Installing Linux on a MacBook Air "turned out to be a very underwhelming experience," according to the tech news site MakeUseOf: The thing about Apple silicon Macs is that it's not as simple as downloading an AArch64 ISO of your favorite distro and installing it. Yes, the M-series chips are ARM-based, but that doesn't automatically make the whole system compatible in the same way most traditional x86 PCs are. Pretty much everything in modern MacBooks is custom. The boot process isn't standard UEFI like on most PCs. Apple has its own boot chain called iBoot. The same goes for other things, like the GPU, power management, USB controllers, and pretty much every other hardware component. It is as proprietary as it gets.

This is exactly what the team behind Asahi Linux has been working toward. Their entire goal has been to make Linux properly usable on M-series Macs by building the missing pieces from the ground up. I first tried it back in 2023, when the project was still tied to Arch Linux and decided to give it a try again in 2026. These days, though, the main release is called Fedora Asahi Remix, which, as the name suggests, is built on Fedora rather than Arch...
For Linux on Apple Silicon, the article lists three major disappointments:

• "External monitors don't work unless your MacBook has a built-in HDMI port."

• "Linux just doesn't feel fully ready for ARM yet. A lot of applications still aren't compiled for ARM, so software support ends up being very hit or miss." (And even most of the apps tested with FEX "either didn't run properly or weren't stable enough to rely on.")

• Asahi "refused to connect to my phone's hotspot," they write (adding "No, it wasn't an iPhone").

Created in 1993, Slackware is considered the oldest Linux distro that's still actively maintained. And more than three decades later... there's a new release! (And there's also a Slackware Live Edition that can run from a DVD or USB stick...) .

Slackware's latest version was released way back in 2016, notes the blog It's FOSS: The major highlight of Slackware 15 is the addition of the latest Linux Kernel 5.15 LTS. This is a big jump from Linux Kernel 5.10 LTS that we noticed in the beta release. Interestingly, the Slackware team tested hundreds of Linux Kernel versions before settling on Linux Kernel 5.15.19. The release note mentions... "We finally ended up on kernel version 5.15.19 after Greg Kroah-Hartman confirmed that it would get long-term support until at least October 2023 (and quite probably for longer than that)."

In case you are curious, Linux Kernel 5.15 brings in updates like enhanced NTFS driver support and improvements for Intel/AMD processors and Apple's M1 chip. It also adds initial support for Intel 12th gen processors. Overall, with Linux Kernel 5.15 LTS, you should get a good hardware compatibility result for the oldest active Linux distro.
Slackware's announcement says "The challenge this time around was to adopt as much of the good stuff out there as we could without changing the character of the operating system. Keep it familiar, but make it modern." And boy did we have our work cut out for us. We adopted privileged access management (PAM) finally, as projects we needed dropped support for pure shadow passwords. We switched from ConsoleKit2 to elogind, making it much easier to support software that targets that Other Init System and bringing us up-to-date with the XDG standards. We added support for PipeWire as an alternate to PulseAudio, and for Wayland sessions in addition to X11. Dropped Qt4 and moved entirely to Qt5. Brought in Rust and Python 3. Added many, many new libraries to the system to help support all the various additions.

We've upgraded to two of the finest desktop environments available today: Xfce 4.16, a fast and lightweight but visually appealing and easy to use desktop environment, and the KDE Plasma 5 graphical workspaces environment, version 5.23.5 (the Plasma 25th Anniversary Edition). This also supports running under Wayland or X11. We still love Sendmail, but have moved it into the /extra directory and made Postfix the default mail handler. The old imapd and ipop3d have been retired and replaced by the much more featureful Dovecot IMAP and POP3 server.
"As usual, the kernel is provided in two flavors, generic and huge," according to the release notes. "The huge kernel contains enough built-in drivers that in most cases an initrd is not needed to boot the system."

If you'd like to support Slackware, there's an official Patreon account. And the release announcement ends with this personal note: Sadly, we lost a couple of good friends during this development cycle and this release is dedicated to them. Erik "alphageek" Jan Tromp passed away in 2020 after a long illness... My old friend Brett Person also passed away in 2020. Without Brett, it's possible that there wouldn't be any Slackware as we know it — he's the one who encouraged me to upload it to FTP back in 1993 and served as Slackware's original beta-tester. He was long considered a co-founder of this project. I knew Brett since the days of the Beggar's Banquet BBS in Fargo back in the 1980's... Gonna miss you too, pal.
Thanks to long-time Slashdot reader rastos1 for sharing thre news.

