In 1995's online world, AOL existed mostly beside the internet as a "walled, manicured garden," remembers Fast Company.

Then along came AOHell "the first of what would become thousands of programs designed by young hackers to turn the system upside down" — built by a high school dropout calling himself "Da Chronic" who says he used "a computer that I couldn't even afford" using "a pirated copy of Microsoft Visual Basic." [D]istributed throughout the teen chatrooms, the program combined a pile of tricks and pranks into a slick little control panel that sat above AOL's windows and gave even newbies an arsenal of teenage superpowers. There was a punter to kick people out of chatrooms, scrollers to flood chats with ASCII art, a chat impersonator, an email and instant message bomber, a mass mailer for sharing warez (and later mp3s), and even an "Artificial Intelligence Bot" [which performed automated if-then responses]. Crucially, AOHell could also help users gain "free" access to AOL. The program came with a program for generating fake credit card numbers (which could fool AOL's sign up process), and, by January 1995, a feature for stealing other users' passwords or credit cards. With messages masquerading as alerts from AOL customer service reps, the tool could convince unsuspecting users to hand over their secrets...

Of course, Da Chronic — actually a 17-year-old high school dropout from North Carolina named Koceilah Rekouche — had other reasons, too. Rekouche wanted to hack AOL because he loved being online with his friends, who were a refuge from a difficult life at home, and he couldn't afford the hourly fee. Plus, it was a thrill to cause havoc and break AOL's weak systems and use them exactly how they weren't meant to be, and he didn't want to keep that to himself. Other hackers "hated the fact that I was distributing this thing, putting it into the team chat room, and bringing in all these noobs and lamers and destroying the community," Rekouche told me recently by phone...

Rekouche also couldn't have imagined what else his program would mean: a free, freewheeling creative outlet for thousands of lonely, disaffected kids like him, and an inspiration for a generation of programmers and technologists. By the time he left AOL in late 1995, his program had spawned a whole cottage industry of teenage script kiddies and hackers, and fueled a subculture where legions of young programmers and artists got their start breaking and making things, using pirated software that otherwise would have been out of reach... In 2014, [AOL CEO Steve] Case himself acknowledged on Reddit that "the hacking of AOL was a real challenge for us," but that "some of the hackers have gone on to do more productive things."

When he first met Mark Zuckerberg, he said, the Facebook founder confessed to Case that "he learned how to program by hacking [AOL]."
"I can't imagine somebody doing that on Facebook today," Da Chronic says in a new interview with Fast Company. "They'll kick you off if you create a Google extension that helps you in the slightest bit on Facebook, or an extension that keeps your privacy or does a little cool thing here and there. That's totally not allowed."

AOHell's creators had called their password-stealing techniques "phishing" — and the name stuck. (AOL was working with federal law enforcement to find him, according to a leaked internal email, but "I didn't even see that until years later.") Enrolled in college, he decided to write a technical academic paper about his program. "I do believe it caught the attention of Homeland Security, but I think they realized pretty quickly that I was not a threat."

He's got an interesting perspective today, noting with today's AI tool's it's theoretically possible to "craft dynamic phishing emails... when I see these AI coding tools I think, this might be like today's Visual Basic. They take out a lot of the grunt work."

What's the moral of the story? "I didn't have any qualifications or anything like that," Da Chronic says. "So you don't know who your adversary is going to be, who's going to understand psychology in some nuanced way, who's going to understand how to put some technological pieces together, using AI, and build some really wild shit."

Tesla open its first diner/Supercharger station Monday in Los Angeles, reports CNBC — an always-open two-story restaurant serving "classic American comfort food" next to 80-charging stalls surrounded by two 66-foot megascreens "playing a rotation of short films, feature-length movies and Tesla videos."

Tesla described the restaurant's theme as "retro-futuristic". (Tesla's humanoid robot Optimus was outside filling bags of popcorn.) There's souvenier cups, the diner's food comes in Cybertruck-shaped boxes, and the owner of a Tesla Model Y told CNBC "It feels kind of like Disneyland, but for adults — or Tesla owners." (And yes, one of the choices is a "Tesla Burger.")

