’t train ChatGPT on your browsing data unless you opt into it, but why would you do that?)

Is Atlas safe to use?

Credit: OpenAI

I’m of the opinion that if the safety of a browser is in question, it’s best not to bother with. That’s the case with Atlas, as well as other AI browsers.

The main issue with browsers that incorporate AI agents is that they are susceptible to indirect prompt injection attacks. Brave has done a lot of research on this subject, particularly with Comet. In short, bad actors can potentially hide malicious instructions on websites that the AI agents see as no different from a typical user request. Because the browser is designed to act on your behalf, these malicious instructions can command the AI to do things you definitely don’t want it to do. You might ask Atlas to summarize a webpage, but because a bad actor hid a command to do something involving your email, bank account, or corporate intranet on the site, it does that instead.

To OpenAI’s credit, the company has compiled a list of safeguards to mitigate risks with Atlas. Atlas cannot run code directly in the browser, nor can it download files or install extensions. The browser has no way to access other apps on your Mac, nor its file system. If agent mode needs to access sensitive sites, like your bank, it’ll pause to make sure “you’re watching.” To that point, you can use Atlas’ agent in logged out mode, which limits its ability to access sensitive data or take actions “as you” on websites. But even OpenAI admits that after thousands of hours of testing, their safeguards “will not stop every attack that emerges as AI agents grow in popularity.” The company says it’ll patch new vulnerabilities as it finds them, but if bad actors find them first, they might trick these AI agents into doing some terrible things.

To me, the risks currently far outweigh the benefits. I don’t yet see much reason to have a bot in my browser do things on my behalf, but even if I did, I wouldn’t use it just yet. The risk that someone injects a website with a malicious command and derails my AI agent—and my digital life—is too great, especially when I can book those flights or order that delivery on my own without issue.

Reddit is suing companies SerApi, OxyLabs, AWMProxy and Perplexity for allegedly scraping its data from search results and using it without a license, The New York Times reports. The new lawsuit follows legal action against AI startup Anthropic, who allegedly used Reddit content to train its Claude chatbot.

 As of 2023, Reddit charges companies looking access to posts and other content in the hopes of making money on data that could be used for AI training. The company has also signed licensing deals with companies like Google and OpenAI, and even built an AI answer machine of its own to leverage the knowledge in users’ posts. Scraping search results for Reddit content avoids those payments, which is why the company is seeking financial damages and a permanent injunction that prevents companies from selling previously scraped Reddit material.

Some of the companies Reddit is focused on, like SerApi, OxyLabs and AWMProxy, are not exactly household names, but they’ve all made collecting data from search results and selling it a key part of their business. Perplexity’s inclusion in the lawsuit might be more obvious. The AI company needs data to train its models, and has already been caught seemingly copying and regurgitating material it hasn’t paid to license. That also includes reportedly ignoring the robots.txt protocol, a way for websites to communicate that they don’t want their material scraped.

Per a copy of the lawsuit provided to Engadget, Reddit had already sent a cease-and-desist to Perplexity asking it to stop scraping posts without a license. The company claimed it didn’t use Reddit data, but it also continued to cite the platform in answers from its chatbot. Reddit says it was able to prove Perplexity was using scraped Reddit content by creating a “test post” that “could only be crawled by Google’s search engine and was not otherwise accessible anywhere on the internet.” Within a few hours, queries made to Perplexity’s answer engine were able to reproduce the content of the post.

“The only way that Perplexity could have obtained that Reddit content and then used it in its ‘answer engine’ is if it and/or its co-defendants scraped Google [search results] for that Reddit content and Perplexity then quickly incorporated that data into its answer engine,” the lawsuit claims.

When asked to comment, Perplexity provided the following statement:

Perplexity has not yet received the lawsuit, but we will always fight vigorously for users’ rights to freely and fairly access public knowledge. Our approach remains principled and responsible as we provide factual answers with accurate AI, and we will not tolerate threats against openness and the public interest.

This new lawsuit fits with the aggressive stance Reddit has taken towards protecting its data, including rate-limiting unknown bots and web crawlers in 2024, and even limiting what access the Internet Archive’s Wayback Machine has to its site in August 2025. The company has also sought to define new terms around how websites are crawled by adopting the Really Simple Licensing standard, which adds licensing terms to robots.txt.

This article originally appeared on Engadget at https://www.engadget.com/social-media/reddit-sues-perplexity-and-three-other-companies-for-allegedly-using-its-content-without-paying-205136436.html?src=rss

Here’s a job pitch you don’t see often.

What if, instead of “work-life balance,” you had no balance at all—your life was your work… and work happened seven days a week?

Did I say days? I actually meant days and nights, because the job I’m talking about wants you to know that you will also work weekends and evenings, and that “it’s ok to send messages at 3am.”

