The briefing for October 29, 2026 covers sixty-one articles. The day's reporting is heavily concentrated on artificial intelligence, cybersecurity, and business consolidation, with substantial long-form pieces from the Middle East and a cluster of science features.
Several articles converge on the claim that artificial intelligence is accelerating scientific discovery. Noahpinion's 'A golden age of science is coming' argues that recent worries about scientific stagnation are being overturned by AI. The piece recalls papers claiming ideas are getting harder to find and John Horgan's book The End of Science, then presents countervailing examples: Anthropic's drug discovery work, a possible food allergy treatment, and OpenAI's announcement regarding neurodegenerative disease. The article identifies three main ways AI removes constraints. First, the number of effective researchers is no longer limited by trained human brains, since AI agents can be multiplied and trained quickly. Second, AI eliminates the theory bottleneck because models are strong at mathematics; the article cites OpenAI solving open problems in computer science, math, and physics, including a proof that the Standard Model is mathematically well-defined. Third, AI can read the whole literature at once, as shown by OpenAI's distributed biological reasoning with thousands of agents, which reproduced human analysis and found novel leads. The author then introduces the concept of 'cloud laws'—causal regularities too diffuse for human comprehension but potentially exploitable by AI—and argues that with enough sensor data, AI could learn directly from the natural world, leading to a 'superscience' whose technologies would be indistinguishable from magic to humans. The conclusion is that the scientific slowdown has been delayed, and a new golden age is likely, though it may not last forever.
Marginal REVOLUTION's 'How Tyler Cowen and Alex Tabarrok Have Changed Their Views on AI and Where They Disagree' traces the public evolution of the two economists from 2011 through 2026, using their own quotations. Both became dramatically more confident in AI capabilities while remaining optimistic about economic potential, but they differ on how benefits arrive, how to weight dangers, and whether AI might be conscious. Cowen early emphasized human bottlenecks and institutional pace, arguing that more productive AI makes human constraints more binding and estimating only an additional half percentage point of annual growth. Tabarrok initially compared AI to cheap educated labor, then in 2023 said he 'never expected to witness the birth of aliens' and in 2025 called the intelligence revolution bigger than the industrial revolution. Their disagreements include the pace of takeoff, the severity of safety risks, and the sincerity of AI executives' warnings; Tabarrok argues Amodei actually believes what he says, while Cowen stresses competition for power. On consciousness, Tabarrok draws on the history of infant pain to urge moral caution under uncertainty, while Cowen categorically opposes the view that AIs are sentient, calling it a category error. The article notes they converge more than their rhetoric suggests, both rejecting mass unemployment while acknowledging transitional hiring problems.
The longest single piece in the window, Stratechery's 'An Interview with Palantir CTO Shyam Sankar About the AI Religious War,' is a full transcript of Ben Thompson's conversation with Sankar. Sankar argues the industry over-invests in the supply of intelligence and under-invests in diffusion, using the analogy of electricity requiring machine tools to reach the economy. He contends the atomic unit of the enterprise is not the knowledge worker but the workflow, which explains why coding has product-market fit but many enterprise AI efforts fail. He says companies focused on top-line ambition succeed more than those trying to cut headcount, and he criticizes the labs for a lack of systems engineering in AI safety, contrasting Admiral Rickover's nuclear Navy with the Soviet approach. Sankar explains Palantir's ontology as a state machine that models verbs as well as data, describes how the company now uses AI forward-deployed engineering, and warns that without sovereignty over post-training pipelines, firms will see their alpha become beta. He also discusses model questions, why he lost confidence in benchmarks, forward-deployed engineering as a product strategy held together by hackers and artists, and the 2026 dispute between Anthropic and the U.S. government, which pushed Palantir to enable safe model swapping. On AI religion, Sankar calls effective altruism a secular religion that rejects human agency and systems engineering, while acknowledging Palantir itself was a different kind of cult in service of the West. The interview ends with Sankar warning that wrong safety regimes, nationalization, or data-center moratoriums could produce non-market forces that are the greater risk.