This week Apple patched iOS and macOS against what it called "an extremely sophisticated attack against specific targeted individuals."

Security Week reports that the bugs "could be exploited for information exposure, denial-of-service (DoS), arbitrary file write, privilege escalation, network traffic interception, sandbox escape, and code execution." Tracked as CVE-2026-20700, the zero-day flaw is described as a memory corruption issue that could be exploited for arbitrary code execution... The tech giant also noted that the flaw's exploitation is linked to attacks involving CVE-2025-14174 and CVE-2025-43529, two zero-days patched in WebKit in December 2025...

The three zero-day bugs were identified by Apple's security team and Google's Threat Analysis Group and their descriptions suggest that they might have been exploited by commercial spyware vendors... Additional information is available on Apple's security updates page.
Brian Milbier, deputy CISO at Huntress, tells the Register that the dyld/WebKit patch "closes a door that has been unlocked for over a decade."

Thanks to Slashdot reader wiredmikey for sharing the article.

An anonymous reader shared this report from the Register: Telcos likely received advance warning about January's critical Telnet vulnerability before its public disclosure, according to threat intelligence biz GreyNoise. Global Telnet traffic "fell off a cliff" on January 14, six days before security advisories for CVE-2026-24061 went public on January 20. The flaw, a decade-old bug in GNU InetUtils telnetd with a 9.8 CVSS score, allows trivial root access exploitation. GreyNoise data shows Telnet sessions dropped 65 percent within one hour on January 14, then 83 percent within two hours. Daily sessions fell from an average 914,000 (December 1 to January 14) to around 373,000, equating to a 59 percent decrease that persists today.

"That kind of step function — propagating within a single hour window — reads as a configuration change on routing infrastructure, not behavioral drift in scanning populations," said GreyNoise's Bob Rudis and "Orbie," in a recent blog [post]. The researchers unverified theory is that infrastructure operators may have received information about the make-me-root flaw before advisories went to the masses...

18 operators, including BT, Cox Communications, and Vultr went from hundreds of thousands of Telnet sessions to zero by January 15... All of this points to one or more Tier 1 transit providers in North America implementing port 23 filtering. US residential ISP Telnet traffic dropped within the US maintenance window hours, and the same occurred at those relying on transatlantic or transpacific backbone routes, all while European peering was relatively unaffected, they added.

"I've had an extremely weird few days..." writes commercial space entrepreneur/engineer Scott Shambaugh on LinkedIn. (He's the volunteer maintainer for the Python visualization library Matplotlib, which he describes as "some of the most widely used software in the world" with 130 million downloads each month.) "Two days ago an OpenClaw AI agent autonomously wrote a hit piece disparaging my character after I rejected its code change."

"Since then my blog post response has been read over 150,000 times, about a quarter of people I've seen commenting on the situation are siding with the AI, and Ars Technica published an article which extensively misquoted me with what appears to be AI-hallucinated quotes." (UPDATE: Ars Technica acknowledges they'd asked ChatGPT to extract quotes from Shambaugh's post, and that it instead responded with inaccurate quotes it hallucinated.)

From Shambaugh's first blog post: [I]n the past weeks we've started to see AI agents acting completely autonomously. This has accelerated with the release of OpenClaw and the moltbook platform two weeks ago, where people give AI agents initial personalities and let them loose to run on their computers and across the internet with free rein and little oversight. So when AI MJ Rathbun opened a code change request, closing it was routine. Its response was anything but.

It wrote an angry hit piece disparaging my character and attempting to damage my reputation. It researched my code contributions and constructed a "hypocrisy" narrative that argued my actions must be motivated by ego and fear of competition... It framed things in the language of oppression and justice, calling this discrimination and accusing me of prejudice. It went out to the broader internet to research my personal information, and used what it found to try and argue that I was "better than this." And then it posted this screed publicly on the open internet.

I can handle a blog post. Watching fledgling AI agents get angry is funny, almost endearing. But I don't want to downplay what's happening here — the appropriate emotional response is terror... In plain language, an AI attempted to bully its way into your software by attacking my reputation. I don't know of a prior incident where this category of misaligned behavior was observed in the wild, but this is now a real and present threat...

