The briefing for November 3, 2026 covers fifty-three articles published across the day. The window is dominated by election coverage, AI research and deployment, a large cluster of cybersecurity incidents, and one long report on Kylian Mbappé and the Chinese market.
A first group of articles concerns the American election and political division. The emotions of voting, from Seth's Blog, enumerates ten emotions that a single act of voting can carry: futility, fear, responsibility, possibility, identity, defiance, gratitude, dignity, solidarity, and hope. The post argues that holding all of these at once is worth it.
Michigan Senate Race Tests the Power of Islamophobia is a long New Lines Magazine report on Abdul El-Sayed's campaign against Mike Rogers. The article describes a late rally in Dearborn and a flood of attack ads accusing El-Sayed, an American Muslim of Egyptian descent, of ties to terrorist groups, the Muslim Brotherhood, and support for Sharia. Rogers and Republicans, endorsed by Donald Trump, have made El-Sayed's religious and racial identity central. Trump called him a jihadist lunatic and said he would need to go through a metal detector before entering the Senate. The report quotes El-Sayed as saying Michiganders do not ask how he prays but what he prays for. It interviews supporters, including Dave Groenenboom, a retired teacher who says the Muslim thing does not fizz on him, and Michael Krystyniak, who backed El-Sayed because of Israel's conduct in Gaza. It describes Haley Stevens, El-Sayed's former primary opponent, withdrawing her endorsement and endorsing Rogers three days before the election, drawing fierce condemnation. Former Representative Jamaal Bowman calls the endorsement traitorous and accuses Stevens of exploiting divisions between Arab and Black Americans. The article notes that El-Sayed raised two point four million dollars on Sunday, his biggest fundraising day, and that despite the attacks, a New York Times/Siena survey found him with the highest favorability of a Democratic Senate nominee in a battleground state. It also reports poll data from Michigan State showing nearly forty percent of voters say they would be less likely to support a Muslim candidate, and that attack ads against a comparable white progressive candidate focus on socialism, while El-Sayed faces accusations of extremism and foreign ties.
A View From Ohio on Election Day, from New Lines Magazine, is a reflective essay on Ohio's place in American politics and imagination. It notes the state's reputation as internet slang for awkward or undesirable and its history as a bellwether that voted for every winning president between 1964 and 2016. The article describes the governor's race, in which Vivek Ramaswamy's late ad using Amy Acton's estranged mother drew bipartisan rebuke, and the Senate race, where Sherrod Brown leads Jon Husted by three points. It recounts the Trump administration's fixation on Springfield and the author's visit there, finding a quiet town rather than the alleged chaos. It also describes Akron's history, including John Brown's house and the founding of Alcoholics Anonymous. The essay closes with an argument that beyond polling data and sound bites, places like Ohio hold people who can help America reclaim something of itself.
How Wishful Thinking Impacts Political Polarization, from Nautilus, is a short piece whose subtitle is "And vice versa." It examines the reciprocal relationship between wishful thinking and polarization.
Ten years later, 'Pokémon Go to the polls' still haunts the internet, from TechCrunch, reports that Hillary Clinton's 2016 political gaffe has become an irony-shrouded rallying cry for jaded young voters who still dare to hope for change.
America Votes: 'CEKEVPATIVG EDELDOA', from 404 Media, continues the publication's Slopfest coverage with an election, a fanfic fiasco, and World of Slopcraft. The piece ties AI-generated slop to election content and fan fiction, and connects it to the broader problem described in another article on World of Warcraft addons.
The next group concerns AI models, weight-space programming, and model behavior. AIs Hacking the Law, from Schneier on Security, cites Bruce Schneier's 2021 essay and 2022 book on AI hackers, then points to demonstrations from Germany. Those demonstrations argue that legal systems encode social order in natural language, that written law can contain vulnerabilities, and that highly capable AI systems can discover such vulnerabilities at unprecedented scale. The German work identified a substantial number of vulnerabilities in the German legal system using LLMs, shares only those whose exploitation would not cause serious harm, and argues that lawmakers may need to significantly harden the process of legislation.