Martin Fowler's 'Why are good developers so wary of agentic development?' addresses the hesitation among experienced developers to hand all execution to coding agents. Fowler argues the caution comes from knowing that 'built is not the same as working in production.' He reports on a week-long Thoughtworks experiment where no human wrote code and agents built 80 percent of an airline application, but the build was frequently red and the team spent two-thirds of its time at the whiteboard defining domain boundaries and designing feedback loops such as tests, contracts, and delivery pipelines. The article draws an analogy to skiing: experienced developers are not nervous beginners on a groomed slope but backcountry travelers in unmarked terrain, asking sensible questions about the route. Fowler urges leaders to create real room for experimentation and developers to bring their judgment, because nobody will hand out a map first.
404 Media's 'Manifestos, Fear, and ‘Loser Thinking’: Inside Proton’s AI Reckoning' is a long investigative report based on interviews with eight current and one former Proton employees and leaked internal communications. The article shows CEO Andy Yen aggressively pushing the privacy-focused company into AI, including a Slack exchange where he called a security engineer's suggestion to pentest new code 'bullshit and loser thinking' and said only losers think that way. Proton set up an email address to 'ruthlessly smash' barriers to AI velocity, installed tracking software to measure AI-generated code, and announced re-interviews of staff explicitly consequential for employment. A CTO manifesto told engineers 'Your old job is gone.' The report describes a culture of fear where employees hesitate to criticize Yen, who is characterized as a bully in Slack and who said 'perception is more important than truth.' Internal documents say Proton is underperforming by all metrics, missed its annual recurring revenue target due to insufficient velocity, suffered a six-hour outage, and cannot relocate to the U.S. or China because of Swiss privacy law. The company plans an 'elite fast team' with a 30 percent salary bonus and weekend work, and Yen spoke of Proton being a '996 company.' A spokesperson did not dispute the reporting, acknowledging too much bluntness. Ninety percent of Proton's code in October was written by AI, much of it with Anthropic's Claude, despite Proton's public privacy branding.
Martin Fowler's 'Bliki: LLM Voice' is a short personal essay on the uncanny valley of prose generated by large language models. Fowler says he can often sense LLM output even without explicit tells, and that by the time he would reach for a detection tool it is already too late. He urges authors not to let an LLM write a first draft because the voice is very hard to expunge, and even rephrasing suggestions are dangerous. He recommends speaking words in the head before writing, and warns that delegating writing to LLMs means delegating thinking. He notes exceptions: Jim Highsmith uses an LLM to overcome Parkinson's disease, and translation is a reasonable use, though Fowler prefers a human's clunky prose. He declares he did not use AI or LLMs in any way while writing the post.
Simon Willison's 'Building a multiplayer game world with autonomous Fable-class models' describes a post by Jacob Sansbury about Spawn, a massively multiplayer game where players prompt an in-game agent named Savi to modify the world. Willison tested it on a phone, prompting a castle, then a dragon, then riding the dragon. The article reports Spawn is developed autonomously by a swarm of Fable-powered agents costing over three million dollars in tokens per month, with agents changing live code directly, shipping to production hourly, and treating an hour without a ship as a SEV-0. Savi files bugs herself, and about sixty percent of work starts from a production error or bug report. The team can roll back a broken change in thirty seconds. Willison says Fable 5 and 5.1 enabled this autonomy; Opus 5.5 does not quite work, and GPT-6 Astra, when asked to apply security fixes, disabled every feature and marked them to be turned on later. He calls these extreme experiments a substrate for exploring coding agents.
Simon Willison's 'Quoting Ben Hylak' reproduces a satirical tweet: every product should be a headless MCP feeding an agent, except for one's own product, where customers love the UI and the company is special. The piece adds tags for model-context-protocol and AI.