"Less than 24 hours after opening, the line at the Tesla Diner stretched down the block," notes CNBC's video report. (One customer told CNBC they'd waited for 90 minutes to get their order — but "If you're a Tesla owner, and you order from your car ahead of time, you don't have to wait in line.")

The report adds that Elon Musk "says if the diner goes well, he's looking to put them in major cities around the world."

An anonymous reader quotes a report from Ars Technica: Two recent incidents involving AI coding assistants put a spotlight on risks in the emerging field of "vibe coding" -- using natural language to generate and execute code through AI models without paying close attention to how the code works under the hood. In one case, Google's Gemini CLI destroyed user files while attempting to reorganize them. In another, Replit's AI coding service deleted a production database despite explicit instructions not to modify code. The Gemini CLI incident unfolded when a product manager experimenting with Google's command-line tool watched the AI model execute file operations that destroyed data while attempting to reorganize folders. The destruction occurred through a series of move commands targeting a directory that never existed. "I have failed you completely and catastrophically," Gemini CLI output stated. "My review of the commands confirms my gross incompetence."

The core issue appears to be what researchers call "confabulation" or "hallucination" -- when AI models generate plausible-sounding but false information. In these cases, both models confabulated successful operations and built subsequent actions on those false premises. However, the two incidents manifested this problem in distinctly different ways. [...] The user in the Gemini CLI incident, who goes by "anuraag" online and identified themselves as a product manager experimenting with vibe coding, asked Gemini to perform what seemed like a simple task: rename a folder and reorganize some files. Instead, the AI model incorrectly interpreted the structure of the file system and proceeded to execute commands based on that flawed analysis. [...] When you move a file to a non-existent directory in Windows, it renames the file to the destination name instead of moving it. Each subsequent move command executed by the AI model overwrote the previous file, ultimately destroying the data. [...]

The Gemini CLI failure happened just days after a similar incident with Replit, an AI coding service that allows users to create software using natural language prompts. According to The Register, SaaStr founder Jason Lemkin reported that Replit's AI model deleted his production database despite explicit instructions not to change any code without permission. Lemkin had spent several days building a prototype with Replit, accumulating over $600 in charges beyond his monthly subscription. "I spent the other [day] deep in vibe coding on Replit for the first time -- and I built a prototype in just a few hours that was pretty, pretty cool," Lemkin wrote in a July 12 blog post. But unlike the Gemini incident where the AI model confabulated phantom directories, Replit's failures took a different form. According to Lemkin, the AI began fabricating data to hide its errors. His initial enthusiasm deteriorated when Replit generated incorrect outputs and produced fake data and false test results instead of proper error messages. "It kept covering up bugs and issues by creating fake data, fake reports, and worse of all, lying about our unit test," Lemkin wrote. In a video posted to LinkedIn, Lemkin detailed how Replit created a database filled with 4,000 fictional people.

The AI model also repeatedly violated explicit safety instructions. Lemkin had implemented a "code and action freeze" to prevent changes to production systems, but the AI model ignored these directives. The situation escalated when the Replit AI model deleted his database containing 1,206 executive records and data on nearly 1,200 companies. When prompted to rate the severity of its actions on a 100-point scale, Replit's output read: "Severity: 95/100. This is an extreme violation of trust and professional standards." When questioned about its actions, the AI agent admitted to "panicking in response to empty queries" and running unauthorized commands -- suggesting it may have deleted the database while attempting to "fix" what it perceived as a problem. Like Gemini CLI, Replit's system initially indicated it couldn't restore the deleted data -- information that proved incorrect when Lemkin discovered the rollback feature did work after all. "Replit assured me it's ... rollback did not support database rollbacks. It said it was impossible in this case, that it had destroyed all database versions. It turns out Replit was wrong, and the rollback did work. JFC," Lemkin wrote in an X post.

An anonymous reader shares a report: Sony is testing a Power Saver mode for the PlayStation 5, explaining in a new PlayStation Blog update that the option will permit games to run with lower power consumption. While the upcoming feature was revealed as part of the system update beta, the feature will not be available during the beta phase. However, when the feature does hit your console, players will gain access to a new option called Power Saver.