Read full article

Comments

Coal use hit a record high around the world last year despite efforts to switch to clean energy, imperilling the world’s attempts to rein in global heating. From a report: The share of coal in electricity generation dropped as renewable energy surged ahead. But the general increase in power demand meant that more coal was used overall, according to the annual State of Climate Action report, published on Wednesday. The report painted a grim picture of the world’s chances of avoiding increasingly severe impacts from the climate crisis. Countries are falling behind the targets they have set for reducing greenhouse gas emissions, which have continued to rise, albeit at a lower rate than before.

Clea Schumer, a research associate at the World Resources Institute thinktank, which led the report, said: “There’s no doubt that we are largely doing the right things. We are just not moving fast enough. One of the most concerning findings from our assessment is that for the fifth report in our series in a row, efforts to phase out coal are well off track.”

Read more of this story at Slashdot.

We may earn a commission from links on this page.

If you’ve been waiting to get your hands on a Samsung Galaxy XR virtual reality headset, now’s your moment: The device is available to order now on Samsung’s or Google’s website for $1,799 and orders are scheduled to arrive Nov. 4. To offset the price, the Galaxy XR comes with a free year of Google AI Pro, YouTube Premium, and Google Play Pass.

Built in partnership with Google and Qualcomm, the Galaxy XR features dual, 4K micro-OLED screens delivering 27 million pixels total and 100-degree horizontal field of view, all powered by Qualcomm Snapdragon XR2+ Gen 2 chip and packed into a lightweight (1.2 pounds) form factor.

The birth of Android XR

The headset will be powered by the new Android XR platform, which will allow users to run standard Android apps as 2D windows. The new ecosystem is designed around system-level integration of Gemini AI and is planned as an infrastructure for future AR and VR devices, including smart glasses. A couple highlights from the operating system: the ability to circle things in the real world and search for them online, and the AI conversion of 2D photos and videos into immersive 3D presentations.

Samsung’s aiming for the space between Apple and Meta

Samsung’s Galaxy XR is stepping into a fairly crowded marketplace, given the public’s relatively tepid reaction to virtual reality. Its main competitors are Meta’s Quest line of headsets and the Apple Vision Pro, although both Meta and Apple are putting much of their AR/VR resources into smart glasses instead of headsets.

The Galaxy XR’s $1,799 price tag is almost exactly half the cost of an Apple Vision Pro and more than three times as expensive as a Meta Quest 3. Samsung seems to be aiming at a tech middle ground: Not prohibitively expense to nearly everyone, but well above the Quest’s “game console” price. The target seems to be pro-level consumers interested in a headset with better hardware than a Quest, but who balk at a $3,500 device. In other words: Samsung is chasing Apple’s polish and Meta’s accessibility, hoping to find a sweet spot between them.

In terms of specs, the Apple Vision Pro’s M5 chip and dedicated R1 spatial processor put it in the lead in terms of raw processing power, though the Galaxy displays 27 million pixels compared to Apple’s 24 million. The Quest 3 displays around 9 million pixels. The Galaxy has a slower top refresh rate compared with Apple, too: 90Hz vs. 12Hz. Ultimately the more expensive headsets are roughly comparable, assuming everything works the way it should; I haven’t tried the Galaxy XR yet, so I can’t say. As for the Quest 3, its very low price isn’t the only advantage to the cheaper system: The Quest’s software library is orders of magnitude larger than the competition.

If Samsung can back up its impressive specs with software and comfort, the XR could give both Apple and Meta something to worry about, but whether any VR headset could generate the kind of enthusiasm the public has for smart glasses remains to be seen.

General Motors held a preview event today to show the world what it’s working on. We’ve already seen some projects, like the further development of lithium manganese-rich battery technology or backup power for EVs that can power a home or support the power grid.

The most significant new announcement is that Cadillac will offer an Escalade IQ with a so-called “Level 3” conditional automated driving system in 2028. GM is referring to it as a “hands off, eyes off” system and says it will integrate advanced digital mapping, use of lidar and other systems, and advanced machine learning to handle the driving duties in a controlled environment up to 80 mph (129 km/h).

This means you can theoretically watch a movie from the driver’s seat while your car takes you down the highway to the airport. Over time, the system’s operation areas will expand to cover even more roads, making driving unnecessary in many situations—unless, of course, you like to drive.

Read full article

Comments

This is a video of a man spinning himself into the Matrix on a piece of playground equipment. We didn’t have that particular apparatus on my school playground growing up. We had seesaws, swings, a jungle gym, a metal slide that squeaked the skin right off your legs as you went down, and a merry-go-round. And I guarantee if you could mod a metal detector to detect teeth, the area around that merry-go-round would be a gold mine. Anyways, after about three seconds of trying to centrifuge all his bodily fluids out his pores this guy doesn’t even look real anymore. I assume that’s when he entered the Matrix. What he saw there is anybody’s guess, because after that much spinning there’s no way this man will ever be able to put another sentence together. I do like how he performed a brief reverse spin at the end though like it would undo all the damage he just did. Doesn’t hurt to try.

Thanks to JustA, who knows what I like, and people doing dumb stuff so I feel less bad about the dumb stuff I do is it.