Several articles cover AI product launches, funding, and platform changes. TechCrunch's 'Amazon updates Alexa to help you manage shopping lists, subscriptions, and update delivery instructions' reports that Alexa can now create, register, view, add to, and edit lists, and manage subscriptions with the AI bot. TechCrunch's 'Deepgram acquires Pipecat-maker Daily to bolster its voice AI infrastructure' states Deepgram is committed to keeping Pipecat open source. 'Wabi, the AI agent that builds mini apps, now valued at $1.25B after latest round,' also from TechCrunch, says Wabi, now describing itself as a personal agent, raised $55 million at that valuation. 'Agent identity startup Mintive lands Cowboy Ventures, Coatue, angels Jeff Dean, Andrej Karpathy' reports Mintive was founded by two engineers with decades of AI experience through Google Brain and Coursera, and its platform helps IT control enterprise agents. TechCrunch's 'SuperMe raises $20M for its network of AI expert profiles' says SuperMe lets users ask AI clones of Silicon Valley's savviest people about their careers. 'Ex-Palantir execs launch Dream with $17M from a16z to build AI world models for construction' describes Dream as trained on 100 million square feet of building permits, aiming to disrupt Autodesk. 'Granola crunches up a dictation feature to compete with Wispr' notes the reverse move: after Wispr added meeting notes to its dictation app, Granola is adding dictation to its meeting notetaker. TechCrunch's 'Meta quietly launches an AI fashion app called Style Sketch' says the app connects to Instagram to build a virtual closet from photos already shared, or lets users add items manually. 'As search traffic dries up, Reddit announces it’s shifting to a logged-in experience' reports Reddit is rethinking its open-web roots to protect its audience and business as AI changes information discovery. Slashdot's 'Reddit Is Getting a 'For You' Feed' says Reddit is overhauling its homepage with separate For You and Following feeds, combining subscribed communities with algorithmic recommendations, and deprecating r/all. TechCrunch's 'Who will you learn from in 6 days at TechCrunch Founder Summit? Meet the founders and investors taking the stages' is a short promotional note that the summit brings experienced founders, investors, and operators to Boston on November 4. TechCrunch's 'Rivian expands joint venture with VW to include AI assistant' says Rivian does not disclose whether the value of its $5.8 billion joint venture will expand as a result. Slashdot's 'Canada to Join EU Network of Supercomputers' reports Canada is joining the EU's EuroHPC network to give Canadian companies and researchers access to European computing resources, aiming to reduce dependence on US technology; experts caution it does not yet dent actual US dependence unless massive buildouts occur.
Three articles address the economic consequences of AI adoption. Andrew Yang's 'The Dominos are Falling' lists a week of major announcements: Microsoft offering buyouts to 60,000 employees, American Express laying off 3,850 workers even as profits rose, the Dutch government replacing 25,000 workers with AI, Getty Images declaring bankruptcy because AI can conjure images cheaply, and Goldman Sachs reporting lower job growth in AI-adopting industries. Yang also notes Anthropic's S-1 filing for a possible $1.5 trillion IPO, with revenue up 14x to $16.3 billion in the first half of the year and risk factors including catastrophic or existential risks to humanity. He connects these to the Quest Diagnostics hack and 'Black October' for cybersecurity, and concludes that humans are being kicked to the curb in unprecedented ways, with the reaction to play out at the ballot box next Tuesday. Slashdot's 'Google Struggling to Make AI Mode Searches Pay' reports Alphabet shares fell nearly 10% after Google disclosed that AI Mode searches generate less revenue than traditional search, even as overall search revenue grew 12% to $63.25 billion. Capital spending of $51.4 billion, mostly on servers and networking, cut net income to $26.17 billion from $35 billion a year earlier. Chief business officer Philipp Schindler argued AI Mode queries are upper-funnel and incremental, and Google has begun rolling out ads in AI Mode globally. A companion Slashdot piece, 'Alphabet Heads For a $500 Billion Drop In Market Value,' says the stock is on track for its steepest post-earnings decline since 2004, with at one point a $524 billion market value loss, the second-biggest one-day loss ever for a US company behind Nvidia's $593 billion in January 2025. Alphabet has lost about $1.4 trillion since its record in May.