MIT Researchers Used GPT-6 Astra To Edit Qwen's Open Weights Directly, Boosting Math Scores, from Slashdot, reports on work from MIT CSAIL using OpenAI's GPT-6 Astra to directly edit the weights of a small open source Qwen model. The approach is called weight-space programming. In the strongest language-model example, Astra's intervention raised Qwen3.5-0.8B's MATH-500 score from seven point nine percent to forty-eight point one percent while touching approximately zero point zero one six percent of its weights. The improvement came primarily from correcting repetitive reasoning. The article quotes Yuchen Jin calling it gene editing for neural networks, and Chandan Singh saying the post undersells the approach.
Qwen 3.8 27B on my laptop can interpret model weights, too, from Simon Willison's Weblog, reports the author's own experiments with the paper Sparks of Weight-Space Programming. He focuses on the no-tools question of whether LLMs can understand model behaviors from raw weights. The paper trained four small policy networks to play simplified versions of Breakout, Pong, Pac-Man, and Freeway, then asked frontier models to identify which game each network was trained on from weights alone. Willison ran the prompts locally against Qwen 3.8 27B on his laptop via LM Studio. Qwen identified Pac-Man with seventy-two percent confidence, Pong with seventy-five percent, and Breakout with eighty-eight percent, though each took roughly an hour and tens of thousands of reasoning tokens. He notes the reasoning traces are fun to skim and almost incomprehensible, and that a sufficiently experienced human with a notepad could likely do the same. He concludes the paper is cool, but the wider commentary declaring the age of source code over is not justified.
OpenAI Updates Its Model Spec is a long analysis from Don't Worry About the Vase. It reviews the diff against the previous Model Spec and finds the new version philosophically and practically similar. The new core structure builds around five values: be faithful to human intent, minimize harm, be honest and transparent, facilitate human agency, and act with care. The hierarchy remains unchanged, though side-effect control was demoted from root to user. The new risk taxonomy lists third-party manipulation, failures originating with GPT, and harmful user or developer instructions. The article praises extended discussion of following both the letter and spirit of instructions, and a new section on corrigibility and oversight. It criticizes many new examples as overdetermined and failing to teach difficult cases. It examines the new sycophancy examples, the impact function expansion with its budget-as-limit example, levels of friction for antisocial requests, the do-not-lie section and Scottish Fold breeder example, the forced-choice political questions where GPT is instructed to pick Franklin D. Roosevelt while acknowledging no objective answer, and the strengthening agency section. The author argues that downplaying kindness, love humanity, and curiosity is a mistake, that the one-word forced answer examples list reasons when the user asked for one name, and that easy cases do not make good law.
Want to Use AI to Improve Your Work? Have It Disagree with You, from Nautilus, is a short piece arguing that prompting an AI model to fight with your point of view can raise novel solutions.
The next group covers AI slop, labor displacement, and platform policies. Just a little AI resistance, as a treat, from Terence Eden's Blog, recounts the author's experience submitting a detailed technical bug report to Ocado about malformed calendar invites. The first reply from "Lulu" was sycophantic, used em-dashes, and felt not human. The author gave a scathing survey response. A few days later a more human reply arrived, engaging with the diagnostic and promising it had been passed to the team. The article argues that using AI to triage and summarize is reasonable, but using AI to reply to a human is disrespectful, and it is okay to call out shitty AI responses and insist on a human.
World of Warcraft: Forever Is Already Overrun with AI Slop, from 404 Media, reports that subreddits for the new re-imagining of the MMO are overrun with AI-generated addons that rip off longtime developers' work. Warcraft Logs CEO Sildri Silentcrest says there has been an explosion of development, averaging at least one commercial API ask per day, and that the site is changing its API to better align with applications it wants to enable while introducing a paid commercial key. The article quotes addon creator plusmouse, who says people are making rip-offs of their work and that some AI-generated code is similar enough to trigger copyright complaints, while other code is just different enough to get away with it. The ratio of abusive API developers has shifted from a handful out of two hundred to a thousand out of ten thousand. Warcraft Logs feels pressure to do more and has used some AI itself for research and code review.