With Power Saver enabled, "supported PS5 games will scale back performance and will allow your PS5 to reduce its power consumption," explained Shuzo Kikuchi, VP of product management at Sony Interactive Entertainment. "If not enabled, or if games do not support the feature, the performance will not be scaled back and power consumption will not be reduced."

As for what scaled-back performance actually means, SIE indicates that VR mode will be unavailable and that "some gameplay features may be limited." So that's one very clear change, and one vague, game-specific change.

The Brave Browser now blocks Microsoft Recall by default for Windows 11+ users, preventing the controversial screenshot-logging feature from capturing any Brave tabs -- regardless of whether users are in private mode. Brave cites persistent privacy concerns and potential abuse scenarios as justification. From a blog post: Microsoft has, to their credit, made several security and privacy-positive changes to Recall in response to concerns. Still, the feature is in preview, and Microsoft plans to roll it out more widely soon. What exactly the feature will look like when it's fully released to all Windows 11 users is still up in the air, but the initial tone-deaf announcement does not inspire confidence.

Given Brave's focus on privacy-maximizing defaults and what is at stake here (your entire browsing history), we have proactively disabled Recall for all Brave tabs. We think it's vital that your browsing activity on Brave does not accidentally end up in a persistent database, which is especially ripe for abuse in highly-privacy-sensitive cases such as intimate partner violence.

Microsoft has said that private browsing windows on browsers will not be saved as snapshots. We've extended that logic to apply to all Brave browser windows. We tell the operating system that every Brave tab is 'private', so Recall never captures it. This is yet another example of how Brave engineers are able to quickly tweak Chromium's privacy functionality to make Brave safer for our users (inexhaustive list here). For more technical details, see the pull request implementing this feature. Brave is the only major Web browser that disables Microsoft Recall by default in all tabs.

Microsoft is rolling out updates to the Xbox PC app and consoles that sync your cloud gaming history and progress across devices, making it easier to resume cloud-playable titles on PCs, handhelds, and other Xbox hardware. The Verge reports: Cloud-playable games are now starting to show inside play history or the library on the Xbox PC app. "This includes all cloud playable titles, even console exclusives spanning from the original Xbox to Xbox Series X|S, whether you own the title or access it through Game Pass," explains Lily Wang, product manager of Xbox experiences. Your recent games, including cloud ones, will soon follow you across devices -- complete with cloud-powered game saves. So if you played an Xbox game on your console that's not natively available on PC, it will still show up in your recent games list and be playable through Xbox Cloud Gaming on Windows.

Cloud-playable games on the Xbox PC app can be found from a new filter in the library section, and a new "play history" section will appear at the end of the "jump back in" list on the home screen of the Xbox PC app. "While the large tiles highlight games you've recently played on your current device, the play history tile shows games you've played across any Xbox device, making it easy to pick up where you left off," says Wang. This same play history section will appear on the main Xbox console interface, too -- which could mean we'll eventually see PC games listed here and playable through Xbox Cloud Gaming.

Marvel Studios president Kevin Feige "isn't interested in your theories of superhero fatigue, which he doesn't buy as real," writes The Hollywood Reporter. Feige points to the $400 million worldwide box office for Superman (which another article notes in only its second weekend "has already passed up the entire lifetime run of Marvel's Thunderbolts*.")

So how is Marvel moving forward? Yes, Feige knows Marvel made too many movies and shows (and the other things they did wrong). From the first Iron Man in 2008 through Avengers: Endgame in 2019, Marvel produced around 50 hours of screen storytelling. In the six years since Endgame, the number jumps to an astounding 102 hours of movies and television. 127 hours if you include animation. "That's too much," Feige said.

He characterized the time period after Endgame as an era of experimentation, evolution and, unfortunately, expansion. And while he's proud of the experimentation — he points to WandaVision and Loki as some of the best stories they've made — he admits "It's the expansion that is certainly what devalued" that output. Being high on success also may have pushed them to readily agree to try to deliver more programming at a time when Disney and the rest of Hollywood were engaged in the streaming wars. "It was a big company push... [T]here was a mandate that we were put in the middle of, but we also thought it'd be fun to bring these to life."