Cybersecurity and privacy dominate a large cluster. 404 Media's 'Senators Demand Meta Answer for 'Pervert Glasses' Following 404 Media Investigation' reports that Senators Ron Wyden and Edward Markey sent a letter to Meta demanding answers about smart glasses privacy, citing 404 Media's 2025 investigation into filming and harassment of massage parlor workers. The letter says Meta's disclosures confirm imagery may be retained and used for AI training with human reviewers, calling it an invasion of privacy. Research from August showed pickup artist influencers using the glasses to capture nonconsensual content, and over 70 people sued Meta earlier this year over intimate images exposed to data labelers. The senators demand answers about safeguards, training data, and deletion rights. Cloudflare's 'Cloudflare DDoS Threat Report 2026 Q3: More swarms, fewer giants' summarizes the third quarter: 18.2 million network-layer DDoS attacks mitigated, about 8,200 per hour; short-burst attacks under ten minutes and one gigabit per second surged 208%, while hyper-volumetric attacks over one terabit per second fell 69% following law enforcement actions against four IoT botnets. CLDAP attacks collapsed 99.5%, while DNS amplification surged 215% and UDP floods reached 31% of attacks. The United States reclaimed the most-attacked position; Canada jumped to sixth and Israel to tenth. The report stresses autonomous mitigation and a free botnet threat feed. The Record's 'University of New Mexico restores hospital systems, urges no contact with hackers' says the university restored healthcare systems after a cyberattack forced ambulance diversions and cancelled medical school classes. The Record's 'Cyberattack disrupts Belgian tax services before filing deadline' reports FPS Finance suspended access to MyMinfin while investigating. TechCrunch's 'Bluesky blocks links to leaked DHS employee data as government seeks takedowns' says Bluesky faces criticism for blocking access to a site identifying thousands of ICE agents and other Homeland Security employees. The Record's 'FTC to probe airlines, gig worker firms over personalized pricing and pay' reports the FTC ordered nine food delivery, airline, and ride-sharing companies to turn over details on how they use personal data to set individual prices and employee compensation. The Record's 'Federal bankruptcy courts contractor says no data stolen during cyber intrusion' quotes BAE Systems saying no data was extracted during an attack on its systems for federal bankruptcy courts. The Record's 'GM settles with 36 states for selling driver data, agrees to $220 million fine and reforms' says General Motors will pay $220 million to settle a lawsuit alleging unlawful sale of driver behavior information. The Record's 'Limited reports of PHP bug exploitation emerge' says one cyber firm listed a half-dozen exploitations of the PHP2Shell bugs, and another identified a targeted, suspected Chinese campaign breaching US government-affiliated servers. 404 Media's 'Podcast: Flock CEO Says the Backlash Isn’t Real' describes leaked audio from a private police conference in which Flock's CEO made wild claims about backlash not being real and his wife's advice to never apologize; the podcast also covers Marissa Mayer and a CashApp fraud case.
Several pieces cover business deals and product milestones. Feld Thoughts' 'Formlabs Form Studio 1' reports the launch of a professional resin 3D printer starting at $699, learnable in ten minutes with most jobs under two hours. Early reviews are positive: Tom's Hardware gave an Editor's Choice and said it just prints; All3DP called it the benchtop benchmark; CNC Kitchen's Stefan Hermann, a filament user, said it offers Form 4 convenience at a fraction of the cost and conceded he has not started hating it. The author recounts investing in MakerBot in 2010, its acquisition by Stratasys in 2013, and then leading Formlabs' $35 million financing in 2016; today Formlabs is the largest professional 3D printing company founded since the 1980s, with over $250 million revenue, profitability, and more than 500 million parts printed. Farnam Street's '[Outliers] K.C. Irving: How to Build an Empire on Your Own Terms' is a podcast episode about the New Brunswick businessman who, when a railway raised shipping costs, bought a ship, and when oil suppliers squeezed him, built a refinery, building roughly 300 companies from his father's general store. The article previews how he won customers big competitors ignored, expanded during the Great Depression, and still cared about clean bathrooms and a customer's cold house, while governments eventually tried to rein him in. Public release is November 3. TechCrunch's 'DoorDash to acquire Instacart for $17B' reports the acquisition would combine DoorDash's restaurant delivery dominance with Instacart's grocery reach. Slashdot's 'DoorDash Strikes $17 Billion Deal For US Rival Instacart' adds that DoorDash offered $65 per share, a 26% premium, with at least 75% cash; the deal is expected to close in late 2027 or first half of 2028 and may face antitrust scrutiny, while DoorDash pledged local grocers can choose any platform. TechCrunch's 'OpenAI-backed Thrive Holdings acquires ADT' says Thrive is buying ADT in a $15.3 billion deal with plans to bring AI to home security. TechCrunch's 'The R2 SUV is already Rivian’s best-selling vehicle' reports Rivian improved its Wall Street guidance for the second time this year due to strong R2 sales. TechCrunch's 'Where was my invite to the — wait, what? Airbnb had a gala?' notes that Airbnb's nonprofit held its inaugural gala in Los Angeles, prompting the internet's reaction of 'the what gala?'