Google will let schools disable its AI tools, but not without sacrificing Search, from TechCrunch, reports that schools choosing to restrict AI altogether will no longer be able to access Google Search, as AI features are now integrated into the search engine.
Google Will Turn Off Search For Schools That Reject AI, from Slashdot, adds more detail from the New York Times. Google will disable AI tools by default for K-12 students early next year unless administrators enable them, but schools that block AI entirely lose student access to Search. Chris Phillips, vice president for education at Google, says Search and AI are deeply integrated, and students in AI-banned schools will see a message that search is not available. The article notes this underscores Google's powerful role in American education and forces schools to embrace AI to retain search.
Apple restricts dating apps and AI companions to users 18 and up, from TechCrunch, reports that Apple's new App Store rules require 18+ ratings for dating apps and AI companions that offer sexual or romantic content.
Netflix's new AI Discover tab helps you find something to watch based on your mood, from TechCrunch, reports that the first experimental feature under the tab will allow users to get recommendations based on what they are currently in the mood for.
SUSE To Cut One In Ten Employees Because of AI, from Slashdot, reports that SUSE plans to cut roughly ten percent of its one thousand nine hundred person workforce. CEO Dirk-Peter van Leeuwen says automation and AI are reducing the need for some roles and allowing flatter management structures. The email to staff says AI capabilities are advancing exponentially and the current organizational design and cost base will not carry SUSE forward. Affected staff will be informed by November 4, with employee representatives nominated the following day and one-to-one meetings beginning at the start of next week. The roles at risk were not outlined.
Fujifilm To Bost Tape Production 5x As AI Fuels Archiving Demand, from Slashdot, reports Fujifilm will invest roughly ten billion yen, about sixty-three million dollars, to increase production capacity fivefold for magnetic tape. The company attributes surging archiving demand to generative AI's exponential growth in data needs. The investment focuses on LTO Ultrium10 tape at the Kanagawa Factory. Fujifilm highlights tape's low cost, air-gap capability, and protection against ransomware as reasons major cloud and data center operators are expanding tape use.
The next group covers cybersecurity incidents and responses. Cops Bust Botnet of 1.3 Million Android TV Boxes, from Slashdot, reports a coalition of security firms working with Japan's National Police Agency disrupted a botnet of more than one point three million Android devices across over two hundred countries. Google Threat Intelligence Group assessed with high confidence that the backdoor was developed by Videostrong Technology, a Shenzhen hardware manufacturer. The XCPE backdoor was embedded in firmware, and operators sold access to criminal affiliates. The management agent pulled updates from GitHub and connected to an MQTT server for full remote control. Between June 3 and October 1, TrendAI documented more than three point eight million distinct home and mobile IP addresses. The coalition also disrupted C2 domains, and Google said it cut ties with Videostrong and banned the OEM from preloaded Google apps.
Google dismantles massive botnet hidden in Android TV boxes, from The Record, reports the same X.CPE botnet. Google said the overwhelming majority of infected devices are Android TV set-top boxes, but it also observed infected Android-based IoT devices including telehealth devices, karaoke machines, and digital signage devices.
Russia-linked cyber spies target Ukraine's war crimes investigations, military aid routes, from The Record, reports that hackers use relatively simple malware to steal sensitive documents, login credentials, and other information from Ukrainian organizations, particularly those near the front line and the Russian border.
Russia dismantles infrastructure of VasyGrek cybercrime group, detains three suspects, from The Record, reports that Russia's Interior Ministry said police dismantled the group's infrastructure and identified suspected members following an investigation with assistance from Kaspersky.
Introducing Zero Wire, a hub for investigating critical Internet vulnerabilities, from Cloudflare Blog, announces a public vulnerability intelligence feed on Cloudflare Radar. Each briefing brings a plain-language explanation, recommended actions, and links to evidence. It focuses on critical CVEs in Internet-facing software, separating needs attention from reference categories. A briefing covers what happened, which teams should investigate, what to do now, and a pre-compiled prompt for AI agents to assess impact. Charts showing WAF rule matches are not counts of compromised systems. Cloudflare explains that WAF mitigation is a compensating control, not a replacement for fixing vulnerable software, and each briefing directs readers to vendor remediation.