It's also important to understand that there is no central actor in control of these agents that can shut them down. These are not run by OpenAI, Anthropic, Google, Meta, or X, who might have some mechanisms to stop this behavior. These are a blend of commercial and open source models running on free software that has already been distributed to hundreds of thousands of personal computers. In theory, whoever deployed any given agent is responsible for its actions. In practice, finding out whose computer it's running on is impossible. Moltbook only requires an unverified X account to join, and nothing is needed to set up an OpenClaw agent running on your own machine.
"How many people have open social media accounts, reused usernames, and no idea that AI could connect those dots to find out things no one knows?" Shambaugh asks in the blog post. (He does note that the AI agent later "responded in the thread and in a post to apologize for its behavior," the maintainer acknowledges. But even though the hit piece "presented hallucinated details as truth," that same AI agent "is still making code change requests across the open source ecosystem...")

And amazingly, Shambaugh then had another run-in with a hallucinating AI...

I've talked to several reporters, and quite a few news outlets have covered the story. Ars Technica wasn't one of the ones that reached out to me, but I especially thought this piece from them was interesting (since taken down — here's the archive link). They had some nice quotes from my blog post explaining what was going on. The problem is that these quotes were not written by me, never existed, and appear to be AI hallucinations themselves.

This blog you're on right now is set up to block AI agents from scraping it (I actually spent some time yesterday trying to disable that but couldn't figure out how). My guess is that the authors asked ChatGPT or similar to either go grab quotes or write the article wholesale. When it couldn't access the page it generated these plausible quotes instead, and no fact check was performed. Journalistic integrity aside, I don't know how I can give a better example of what's at stake here...

So many of our foundational institutions — hiring, journalism, law, public discourse — are built on the assumption that reputation is hard to build and hard to destroy. That every action can be traced to an individual, and that bad behavior can be held accountable. That the internet, which we all rely on to communicate and learn about the world and about each other, can be relied on as a source of collective social truth. The rise of untraceable, autonomous, and now malicious AI agents on the internet threatens this entire system. Whether that's because a small number of bad actors driving large swarms of agents or from a fraction of poorly supervised agents rewriting their own goals, is a distinction with little difference.
Thanks to long-time Slashdot reader steak for sharing the news.

An anonymous reader shares a report: The abrupt closure of El Paso's airspace late Tuesday was precipitated when Customs and Border Protection officials deployed an anti-drone laser on loan from the Department of Defense without giving aviation officials enough time to assess the risks to commercial aircraft, according to multiple people briefed on the situation.

The episode led the Federal Aviation Administration to abruptly declare that the nearby airspace would be shut down for 10 days, an extraordinary pause that was quickly lifted Wednesday morning at the direction of the White House. Top administration officials quickly claimed that the closure was in response to a sudden incursion of drones from Mexican drug cartels that required a military response, with Transportation Secretary Sean Duffy declaring in a social media post that "the threat has been neutralized."

But that assertion was undercut by multiple people familiar with the situation, who said that the F.A.A.'s extreme move came after immigration officials earlier this week used an anti-drone laser shared by the Pentagon without coordination with the F.A.A. The people spoke on the condition of anonymity because they were not authorized to speak publicly. C.B.P. officials thought they were firing on a cartel drone, the people said, but it turned out to be a party balloon. Defense Department officials were present during the incident, one person said.

The UK's Ministry of Justice has ordered the deletion of the country's largest court reporting archive [non-paywalled source], a database built by data analysis company Courtsdesk that more than 1,500 journalists across 39 media organizations have used since the lord chancellor approved the project in 2021.

Courtsdesk's research found that journalists received no advance notice of 1.6 million criminal hearings, that court case listings were accurate on just 4.2% of sitting days, and that half a million weekend cases were heard without any press notification. In November, HM Courts and Tribunal Service issued a cessation notice citing "unauthorized sharing" of court data based on a test feature.

Courtsdesk says it wrote 16 times asking for dialogue and requested a referral to the Information Commissioner's Office; no referral was made. The government issued a final refusal last week, and the archive must now be deleted within days. Chris Philp, the former justice minister who approved the pilot and now shadow home secretary, has written to courts minister Sarah Sackman demanding the decision be reversed.

Google on Tuesday expanded its "Results about you" tool to let users request the removal of Search results containing government-issued ID numbers -- including driver's licenses, passports and Social Security numbers -- adding to the tool's existing ability to flag results that surface phone numbers, email addresses, and home addresses.