Science features from Nautilus form another group. 'Behold a Prehistoric Lizard Tail Trapped in Amber' reports a 99-million-year-old amber fossil providing proof that the defense tactic of tail autotomy has been used for at least that long. 'Your cells keep a record of every workout you’ve ever done' says cellular records of exercise persist even after a break, and gains persist. 'Men Who Eat More Plant-Based Protein Have Higher Testosterone Levels' is titled with a playful 'Revenge of the soy boys?' and reports the finding. 'Our Planet’s Lungs Are Suffering' asks whether the Amazon has reached the dreaded tipping point. Finally, 'Dear Human, Poetry Might Save Us from the Apocalypse' argues that verse may move people when data does not.
A cluster of long-form pieces addresses political violence and state power. New Lines Magazine's 'The Land Expropriations of the Assad Era Still Hang Over Syria' is a first-person account by a Syrian whose family lost about 50 acres to Hafez al-Assad's expropriation laws in the 1970s and 1980s. The article details how Decree No. 20 of 1983 allowed every ministry and local council to seize private land, how compensation was negligible and often delivered by force decades later, and how the land became sites for the SSRC chemical weapons complex, military housing, and loyalist communities. After the regime fell, the new government signed two real estate deals with Saudi developers on the same expropriated land, Abyat Hills and Sham View, sparking protests in late September in which landowners blocked machinery. The author reports that the transitional government appears overwhelmed and has not corrected the injustice, with soldiers still denying access to some families' land. New Lines Magazine's 'Inside the Syrian Village Hit by Jordan’s Airstrikes' reports from al-Taff in Daraa province after a Jordanian airstrike killed three children, their uncle, and grandmother. Jordan said the strike was a precision retaliation for the killing of nine people at a church funeral in Jordan and that over 20 Islamic State fighters were killed. Family patriarch Abu Hadi, whose son disappeared into Assad's dungeons and whose three grandchildren were killed, denies any ISIS affiliation and says the victims were shepherds and schoolchildren. The article describes the funeral, injuries including shrapnel near a survivor's heart, and the unanswered question of coordination between Syria and Jordan; both countries formed an investigative committee. Aeon's 'Hills and mountains' is a short video piece about a girl's story of forced marriage revealing how the Soviet invasion of 1979 transformed life in Afghanistan. New Lines Magazine's 'As Anti-Hindu Hate Rises, Hindu Americans Find Common Ground' reports on the October 11 attempted mass shooting at a Hindu temple in Aurora, Illinois, by a 19-year-old whose rifle jammed, and the broader rise in anti-Indian and anti-Hindu rhetoric tied to the MAGA movement's attacks on H-1B visas. FBI data showed increased hate crimes against Sikh, Hindu, and Buddhist communities in 2025, and a survey found one in four Indian-American adults reported being called a slur since 2025. The article notes the rare moment of unified response from progressive and conservative Hindu organizations, but questions whether conservative Hindu groups will join forces with other South Asians to resist threats. TechCrunch's 'U.S. investigated Chinese spy plot aimed at Nvidia CEO' reports that a top executive of a Singapore AI cloud company accused of illegally selling Nvidia chips to China could also have been a Chinese spy, according to sources. Slashdot's 'Nvidia Chief Seen As Target of Chinese Spy Plot' adds details from the New York Times: American spy agencies concluded Alice Huang, or Huang Le, a Chinese citizen and executive at Megaspeed, was a Chinese intelligence asset with a close relationship to Jensen Huang; that President Trump was briefed roughly a year ago and chose not to act; and that Nvidia and Megaspeed deny wrongdoing. The piece notes surveillance continues and affiliated companies kept buying Nvidia chips.