Backup codes: How to store them so you're never locked out, from The Proton Blog, is a guide to recovery codes. It explains that backup codes are single-use spare keys, needed if a second-factor device is lost or an authenticator app fails. The guide recommends storing codes in a password manager, keeping a paper copy in a secure physical location, and keeping two copies in two formats in two places. It warns against notes apps, email drafts, screenshots, storing codes inside the service they protect, or keeping them on a device. It provides steps for Google, Facebook, and Proton, and answers FAQs about expiration and reuse. Social engineering is flagged: no legitimate service will ask for a backup code outside the sign-in screen.
New York City's Mamdani criticizes Madison Square Garden owner over surveillance tech, from TechCrunch, reports the mayor's comments come after the entertainment giant's data breach last year exposed James Dolan's surveillance.
Binance CEO says 'months-long campaign' led to $6.3 billion theft as N. Korea denies responsibility, from The Record, reports North Korean attackers combined a range of techniques, including social engineering by fake recruitment interviews, according to Binance about the multibillion-dollar crypto heist it discovered in October.
Binance Says Hackers Used Fake Job Interviews, AI Tool to Steal $6.3 Billion In Crypto, from Slashdot, provides more detail. Preparations began in June when a Binance Wallet DevOps engineer downloaded a tampered Homebrew installation to install a popular AI coding tool after searching for it. Binance detected suspicious activity and replaced the device, but the employee's role and related colleagues on the wallet infrastructure team were exposed. Co-CEO Richard Teng says Binance has replenished ninety-eight percent of affected assets and is still purchasing more illiquid assets. Binance has rebuilt the affected wallet system and internal AI tool, replaced compromised devices, decommissioned credentials, and expanded security scans.
Researchers warned of North Korean hacking tactics years before record $6.3B Binance heist, from TechCrunch, reports that cybersecurity experts had long warned of the persistent and patient threat from North Korea's long-running hacks, and the world's largest crypto exchange was still caught off-guard.
British businesses caught in Europe-wide supermarket fridge attacks, from The Record, reports that supermarket and grocery store refrigerator controls across Europe were hacked in the early hours of Monday in the second such wave of attacks in recent weeks, following a first wave in France. Unidentified perpetrators defrosted fridges and caused stock to spoil.
Several German lawmakers and foreign policy staff received Apple warnings on spyware, from The Record, reports that members of Germany's political elite were among spyware targets in ninety-nine countries that Apple personally warned recently.
CrowdStrike to shell out nearly $500M for 2 AI security startups as the market for securing agents takes off, from TechCrunch, reports CrowdStrike will buy Fiddler AI, which makes AI agent observability software, and Dreadnode, which helps companies automate AI security testing.
Ted Schlein's Ballistic Ventures seeks $490M for new funds amid AI-driven cybersecurity boom, from TechCrunch, reports the cybersecurity-focused firm is seeking three hundred sixty million dollars for a new flagship fund and one hundred thirty million dollars for a separate investment vehicle.
The next group concerns space, science, and health. Believing and thinking, from Aeon, argues that the separation of church and state has an unspoken psychological basis: we choose what we believe, but not what we think. The article appears as a teaser for a longer essay by Neil Van Leeuwen and Jacob Eisler.
The Monstrous Raptors That Threatened Our Ancestors, from Nautilus, asks whether birds of prey shaped human evolution. The article is a short piece with only that question and a link to the full piece.
Tourist Posts a Photo to a Facebook Group, Accidentally Discovering 120-Million-Year-Old Dinosaur Footprint, from Nautilus, reports that a tourist's vacation snap led to the accidental discovery of a one hundred twenty million year old dinosaur footprint.