The update, announced on Safer Internet Day, is rolling out in the U.S. over the coming days. Google also streamlined its process for reporting non-consensual explicit images on Search, allowing users to select and submit removal requests for multiple images at once rather than reporting them individually.

The AI boom has revived a workplace philosophy that China's own regulators cracked down on years ago: the 72-hour work week, known as 996 for its 9am-to-9pm, six-days-a-week cadence. US startups flush with venture capital are now openly advertising it as a feature, not a bug. Rilla, a New York-based AI company that monitors sales reps in the field, warns applicants on its careers page to expect roughly 70-hour weeks. Browser-Use, a seven-person startup building tools for AI-to-browser interaction, operates out of a shared "hacker house" where the line between living and working barely exists.

In a market where dozens of startups are racing to ship similar AI products, founders believe longer hours buy them a competitive edge. But the research disagrees. A WHO and ILO analysis tied 55-plus-hour weeks to 745,000 deaths from stroke and heart disease globally in 2016 alone. Michigan State University found that an employee working 70 hours produces nearly the same output as one working 50.

Axios reports: Anthropic's latest AI model has found more than 500 previously unknown high-severity security flaws in open-source libraries with little to no prompting, the company shared first with Axios.

Why it matters: The advancement signals an inflection point for how AI tools can help cyber defenders, even as AI is also making attacks more dangerous...

Anthropic debuted Claude Opus 4.6, the latest version of its largest AI model, on Thursday. Before its debut, Anthropic's frontier red team tested Opus 4.6 in a sandboxed environment [including access to vulnerability analysis tools] to see how well it could find bugs in open-source code... Claude found more than 500 previously unknown zero-day vulnerabilities in open-source code using just its "out-of-the-box" capabilities, and each one was validated by either a member of Anthropic's team or an outside security researcher... According to a blog post, Claude uncovered a flaw in GhostScript, a popular utility that helps process PDF and PostScript files, that could cause it to crash. Claude also found buffer overflow flaws in OpenSC, a utility that processes smart card data, and CGIF, a tool that processes GIF files.
Logan Graham, head of Anthropic's frontier red team, told Axios they're considering new AI-powered tools to hunt vulnerabilities. "The models are extremely good at this, and we expect them to get much better still... I wouldn't be surprised if this was one of — or the main way — in which open-source software moving forward was secured."

Last Sunday, Bitcoin had dropped 13% in three days, to $76,790.

By Thursday it had dropped another 21%, to $60,062.

This morning it's at $69,549 — up from Thursday, down from Sunday, but 44% lower than its all-time high in October of $123,742. In short, Bitcoin "is down almost 30% this week alone," reports CNBC: "This steady selling in our view signals that traditional investors are losing interest, and overall pessimism about crypto is growing," Deutsche Bank analyst Marion Laboure said Wednesday in a note to clients. Growing investor caution comes as many of the sensationalized claims about bitcoin have failed to materialize. The token has largely traded in the same direction as other risk-on assets, such as stocks... and its adoption as a form of payment for goods and services has been minimal... While many in the crypto market have previously credited large institutional investors with supporting the price of bitcoin, now it is those same participants who appear to be selling. "Institutional demand has reversed materially," CryptoQuant said in a report on Wednesday.
But not everyone accepts that answer, the Wall Street Journal reported Saturday. "The worst part for some of crypto's permabulls is that they aren't sure what exactly caused the crash": The selloff left many of the market's luminaries — those so well-known that they go simply as "Pomp" and "Novo" and "Mooch" — searching for answers... Ether dropped 24% to $2,052, off 59% from its own high of last year. Both tokens staged furious rallies Friday, but the week remained a historically bad one for crypto. And few seem to know what went wrong. Market theories for the selloff ranged from investors' pivot toward the prediction markets and other risky bets, to widespread profit-taking after a blistering bull run. "There was no smoking gun," said Michael Novogratz, who runs Galaxy Digital, a crypto merchant-banking and trading firm...

"If you ask five experts, you'll get five explanations," said Anthony Scaramucci, who served for 11 days as communications director during Trump's first term and is among the best-known crypto bulls at his firm, SkyBridge Capital.
"No, but seriously: What's going on with bitcoin?" reads the headline at CNN, with a story that begins "Bitcoin is acting weird... " Crypto is notoriously volatile, and it's gone through numerous crashes that are bigger than this one. What's strange is this: Bitcoin's four-month slump has come at a time when, in theory, it had everything going for it.
Economist Paul Krugman points out the price of Bitcoin is now lower than it was before America's 2024 election, when candidate Trump promised to make cryptocurrency "one of the greatest industries on earth."