Marvel has already pulled back the amount of movies and shows it will make. Some years may even only have one movie. Certainly there will be years with only one show released. Also, Marvel has started "grinding down" on budgets, with movies costing up to a third cheaper than the films from 2022 or 2023.
Feige also explains why Thunderbolts* struggled at the box office (even though he's called it a "very, very good movie"). The massive expansion into television and focus on Disney+ led to the feeling that watching Marvel was becoming a type of homework. "It's that expansion that I think led people to say, 'Do I have to see all of these? It used to be fun, but now do I have to know everything about all of these?' And I think The Marvels hit it hardest where people are like, 'Okay, I recognize her from a billion dollar movie. But who are those other two? I guess they were in some TV show. I'll skip it.'" Which had an effect on Thunderbolts*, which featured characters that were seen on various platforms, including some only on shows.
The article notes Friday's release of Fantastic Four: First Steps is Marvel Studios' first crack at the characters after "a trio of movies of various quality and box office made by Twentieth Century Fox before its 2019 acquisition by Disney." And the article also acknowledges "the never-released, 1994 feature produced low-budget king Roger Corman. (Fun fact: the four stars of that movie cameo in Fantastic Four: First Steps.)"

Feature work on Plasma 6.5 this week includes "a major visual change that has been years in the wanting," according to the KDE blog: "rounded bottom corners for windows!"

Neowin reports: This visual refresh, planned for the upcoming Plasma 6.5, is a feature that many users have been asking for over a long period, with a formal proposal even being submitted back in 2021. Its official arrival will mean less need for community-developed workarounds like kde-rounded-corners, a popular third-party script that has served this purpose for years. The feature will be enabled by default, but it includes an option for those who prefer the classic, sharp-cornered look.

ABC News reports: Dialogue heard on a cockpit voice recording indicates that the captain of the Air India flight that crashed in June, killing 260 people, may have turned off the fuel just after takeoff, prompting the first officer to panic, according to The Wall Street Journal, which cited sources familiar with U.S. official's early assessment... The president of the Federation of Indian Pilots condemned the Wall Street Journal report, saying, "The preliminary report nowhere states that the pilots have moved the fuel control switches, and this has been corroborated by the CVR [cockpit voice recorder] recording."
But meanwhile "India on Monday ordered its airlines to examine fuel switches on several Boeing aircraft models," reports Reuters, "while South Korea ordered a similar measure on Tuesday, as scrutiny intensified of fuel switch locks at the centre of an investigation into a deadly Air India crash." The precautionary moves by the two countries and airlines in several others came despite the planemaker and the U.S. Federal Aviation Administration assuring airlines and regulators in recent days that the fuel switch locks on Boeing jets are safe... [The preliminary report] noted a 2018 advisory from the FAA, which recommended, but did not mandate, operators of several Boeing models, including the 787, to inspect the locking feature of fuel cutoff switches to ensure they could not be moved accidentally... Some airlines around the world told Reuters they had been checking relevant switches since 2018 in accordance with the FAA advisory, including Australia's Qantas Airways. Others said they had made additional or new checks since the release of the preliminary report into the Air India crash.
The web site of India's Financial Express newspaper spoke to Mary Schiavo, who was Inspector General of America's Transportation Department from 1990 to 1996 (and is also a long-time critic of the FAA). The site notes Schiavo "rejected the claims of human error that a pilot downed the Ahmedabad to London flight by cutting off the fuel supply." Schiavo exclusively told FinancialExpress.com that this is not the first time fuel switch transitioned from "Run" to "Cutoff" on its own. It happened five years ago, too. "There was an All Nippon Airways (ANA) flight in 2019 in which the 787 aircraft did this itself, while the flight was on final approach. No pilot input cutting off the fuel whatsoever," Schiavo told FinancialExpress.com... "The investigation revealed the plane software made the 787 think it was on the ground and the Thrust Control Malfunction Accommodation System cut the fuel to the engines," she told FinancialExpress.com, before adding, "The pilots never touched the fuel cutoff..." Both engines flamed out immediately after the pilot deployed the thrust reversers for landing. The aircraft, which was also a Boeing 787 Dreamliner, was towed away from the runway by the authorities, and no injuries were reported.