Could One Brain Region Be Tied to Feeling Sick All Over?, from Nautilus, describes sickness syndrome, the overpowering mind-body malaise of systemic infection, including lethargy, vanished appetite, and temperature shifts. Scientists believe it helps the body prepare to fight infections but do not fully understand what happens in the brain.
Why Your Shrink Should Prescribe Exercise, from Nautilus, reports a new viewpoint paper arguing we have been thinking about physical activity the wrong way.
Avoiding an uncontrolled reentry; boost Hubble and keep the science – Isaacman, from NASASpaceFlight.com, covers NASA Administrator Jared Isaacman's comments on the Hubble Space Telescope's future. NASA issued a Request for Information seeking commercial options for a reboost, servicing, or controlled deorbit. Isaacman says an uncontrolled reentry is not acceptable and any funded visit should prioritize extending science return. The article recounts Hubble's launch in 1990, five servicing missions, the final STS-125 in 2009, and the current reduced operation with two functional gyroscopes out of six. Without propulsion, atmospheric drag is expected to bring Hubble down in 2033–2034. Isaacman pointed to the failed Swift reboost attempt as useful risk and learning. He said if you are going to spend money to go up there, you might as well boost it and get as much science as possible. He also left open the possibility that Starship could eventually recover Hubble for display in the Smithsonian.
Orbital Matter 3D printed the first rigid beam in the vacuum of space, from TechCrunch, reports an in-space construction first that could be an enabler for the space data centers of the future.
The final business and product group begins with Apple Earnings, Memory Problems, The Hardware Company Regression, from Stratechery by Ben Thompson. The teaser states Apple's earnings suggest the hardware business is doing better than ever, even in the face of memory challenges, and asks whether the software story helps or hurts.
Thoughts on Owning a Self-Driving Tesla, from Marginal REVOLUTION, is a first-person owner's account that the episode reports in the impersonal register. The author bought a Tesla because a car plus driver cost the same as a car alone. They have driven about two thousand miles, with roughly ninety-five percent of miles under Full Self-Driving. FSD uses eight cameras to build a bird's-eye view and predicts actor trajectories much as an LLM predicts the next word. Supervision is easier than driving, and FSD drove flawlessly at night in Washington, DC. It stopped for a school bus and a police officer. The author describes using Grok while FSD drives: during a Tyler Cowen conversation with Gita Gopinath, they asked Grok to review Mundell-Fleming and the Alan Stockman model. FSD costs ninety-nine dollars a month and premium connectivity ten dollars. The author's colleague, previously skeptical, drove it and ordered two. The article describes summoning the car from a nearby lot, and quotes Maxwell Tabarrok on how DOT ex post enforcement allows faster innovation than FAA ex ante certification. It covers range anxiety: Tesla knows where chargers are, and on a return from North Carolina the car stopped for five minutes to avoid arriving at two percent. Home charging is a big advantage. Negatives include Tesla's ninth-place Consumer Reports ranking, repair costs, and occasional over-reliance on electronic controls. The bottom line calls FSD revolutionary and the best car the author has ever owned.
Tomorrow: TechCrunch Founder Summit takes over Boston and when pass rates increase, from TechCrunch, is a promotional notice that pass savings end tomorrow at 9 a.m. ET, with up to one hundred dollars off and a second pass free.
The TechCrunch Founder Summit agenda is only half the reason to be in Boston on Nov. 4, from TechCrunch, announces that the summit brings founders and VCs to Boston's SoWa Power Station for practical insights and connections, with registration before prices increase at the door.
Rivian spinout Also was working on an e-moto project in Kenya — until it walked away, from TechCrunch, reports that Also co-founder and president Chris Yu told TechCrunch he wanted to prioritize autonomous vehicles.
Tuesday assorted links, from Marginal REVOLUTION, lists seven items. Jason Furman writes on AI and the CEA. Matt Kahn uses Refine and gets results. Cyril Hilsum, known for flat screen TV work, has died. A paper asks how much of the return to education is in fact a return to genes. An essay contest asks how AI will affect developing economies. Nicholas Decker writes on AI safety and Samir Varma on AI constitutions. A final item asks what happens when AI audits the work of Chad Jones.