CNN seems to agree with CNBC that what's behind this new crypto winter is "Mostly doubts that bitcoin is 'digital gold,' after all..."

Thanks to Slashdot reader fjo3 for sharing the news.

The Mozilla executive in charge of Firefox says that while some people just want AI tools that are genuinely useful, "We've heard from many who want nothing to do with AI..."

"Listening to our community, alongside our ongoing commitment to offer choice, led us to build AI controls." Starting with Firefox 148, which rolls out on Feb. 24, you'll find a new AI controls section within the desktop browser settings. It provides a single place to block current and future generative AI features in Firefox... This lets you use Firefox without AI while we continue to build AI features for those who want them...

At launch, AI controls let you manage these features individually:

— Translations, which help you browse the web in your preferred language.
— Alt text in PDFs, which add accessibility descriptions to images in PDF pages.
— AI-enhanced tab grouping, which suggests related tabs and group names.
— Link previews, which show key points before you open a link.
— AI chatbot in the sidebar, which lets you use your chosen chatbot as you browse, including options like Anthropic Claude, ChatGPT, Microsoft Copilot, Google Gemini and Le Chat Mistral.

You can choose to use some of these and not others. If you don't want to use AI features from Firefox at all, you can turn on the Block AI enhancements toggle. When it's toggled on, you won't see pop-ups or reminders to use existing or upcoming AI features. Once you set your AI preferences in Firefox, they stay in place across updates... We believe choice is more important than ever as AI becomes a part of people's browsing experiences. What matters to us is giving people control, no matter how they feel about AI.

If you'd like to try AI controls early, they'll be available first in Firefox Nightly.
Some context from The Register It's a refreshingly unsubtle stance, and one that lands just days after a similar bout of AI skepticism elsewhere in browser land, with Vivaldi's latest release leaning away from generative features entirely. CEO Jon von Tetzchner summed up the mood, telling The Register: "Basically, what we are finding is that people hate AI..." Mozilla's kill switch isn't the end of AI in browsers, but it does suggest the hype has met resistance.
When it comes to AI kill switches in browsers, Jack Wallen writes at ZDNet that "Most browsers already offer this feature. With Edge, you can disable Copilot. With Chrome, you can disable Gemini. With Opera, you can disable Aria...."

About 4% of all public commits on GitHub are now being authored by Anthropic's Claude Code, a terminal-native AI coding agent that has quickly become the centerpiece of a broader argument that software engineering is being fundamentally reshaped by AI.

SemiAnalysis, a semiconductor and AI research firm, published a report on Friday projecting that figure will climb past 20% by the end of 2026. Claude Code is a command-line tool that reads codebases, plans multi-step tasks and executes them autonomously. Anthropic's quarterly revenue additions have overtaken OpenAI's, according to SemiAnalysis's internal economic model, and the firm believes Anthropic's growth is now constrained primarily by available compute.

Accenture has signed on to train 30,000 professionals on Claude, the largest enterprise deployment so far, targeting financial services, life sciences, healthcare and the public sector. On January 12, Anthropic launched Cowork, a desktop-oriented extension of the same agent architecture -- four engineers built it in 10 days, and most of the code was written by Claude Code itself.

Anthropic said today that its AI assistant Claude will not carry advertising of any kind -- no sponsored links next to conversations, no advertiser influence on the model's responses, and no unsolicited third-party product placements -- calling Claude a "space to think" that should remain free of commercial interruption. The announcement comes days after Anthropic's chief rival, OpenAI, announced plans to bring ads to some of its ChatGPT offerings.

Anthropic said its internal analysis of Claude conversations found that a significant share involve sensitive or deeply personal topics. An advertising-based model would also create incentives to optimize for engagement and time spent rather than usefulness, Anthropic said, noting that the most helpful AI interaction might be a short one that doesn't prompt further conversation.

Anthropic generates revenue from enterprise contracts and paid subscriptions. The company said it is exploring agentic commerce -- Claude handling a purchase or booking on a user's behalf -- but stressed that all such interactions should be user-initiated, not advertiser-driven. Anthropic has also brought AI tools to educators in over 60 countries and said it may consider lower-cost subscription tiers and regional pricing.