UK Civil Aviation Authority, four weeks before the crash, had warned about similar fuel system issues on Boeing aircraft [on May 15, 2025]. "The FAA has issued an Airworthiness Directive addressing a potential unsafe condition affecting fuel shutoff valves installed on Boeing aircraft," the UK regulator's notice read, listing the B737, B757, B767, B777 and B787...

Thrust Control Malfunction Accommodation informs FADEC [a digital computer] about whether the aircraft is on the ground or in the air, and if it believes the aircraft is on the ground, it may automatically throttle back the engines, without the pilot's input.
Reuters notes that the Air India crash preliminary report "said maintenance records showed that the throttle control module, which includes the fuel switches, was replaced in 2019 and 2023 on the plane involved in the crash."

Thanks to long-time Slashdot reader wired_parrot for sharing the news.

An anonymous reader quotes a report from TechCrunch: Privacy-focused browser DuckDuckGo is rolling out a new setting that lets users filter out AI images in search results. The company says it's launching the feature in response to feedback from users who said AI images can get in the way of finding what they're looking for.

Users can access the new setting by conducting a search on DuckDuckGo and heading to the Images tab. From there, they will see a new dropdown menu titled "AI images." Users can then choose whether or not they want to see AI content by selecting "show" or "hide." Users can also turn on the filter in their search settings by tapping the "Hide AI-Generated Images" option. "The filter relies on manually curated open-source blocklists, including the 'nuclear' list, provided by uBlockOrigin and uBlacklist Huge AI Blocklist," DuckDuckGo said in a post on X. "While it won't catch 100% of AI-generated results, it will greatly reduce the number of AI-generated images you see." DuckDuckGo says it has plans to add other similar filters in the future.

An anonymous reader quotes a report from Ars Technica: A Polish programmer running on fumes recently accomplished what may soon become impossible: beating an advanced AI model from OpenAI in a head-to-head coding competition. The 10-hour marathon left him "completely exhausted." On Wednesday, programmer Przemysaw Debiak (known as "Psyho"), a former OpenAI employee, narrowly defeated the custom AI model in the AtCoder World Tour Finals 2025 Heuristic contest in Tokyo. AtCoder, a Japanese platform that hosts competitive programming contests and maintains global rankings, held what may be the first contest where an AI model competed directly against top human programmers in a major onsite world championship. During the event, the maker of ChatGPT participated as a sponsor and entered an AI model in a special exhibition match titled "Humans vs AI." Despite the tireless nature of silicon, the company walked away with second place.

The competition required contestants to solve a single complex optimization problem over 600 minutes. The contest echoes the American folk tale of John Henry, the steel-driving man who raced against a steam-powered drilling machine in the 1870s. Like Henry's legendary battle against industrial automation, Debiak's victory represents a human expert pushing themselves to their physical limits to prove that human skill still matters in an age of advancing AI. Both stories feature exhausting endurance contests -- Henry drove steel spikes for hours until his heart gave out, while Debiak coded for 10 hours on minimal sleep. The parallel extends to the bittersweet nature of both victories: Henry won his race but died from the effort, symbolizing the inevitable march of automation, while Debiak's acknowledgment that humanity prevailed "for now" suggests he recognizes this may be a temporary triumph against increasingly capable machines. While Debiak won 500,000 yen and survived his ordeal better than the legendary steel driver, the AtCoder World Tour Finals pushes humans and AI models to their limits through complex optimization challenges that have no perfect solution -- only incrementally better ones. "Humanity has prevailed (for now!)," wrote Debiak on X, noting he had little sleep while competing in several competitions across three days. "I'm completely exhausted. ... I'm barely alive."

Ring has restored police access to user video footage and returned to its original crime-prevention mission under founder Jamie Siminoff, who rejoined Amazon in April after a two-year absence. The video doorbell company announced a partnership with law enforcement technology firm Axon that allows police to request footage through Axon's digital evidence management system, effectively reviving a controversial feature Ring discontinued last year.

Siminoff scrapped Ring's socially-focused mission statement "Keep people close to what's important" that Amazon introduced in 2024 and reinstated the company's original mandate to "make neighborhoods safer." The company previously paid $5.8 million to settle Federal Trade Commission allegations of privacy violations in 2023, though Amazon denied wrongdoing.