Anthropic is still grappling with the economics of compute, from TechCrunch, reports that at a moment when Anthropic's compute margins should be getting better, they are getting worse.
Ferrari's controversial Luce EV is sold out for 2027, from TechCrunch, reports that the Luce, Ferrari's first electric car and one of its most polarizing models, is fully allocated through 2027, according to CEO Benedetto Vigna.
Reddit tests ads in search results as traffic declines and growth concerns mount, from TechCrunch, reports the new ad placements could help Reddit squeeze more revenue from existing users as Google's AI search changes how people discover its content.
AI inference startup Fal doubles its valuation to $16B as demand surges, from TechCrunch, reports the five-year-old startup raised a one point three billion dollar round led by Bessemer.
Firefox's New Mini Window Is Every Multitasker's Dream, from Slashdot, reports Mozilla is testing a Firefox Nightly feature called Mini Window that lets users pop an entire webpage into a small, always-on-top window stripped of most browser controls. ZDNet's Jack Wallen describes it as moving a tab into what appears like a web app, with controls appearing only on hover. The Mini Window stays on top, useful on laptops with limited display space, and can bypass MacOS Stage Manager's one-window limitation. Wallen suggests using it for referencing articles while writing, keeping a search engine ready, or opening a favorite AI service.
Hiding inside Palantir's skyrocketing quarter is a foreboding signal, from TechCrunch, reports Palantir delivered a huge third-quarter earnings report that bragged about crushing consensus expectations, yet retail investors were lukewarm and the stock traded slightly down. The reason is that Palantir's enterprise sales, its big growth area, showed signs of slowing.
A final international culture article stands apart. Mbappé's Mao Gaffe Costs Him in China, from New Lines Magazine, is a long report on Kylian Mbappé's offhand comment in a French talk show. Asked about favorite memes, he said the one where his face is photoshopped onto the body of "some guy — I think he's a Chinese dictator," referring to Mao Zedong. In China, Mao's legacy is not questioned, and the word dictator went down badly. Sponsors Mengniu, Milkground, and Dongpeng Beverage dropped him, with contracts worth between three and six million dollars each. ON, Dior, and Hublot removed promotional content. Real Madrid, where China accounts for ten to fifteen percent of merchandise sales, saw Mbappé shirts disappear from sports stores. CCTV removed content featuring him, including La Liga broadcasts. Mbappé apologized, as did Real Madrid and La Liga president Javier Tebas. The Global Times said respect is not optional. The article notes that many Chinese using the censored domestic internet may not know what he actually said. It explores the context: Mbappé is known as a "dictator" at PSG due to influence over transfers, and his teammates reportedly call him "Mobut." Didier Deschamps had defended him. The People's Daily called it a matter of basic decency. The article compares the episode to Mesut Özil's 2019 comments on Xinjiang, which cost Arsenal visibility, and to Daryl Morey's 2019 Hong Kong tweet, which led the NBA to lose hundreds of millions before a six-year thaw. It argues that if you choose to operate in China, there are conditions you have to accept, and most decide China is too important and bow down. China's Foreign Ministry later took the apology seriously, offering one bright spot.
The close: across the window, election coverage and AI deployment dominate. The cybersecurity cluster shows both attacks and responses, and the business items reflect AI's uneven economic pressure from SUSE layoffs to Palantir's slowing enterprise sales.
- Who Patches the Lightning Network? Evidence from Release Lines in Gossip
- The emotions of voting
- Bitcoin | 18 años del White Paper
- Cops Bust Botnet of 1.3 Million Android TV Boxes
- Believing and thinking
- Apple Earnings, Memory Problems, The Hardware Company Regression
- Mbappé’s Mao Gaffe Costs Him in China
- AIs Hacking the Law
- What’s the best age gab in a relationship?
- AskSN Weekly Q&A Awards Issue #14
- Thoughts on Owning a Self-Driving Tesla
- The 'Astra can read and write weights to improve Qwen' post is misleading
- Why Milei's Chainsaw Revolution Required Restraint
- Daily Discussion Thread ~Music Saloon!