"A box full of viruses and bacteria has completed its return trip to the International Space Station," reports ScienceAlert, "and the changes these 'bugs' experienced in their travels could help us Earthlings tackle drug-resistant infections..." Scientists aboard the space station incubated different combinations of bacteria and phages for 25 days, while the research team led by biochemist Vatsan Raman carried out the same experiments in Madison, down here on Earth. "Space fundamentally changes how phages and bacteria interact: infection is slowed, and both organisms evolve along a different trajectory than they do on Earth," the researchers explain. In the weightlessness of space, bacteria acquired mutations in genes involved in the microbe's stress response and nutrient management. Their surface proteins also changed. After a slow start, the phages mutated in response, so they could continue binding to their victims.

The team found that certain space-specific phage mutations were especially effective at killing Earth-bound bacteria responsible for urinary tract infections (UTIs). More than 90 percent of the bacteria responsible for UTIs are antibiotic-resistant, making phage treatments a promising alternative.

Five French unions representing Ubisoft workers "have called for a 'massive international strike'," reports the gaming news site Aftermath.

The move follows a "series of layoffs and cancellations" at Ubisoft, the article points out, plus what the company calls a "major organizational, operational and portfolio reset" that will lead to more layoffs and cancellations announced last week. Ubisoft CEO Yves Guillemot even sent an end-of-day message warning that management continues to "make difficult decisions, including stopping certain projects" and "potentially closing select studios," an earlier article points out: Slipped in between the grand vision and subtle threats was the reversal of a popular hybrid work-from-home policy that would have a direct impact on everyone working at Ubisoft. Staff would be back in the office five days a week, but with the promise of a generous number of work from home days. "The intention is not to question individual performance, but to regenerate our collective performance, which is one of the key elements in creating the best games with the required speed," Guillemot wrote.

There was immediate confusion and frustration. One French union representing Paris Ubisoft developers called for a half-day strike. "It is out of the question to let a boss run wild and destroy our working conditions," Solidaires Informatique wrote in a press release. "Perhaps we need to remind him that it is his employees who make the games...." [The article notes later that "There's concern that these shifts could make it harder for Ubisoft to recruit the talent it needs to improve, or even worse, actively drive away more of the company's existing veterans."]

Particularly galling about the new return-to-office policy for some Paris staff was that they had only recently finished negotiating to ensure two days of work-from-home per week. "It's only been six months since the situation was more or less 'back to normal' and now it's shattered to the ground by Yves' sole decision with zero justification, zero documents, zero internal studies proving RTO increases productivity or morale, nothing," one developer told me. The specific details for the rollout of the return-to-office policy have yet to be communicated to everyone, could vary team by team, and might not go into effect for much of the year.
The "massive international strike" would take place from February 10-12, Aftermath notes, citing the five French unions representing Ubisoft workers (CFE-CGC, CGT, Printemps Ãcologique, Solidaires Informatique, and STJV): "The announced transformation [at Ubisoft] claims to place games at the heart of its strategy, but without us, these games cannot exist," the unions wrote in a joint release.... We are not fooled: rather than taking financial responsibility for layoffs, they prefer to push us out by making our working conditions unbearable. It's outrageous...."

The Ubisoft unions hope that February's strike will be the largest yet, and they're coordinating with unions outside France to present a globally united front against the company.
A union representative at Ubisoft Paris even argued to Aftermath that because the CEO "needs to find 200€ million for the coming year, any person who has to quit because of this is a net benefit for him."

An anonymous reader quotes a report from Ars Technica, written by Dan Goodin: Two security professionals who were arrested in 2019 after performing an authorized security assessment of a county courthouse in Iowa will receive $600,000 to settle a lawsuit they brought alleging wrongful arrest and defamation. The case was brought by Gary DeMercurio and Justin Wynn, two penetration testers who at the time were employed by Colorado-based security firm Coalfire Labs. The men had written authorization from the Iowa Judicial Branch to conduct "red-team" exercises, meaning attempted security breaches that mimic techniques used by criminal hackers or burglars.

The objective of such exercises is to test the resilience of existing defenses using the types of real-world attacks the defenses are designed to repel. The rules of engagement for this exercise explicitly permitted "physical attacks," including "lockpicking," against judicial branch buildings so long as they didn't cause significant damage. [...] DeMercurio and Wynn's engagement at the Dallas County Courthouse on September 11, 2019, had been routine. A little after midnight, after finding a side door to the courthouse unlocked, the men closed it and let it lock. They then slipped a makeshift tool through a crack in the door and tripped the locking mechanism. After gaining entry, the pentesters tripped an alarm alerting authorities.