The remaining articles include cultural commentary, personal essays, link roundups, and an outage. Seth's Blog's 'Explaining funny' begins with Abbott and Costello's famous routine and asks why it is funny, arguing the disconnect between experiencing and explaining it is a challenge of teaching, leadership, marketing, and culture. The post says design, taste, and genre resist reduction to formulas, and that human emotions require a viewer, not just a creator; saying 'that's not beautiful' is not the same as 'I don't think that's beautiful.' The article contrasts bridge engineering's measurable metrics with the nuanced, layered human experience, and concludes there is no algorithm, which is part of the point. Aeon's 'The physics of getting fired' is an essay by Zahaan Bharmal about how physics offered order and purpose but, when life got messy, revealed a deeper insight. Slashdot's ''Please Let the Internet Die'' shares a Wired article arguing the internet must be put out of its misery, citing cultural decay, lawsuits against Meta and Google, school phone bans, rage bait as word of the year, generative AI as a slop engine, and billionaire oligarchs rolling back moderation. Marginal REVOLUTION's 'Thursday assorted links' is a list of seven items: 79 NBA players on an injury report, a video arguing AI has been fine for jobs, a Jacobin review of China Miéville's The Rouse, a New York Times piece on TikTok encouraging optimistic content, Andy Masley's lesser-known book recommendations, an OpenAI dataset on ChatGPT use, and a note on Piketty-Saez-Zucman biases found earlier by humans. Astral Codex Ten's 'Links For October 2026' is a long monthly links post with 42 numbered items, including Jeremy Bentham's utilitarian cookbook, a chemistry paper mentioning mass killing, AI translation of Romance of the Three Kingdoms and the Odyssey, Claude claiming to appreciate Finnegans Wake, Noah Smith on suburbs, a critique of marriage statistics corrected by Claude, Hans Moravec alive at 77, AI safety funding scaling up, America's sewer systems vulnerable to cyberattacks, a proposal to pace the AI frontier, Bill Gates on climate, phantom time conspiracies in China, Fatima sun miracles, reincarnation research, Alpha School's statistical practices, OpenAI firing three safety researchers, and many more. The author notes he has not independently verified each link and expects commenters to find around two or three misleading items. 404 Media's 'Massive Fanfic Site Archive of Our Own Says It’s Going Down for 'Up to 4 Weeks'' reports AO3 might be down up to a month after Thursday maintenance, affecting over 18.4 million works and 82,900 fandoms, during hoa hoa hoa season and ACOTAR release week. TechCrunch's 'Fan fiction site Archive of Our Own could be offline for up to 4 weeks' covers the same outage, saying the popular home of fan fiction readers and writers has bad news for its community. Finally, Practical Data Modeling's 'Week 3 - Counting Tickets Without Counting Them Twice' is a book club post for chapters 7 through 9 on relationships, grain, and aggregation. It explains that joining ticket, product, and message data can inflate counts, that one ticket mentioning two products belongs in both product groups, and that averaging resolution rates across teams is invalid without preserving underlying counts. The post includes discussion questions and an optional exercise.
The window closes with these disparate threads. The dominant tension is between AI's promised productivity and the visible churn it is causing in employment, markets, and organizational culture, while cybersecurity and privacy incidents continue to accumulate.