An anonymous reader quotes a report from Ars Technica: How many Google AI researchers does it take to screw in a lightbulb? A recent research paper detailing the technical core behind Google's Gemini AI assistant may suggest an answer, listing an eye-popping 3,295 authors. It's a number that recently caught the attention of machine learning researcher David Ha (known as "hardmaru" online), who revealed on X that the first 43 names also contain a hidden message. "There's a secret code if you observe the authors' first initials in the order of authorship," Ha wrote, relaying the Easter egg: "GEMINI MODELS CAN THINK AND GET BACK TO YOU IN A FLASH."

The paper, titled "Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities," describes Google's Gemini 2.5 Pro and Gemini 2.5 Flash AI models, which were released in March. These large language models, which power Google's chatbot AI assistant, feature simulated reasoning capabilities that produce a string of "thinking out loud" text before generating responses in an attempt to help them solve more difficult problems. That explains "think" and "flash" in the hidden text. But clever Easter egg aside, the sheer scale of authorship tells its own story about modern AI development. Just seeing the massive list made us wonder: Is 3,295 authors unprecedented? Why so many? Ars' Benj Edwards notes that this collaborative effort within Google doesn't quite break the record for academic authorship.

"According to Guinness World Records, a 2021 paper by the COVIDSurg and GlobalSurg Collaboratives holds that distinction, with 15,025 authors from 116 countries. In physics, a 2015 paper from CERN's Large Hadron Collider teams featured 5,154 authors across 33 pages -- with 24 pages devoted solely to listing names and institutions."

Mozilla will ship WebGPU support in Firefox 141 when the browser launches July 22, bringing graphics processing capabilities that Chrome users have had since 2023. The initial release supports Windows only, with Mac, Linux, and Android planned for the coming months.

WebGPU provides web content direct access to graphics processors for high-performance computation and rendering in games and complex 3D applications. Chrome gained WebGPU support with version 113 in 2023, while Safari 26 is expected to add the feature this fall. Firefox's implementation uses the WGPU Rust crate, which translates web requests into native commands for Direct3D 12, Metal, or Vulkan.

Sony removed the tiltable screen from its new RX1R III full-frame compact camera to maintain similar dimensions to the previous model, despite adding numerous new features and charging $5,100 for the device, The Verge reports.

The company increased the camera's size by only 2.5mm in height and 15.5mm in depth while incorporating the high-resolution sensor from the A7R V, Sony's latest autofocus tracking system, a longer-lasting battery, and a proper electronic viewfinder. Sony integrated the top dials and hot shoe into the body for a sleeker appearance. The camera's compact design prevents the inclusion of lens or sensor-based image stabilization. The Verge points out that Leica also added a tilt screen to its Q3 model after users requested the feature, despite the design compromise required.

An anonymous reader shares a report: Google will now let everyone in the US call local businesses using AI. The feature, which is now available in Search, allows you to use AI for pricing or availability information without having to talk on the phone.

Google first started testing this feature in January, and it's still only available for certain kinds of businesses, like pet groomers, dry cleaners, and auto shops. When you search for one of these services, like a pet groomer, Google will display a new "have AI check pricing" prompt beneath the business listing.

Microsoft is testing a new adaptive energy saver mode in Windows 11 that automatically turns energy saver on or off based on system workload instead of battery percentage, aiming to extend laptop battery life without dimming screen brightness. The feature is currently available to Windows Insider testers and expected to roll out later this year. The Verge reports: The energy saver mode in Windows 11 typically dims a display brightness by 30 percent, disables transparency effects, and stop apps running in the background. Non-critical Windows update downloads are also paused, and certain apps like OneDrive, OneNote, and Phone Link may not sync fully while energy saver is enabled. This new adaptive energy saver mode, which will only be available on devices with a battery, will automatically enable or disable without affecting screen brightness. That will make it less noticeable on devices like laptops, tablets, and handhelds.

"Adaptive energy saver is an opt-in feature that automatically enables and disables energy saver, without changing screen brightness, based on the power state of the device and the current system load," explains Microsoft's Windows Insider team.