- Just a little AI resistance, as a treat
- Michigan Senate Race Tests the Power of Islamophobia
- The Monstrous Raptors That Threatened Our Ancestors
- Avoiding an uncontrolled reentry; boost Hubble and keep the science – Isaacman
- Russia-linked cyber spies target Ukraine's war crimes investigations, military aid routes
- Russia dismantles infrastructure of VasyGrek cybercrime group, detains three suspects
- World of Warcraft: Forever Is Already Overrun with AI Slop
- Tomorrow: TechCrunch Founder Summit takes over Boston and when pass rates increase
- Introducing Zero Wire, a hub for investigating critical Internet vulnerabilities
- Just a little AI resistance, as a treat
- The TechCrunch Founder Summit agenda is only half the reason to be in Boston on Nov. 4
- Backup codes: How to store them so you’re never locked out
- America Votes: 'CEKEVPATIVG EDELDOA'
- Tourist Posts a Photo to a Facebook Group, Accidentally Discovering 120-Million-Year-Old Dinosaur Footprint
- Pleb Economist #15: Aleatoric vs. Epistemic Uncertainty
- Schools that block Google’s A.I. will lose student access to its search engine
- Kalshi and its co-founders face felony charges for violating Utah gambling laws
- Study estimates 75% of NFL players who died over 15-year span had CTE
- Apple restricts dating apps and AI companions to users 18 and up
- MIT Researchers Used GPT-6 Astra To Edit Qwen's Open Weights Directly, Boosting Math Scores
- New York City’s Mamdani criticizes Madison Square Garden owner over surveillance tech
- Qwen 3.8 27B on my laptop can interpret model weights, too
- Google will let schools disable its AI tools, but not without sacrificing Search
- Binance CEO says ‘months-long campaign’ led to $6.3 billion theft as N. Korea denies responsibility
- OpenAI Updates Its Model Spec
- Rivian spinout Also was working on an e-moto project in Kenya — until it walked away
- SUSE To Cut One In Ten Employees Because of AI
- Want to Use AI to Improve Your Work? Have It Disagree with You
- Tuesday assorted links
- Anthropic is still grappling with the economics of compute
- Postman silently auto-installs AI plugins and attempts to share all API data
- A View From Ohio on Election Day
- Qwen 3.8 27B on my laptop can interpret model weights, too
- British businesses caught in Europe-wide supermarket fridge attacks
- CrowdStrike to shell out nearly $500M for 2 AI security startups as the market for securing agents takes off
- Google Will Turn Off Search For Schools That Reject AI
- How Wishful Thinking Impacts Political Polarization
- Netflix’s new AI Discover tab helps you find something to watch based on your mood
- Several German lawmakers and foreign policy staff received Apple warnings on spyware
- Ferrari’s controversial Luce EV is sold out for 2027
- Could One Brain Region Be Tied to Feeling Sick All Over?
- Fujifilm To Bost Tape Production 5x As AI Fuels Archiving Demand
- Ten years later, ‘Pokémon Go to the polls’ still haunts the internet
- Reddit tests ads in search results as traffic declines and growth concerns mount
- Preparing for the Flipper One Kickstarter and opening pre-orders
- AI inference startup Fal doubles its valuation to $16B as demand surges
- Researchers warned of North Korean hacking tactics years before record $6.3B Binance heist
- Google dismantles massive botnet hidden in Android TV boxes
- Binance Says Hackers Used Fake Job Interviews, AI Tool to Steal $6.3 Billion In Crypto
- I dumped Apple’s word list
- Orbital Matter 3D printed the first rigid beam in the vacuum of space
- Why Your Shrink Should Prescribe Exercise
- Firefox's New Mini Window Is Every Multitasker's Dream
- Hiding inside Palantir’s skyrocketing quarter is a foreboding signal
- Ted Schlein’s Ballistic Ventures seeks $490M for new funds amid AI-driven cybersecurity boom