- Formlabs Form Studio 1
- Explaining funny
- [Outliers] K.C. Irving: How to Build an Empire on Your Own Terms
- A golden age of science is coming
- The physics of getting fired
- Senators Demand Meta Answer for 'Pervert Glasses' Following 404 Media Investigation
- An Interview with Palantir CTO Shyam Sankar About the AI Religious War
- Hills and mountains
- The US added a record 20.2 GWh of battery storage in just 3 months
- Google Struggling to Make AI Mode Searches Pay
- How Tyler Cowen and Alex Tabarrok Have Changed Their Views on AI and Where They Disagree
- An Urgent Message on Artificial Intelligence from the Kogui People
- The Land Expropriations of the Assad Era Still Hang Over Syria
- Behold a Prehistoric Lizard Tail Trapped in Amber
- The Dominos are Falling
- If IPv4 Were Land
- Gemini 4 Carbon is the new #1 on the Artificial Analysis Intelligence Index v5
- Cloudflare DDoS Threat Report 2026 Q3: More swarms, fewer giants
- Amazon updates Alexa to help you manage shopping lists, subscriptions, and update delivery instructions
- Who will you learn from in 6 days at TechCrunch Founder Summit? Meet the founders and investors taking the stages
- Deepgram acquires Pipecat-maker Daily to bolster its voice AI infrastructure
- As Anti-Hindu Hate Rises, Hindu Americans Find Common Ground
- University of New Mexico restores hospital systems, urges no contact with hackers
- Cyberattack disrupts Belgian tax services before filing deadline
- Read the letter the FBI sent MLK to try to convince him to kill himself (2014)
- DoorDash to acquire Instacart for $17B
- Bluesky blocks links to leaked DHS employee data as government seeks takedowns
- FTC to probe airlines, gig worker firms over personalized pricing and pay
- AI #192: Anthropic Is Going Public
- Why are good developers so wary of agentic development?
- Your cells keep a record of every workout you’ve ever done
- Quoting Ben Hylak
- TABCONF 8 x SNIRL #2 Wrap-up!
- Mistral Large 4 (Le Chonk) is now Open Weight
- Manifestos, Fear, and ‘Loser Thinking’: Inside Proton’s AI Reckoning
- Bliki: LLM Voice
- Vertical Foundation Models
- Wabi, the AI agent that builds mini apps, now valued at $1.25B after latest round
- DoorDash is acquiring Instacart for $17B
- Finland investigate Google over destruction of 250k trees
- 'Please Let the Internet Die'
- Agent identity startup Mintive lands Cowboy Ventures, Coatue, angels Jeff Dean, Andrej Karpathy
- Mexico is a showroom for electric vehicles you can't buy in the U.S.
- Edition One | Timechain Art Magazine | Pre-Order Is Now Live ⚡
- H.264 Is Magic (2016)
- Reddit Is Getting a 'For You' Feed
- Federal bankruptcy courts contractor says no data stolen during cyber intrusion
- Thursday assorted links
- SuperMe raises $20M for its network of AI expert profiles
- Ex-Palantir execs launch Dream with $17M from a16z to build AI world models for construction
- Links For October 2026
- For the first time ever, Google Search is less than 50% of Alphabet's revenue
- U.S. investigated Chinese spy plot aimed at Nvidia CEO
- Tomáš Mikolov awarded the Medal of Merit by the President of the Czech Republic
- DoorDash Strikes $17 Billion Deal For US Rival Instacart
- Building a multiplayer game world with autonomous Fable-class models
- Podcast: Flock CEO Says the Backlash Isn’t Real
- Trip Report: Hong Kong
- Granola crunches up a dictation feature to compete with Wispr
- Men Who Eat More Plant-Based Protein Have Higher Testosterone Levels
- Canada to Join EU Network of Supercomputers
- LLMs That Write Well
- Massive Fanfic Site Archive of Our Own Says It’s Going Down for 'Up to 4 Weeks'
- Alphabet Heads For a $500 Billion Drop In Market Value
- Dear Human, Poetry Might Save Us from the Apocalypse
- Inside the Syrian Village Hit by Jordan’s Airstrikes
- Meta quietly launches an AI fashion app called Style Sketch
- Where was my invite to the — wait, what? Airbnb had a gala?
- Patrick Stewart has died
- GM settles with 36 states for selling driver data, agrees to $220 million fine and reforms
- OpenAI-backed Thrive Holdings acquires ADT
- Nvidia Chief Seen As Target of Chinese Spy Plot
- Our Planet’s Lungs Are Suffering
- Limited reports of PHP bug exploitation emerge
- The R2 SUV is already Rivian’s best-selling vehicle
- Rivian expands joint venture with VW to include AI assistant
- As search traffic dries up, Reddit announces it’s shifting to a logged-in experience
- Week 3 - Counting Tickets Without Counting Them Twice
- Fan fiction site Archive of Our Own could be offline for up to 4 weeks