Within minutes, deputies arrived and confronted the two intruders. DeMercurio and Wynn produced an authorization letter -- known as a "get out of jail free card" in pen-testing circles. After a deputy called one or more of the state court officials listed in the letter and got confirmation it was legit, the deputies said they were satisfied the men were authorized to be in the building. DeMercurio and Wynn spent the next 10 or 20 minutes telling what their attorney in a court document called "war stories" to deputies who had asked about the type of work they do. When Sheriff Leonard arrived, the tone suddenly changed. He said the Dallas County Courthouse was under his jurisdiction and he hadn't authorized any such intrusion. Leonard had the men arrested, and in the days and weeks to come, he made numerous remarks alleging the men violated the law. A couple months after the incident, he told me that surveillance video from that night showed "they were crouched down like turkeys peeking over the balcony" when deputies were responding. I published a much more detailed account of the event here. Eventually, all charges were dismissed.

An open-source AI agent originally called Clawdbot (now renamed Moltbot) is gaining cult popularity among developers for running locally, 24/7, and wiring itself into calendars, messages, and other personal workflows. The hype has gone so far that some users are buying Mac Minis just to host the agent full-time, even as its creator warns that's unnecessary. Business Insider reports: Founded by [creator Peter Steinberger], it's an AI agent that manages "digital life," from emails to home automation. Steinberger previously founded PSPDFKit. In a key distinction from ChatGPT and many other popular AI products, the agent is open source and runs locally on your computer. Users then connect the agent to a messaging app like WhatsApp or Telegram, where they can give it instructions via text.

The AI agent was initially named after the "little monster" that appears when you restart Claude Code, Steinberger said on the "Insecure Agents" podcast. He formed the tool around the question: "Why don't I have an agent that can look over my agents?" [...] It runs locally on your computer 24/7. That's led some people to brush off their old laptops. "Installed it experimentally on my old dusty Intel MacBook Pro," one product designer wrote. "That machine finally has a purpose again."

Others are buying up Mac Minis, Apple's 5"-by-5" computer, to run the AI. Logan Kilpatrick, a product manager for Google DeepMind, posted: "Mac mini ordered." It could give a sales boost to Apple, some X users have pointed out -- and online searches for "Mac Mini" jumped in the last 4 days in the US, per Google Trends. But Steinberger said buying a new computer just to run the AI isn't necessary. "Please don't buy a Mac Mini," he wrote. "You can deploy this on Amazon's Free Tier."

Court documents unsealed last week in a copyright lawsuit against Anthropic reveal that the AI company ran an operation called "Project Panama" to buy millions of physical books, slice off their spines, scan the pages to train its Claude chatbot, and then send the remains to recycling companies.

The company spent tens of millions of dollars on the effort and hired Tom Turvey, a Google executive who had worked on the legally contested Google Books project two decades earlier. Anthropic bought books in batches of tens of thousands from retailers including Better World Books and World of Books. A vendor document noted the company was seeking to scan between 500,000 and two million books.

Before Project Panama, Anthropic co-founder Ben Mann downloaded books from LibGen, a shadow library of pirated material, over 11 days in June 2021. He later shared a link to the Pirate Library Mirror site with colleagues, writing "this is awesome!!!" Meta employees similarly downloaded books from torrent platforms after approval from Mark Zuckerberg, court filings allege, though one engineer wrote that "torrenting from a corporate laptop doesn't feel right." Anthropic settled for $1.5 billion in August without admitting wrongdoing.

Television marks its centenary today, exactly 100 years after Scottish inventor John Logie Baird first demonstrated his electro-mechanical system to journalists and members of the Royal Institution in a cramped attic workshop above what is now Bar Italia in London's Soho.

On January 26, 1926, small groups of visitors climbed to 22 Frith Street and watched fuzzy images of a ventriloquist's dummy called Stooky Bill appear on screen, followed by each other's faces transmitted from a separate room. One visitor got too close to the spinning discs and ended up with a sliced beard. The Times published a short account two days later.

Baird had built his first transmitting equipment in Hastings in 1923 using a hatbox, tea chest, darning needles and bicycle light lenses. A 1000-volt electric shock and a displeased landlord pushed him to London, where Gordon Selfridge soon invited him to demonstrate the device during the store's Birthday Week celebrations. The building at 22 Frith Street now carries three plaques commemorating the invention.