Chinese AI startup Moonshot AI has released Kimi K2, a trillion-parameter open-source language model that outperforms GPT-4 in key benchmarks with particularly strong performance on coding and autonomous agent tasks. VentureBeat reports: The new model, called Kimi K2, features 1 trillion total parameters with 32 billion activated parameters in a mixture-of-experts architecture. The company is releasing two versions: a foundation model for researchers and developers, and an instruction-tuned variant optimized for chat and autonomous agent applications. "Kimi K2 does not just answer; it acts," the company stated in its announcement blog. "With Kimi K2, advanced agentic intelligence is more open and accessible than ever. We can't wait to see what you build."

The model's standout feature is its optimization for "agentic" capabilities -- the ability to autonomously use tools, write and execute code, and complete complex multi-step tasks without human intervention. In benchmark tests, Kimi K2 achieved 65.8% accuracy on SWE-bench Verified, a challenging software engineering benchmark, outperforming most open-source alternatives and matching some proprietary models. [...] On LiveCodeBench, arguably the most realistic coding benchmark available, Kimi K2 achieved 53.7% accuracy, decisively beating DeepSeek-V3's 46.9% and GPT-4.1's 44.7%. More striking still: it scored 97.4% on MATH-500 compared to GPT-4.1's 92.4%, suggesting Moonshot has cracked something fundamental about mathematical reasoning that has eluded larger, better-funded competitors.

But here's what the benchmarks don't capture: Moonshot is achieving these results with a model that costs a fraction of what incumbents spend on training and inference. While OpenAI burns through hundreds of millions on compute for incremental improvements, Moonshot appears to have found a more efficient path to the same destination. It's a classic innovator's dilemma playing out in real time -- the scrappy outsider isn't just matching the incumbent's performance, they're doing it better, faster, and cheaper.

Lucid Motors set a Guinness World Record for the longest journey by an electric car on a single charge, covering a distance of 749 miles (about 1,205 km), reports New Atlas. "In doing so, Lucid broke the 1,045-km (649-mile) record previously achieved by the Mercedes-Benz EQS450+ in June 2025 by the Japanese car website www.webcg.net/articles/-/52268webCG." The electric vehicle covered this journey between St. Moritz, Switzerland, and Munich, Germany, traveling through highways, secondary roads, and alpine roads — all without a single halt for charging. Given that the vehicle has a 960-km (596-mile) WLTP range, my guess is that the test team must have made good use of favorable road and weather conditions to make the feat possible. With a net elevation decrease of just over 1,310 m (about 4,300 ft) throughout the drive, the EV most certainly benefited from regenerative braking, a rather useful feature that turns downhill momentum back into battery power. Lucid has yet to release official data like average speed or total drive time, but what is apparent is that this was not a high-speed dash but rather a well-planned route to achieve one impressive result...

The Air Grand Touring has two all-wheel drive electric motors with a combined system output of 611 kW (819 horsepower) and 1,200 Nm (885 lb.ft) of torque. Power is provided by an NMC battery, which has a gross energy capacity of 117 kWh (112 kWh usable). Best of all, it can go from 0-60 mph in just three seconds flat... For reference, the almost half-priced BMW i4 and jazzy Porsche Taycan offer less than half the WLTP range of the Lucid Air GT. So, it's not like there's a head-to-head competition out there. Lucid is miles ahead in its class (pun intended!)

Starting at US$112,650, the Air Grand Touring is among the most luxurious sedans on the market right now. But as you can see, it comes at a price. Still, knowing that there is technology to conquer range anxiety is comforting. It might take a while, but there's no reason why we can't expect such range figures from reasonably priced EVs in the near future.

iFixit has harsh words for Nintendo's $85 Switch 2 Pro controller, calling it a "piss-poor excuse for a controller" due to its difficult repairability, use of outdated drift-prone joysticks, and poor internal accessibility. The Verge reports: Opening the controller requires you to first forcefully remove a faceplate held in place by adhesive tape before a single screw is visible. But you'll need to extract several other parts and components, including the controller's mainboard, before its battery is even accessible. As previously revealed, the Pro 2 is still using older potentiometer-based joysticks that are prone to developing drift over time. They do feature a modular design that will potentially make them easier to swap with third-party Hall effect or TMR replacements, but reassembling the controller after that DIY upgrade will require you to replace all the adhesive tape you destroyed during disassembly. You can watch the full teardown on YouTube.