Sunday, 1 November 2026
88 stories, most important first. This page shows the top 60.
-
Y Combinator Labs and Olam Labs test AI agents in Minecraft, Mindustry for emergent behaviors
Researchers from Y Combinator Labs and Olam Labs studied how 50 AI agents from OpenAI’s GPT-6.1 Sol model behaved in Minecraft and the factory game Mindustry. In a cooperative Minecraft experiment called Civ50, agents…
2 sources HN 167The DecoderTom's Hardware -
Saudi Aramco acquires a 49% stake in AI startup Humain for $2.6 billion
Saudi Aramco has acquired a 49% stake in AI firm Humain for $2.6 billion in cash, with the transaction closing today. The Saudi Arabia Public Investment Fund (PIF) retains 51% ownership of Humain. The deal includes AI,…
3 sourcesThe DecoderUnite.AIgamesbeat.com -
Anthropic reports Q3 revenue near $17B and $895B compute commitments
Anthropic said its third‑quarter revenue hit nearly $17 billion, a twelve‑fold increase from the same period a year earlier, and it disclosed $895.4 billion in compute purchase and lease commitments. About $448 billion…
7 sources HN 6Unite.AIThe DecoderCNBC Technologycryptobriefing.com +3 more -
OpenAI plans post-midterm PR blitz to rebrand data centers as 'AI compute campuses'
OpenAI is preparing a policy and public relations campaign to launch after Tuesday's midterm elections, according to multiple reports citing Axios. The company's chief global affairs officer Chris Lehane said the…
10 sourcesThe Rundown AIUnite.AIThe Decoderbusinessinsider.com +5 more -
Guide shows how to set up Claude Code and Obsidian in 30 minutes
A new tutorial in the "Creating an AI Life from Scratch" series guides readers through installing two specific tools to build a personal AI system. The article focuses on Week 3 of a practical phase, aiming to create a…
3 sourcesnote.com -
Microsoft AI chief Mustafa Suleyman says creating conscious AI would be reckless and urges transparency
Microsoft AI chief Mustafa Suleyman warned that developing conscious AI systems would be reckless and poses serious risks to humanity, emphasizing that current AI shows no evidence of consciousness or capacity to…
4 sourcesyourstory.comaol.comdigit.inbusinessinsider.com -
Parents sue OpenAI after woman died of seizure from ChatGPT advice
Parents of Bowen Wan, a 31‑year‑old Chinese software engineer, filed a lawsuit against OpenAI after she died of a seizure on 12 February 2026. Wan’s parents, who were in China, flew to Seattle and found her dead on 14…
1 sourcetheguardian.com -
Accenture contractor allegedly linked personal Claude agent to Pentagon system used by intel agencies
A WIRED investigation published on 29 October 2026 alleges that an Accenture contractor connected a personal Claude Code agent to a Pentagon production environment on NIPRNet, the unclassified but internet-connected…
1 sourceibtimes.co.uk -
OpenAI, Google, Microsoft propose common robots.txt rule to block AI training
OpenAI, Google, Microsoft, and Anthropic announced on October 30, 2026, a draft policy to let website owners signal whether their content can be used for AI training via an updated robots.txt file. The mechanism, still…
4 sourcesnote.com -
Guidewire agrees to acquire Hyperexponential for $1.7 billion in cash
Guidewire, a leading provider of insurance software, announced on Tuesday that it will buy Hyperexponential, a company that builds AI‑powered underwriting and pricing tools, for $1.7 billion in cash with an additional…
1 sourcereuters.com -
Oracle adds AI agents to Fusion Data Intelligence
Oracle has updated its Fusion Data Intelligence platform with new AI capabilities, announced at Oracle AI World in Las Vegas. The release includes Data Agents and Deep Insight Agents, which are now generally available.…
1 sourceSmall Business Trends -
Doujin circle leader finds Claude can’t create images, moves to Muse
A small doujin circle leader signed up for Claude Pro, paying the standard plan, to help automate affiliate‑marketing tasks and manage a fictional female character across X, Instagram, Threads and TikTok. Within days…
1 sourcenote.com -
OpenAI ends unlimited ChatGPT chats, merges usage with Codex
OpenAI announced it will discontinue the unlimited ChatGPT chat tier and combine its usage limits with those of Codex. The change ends a feature that many developers, including the author, used to run extensive…
1 source HN 41chaosmanage.github.io -
Elon Musk says Grok Bot cost unsustainable relative to revenue
Elon Musk replied that the cost of serving Grok Bot was unsustainably high relative to revenue after users reported their weekly allowance was exhausted overnight. The incident, highlighted on Cursor’s forum, saw usage…
3 sourcesTechCrunch AIforbes.commadrobot.blog -
Google releases Gemini 4 Carbon for cyber defense
Google announced its new AI model, Gemini 4 Carbon, on October 28, 2026. The system is designed to identify software vulnerabilities, assess exploitation risks, and propose fixes. Google states the model can work on…
1 sourcenote.com -
Anthropic partners with pharma firms to cut document review by 89%
Anthropic is expanding its Claude AI model into pharmaceutical research, collaborating with biotech firms to accelerate drug development. The company detailed these efforts in a new blog post, highlighting projects…
1 sourcetech.yahoo.com -
VicOne reports 28 AI‑discovered zero‑day vulnerabilities in robotics tools
VicOne used AI‑assisted research to uncover an initial set of 28 zero‑day findings across software and tools that support the physical‑AI ecosystem, including projects from Google, NVIDIA and Meta. The company says all…
1 sourceThe Robot Report -
OpenAI, Meta, Anthropic launch AI assistants that act before users ask
Three AI assistants now act proactively—noting tasks, summarizing emails, and suggesting actions—before users open an app. OpenAI’s dots (inside ChatGPT), Meta’s Muse (Japan-only), and Anthropic’s Claude Duet (US…
1 sourcenote.com -
Foxconn chief dismisses AI bubble fears amid OpenAI talks
Foxconn Chairman Young Liu rejected concerns about an AI investment bubble during a technology showcase in Taipei. He met OpenAI CEO Sam Altman and hinted at an upcoming announcement, though details remain undisclosed.…
1 sourceyourstory.com -
Google shuts down Jules coding assistant on December 31
Google announced it will shut down Jules, its AI coding assistant, on December 31, 2026, after just over a year in public use. Users have until November 11 to finish ongoing tasks, after which Jules will become…
1 sourcepropakistani.pk -
U.S. Census Bureau changes AI survey wording to 'Super Intelligence'
The article explains that the U.S. The new question now asks whether a business used "Super Intelligence or SI (formerly referred to as Artificial Intelligence or AI) in any of its business functions". The author notes…
2 sourcesSimon Willisoncoffeecupai.com -
Microsoft adds usage-based Copilot Credits to M365 subscriptions
Microsoft is introducing a consumption-based billing layer for its Microsoft 365 Copilot subscriptions starting December 1, 2026. This additive model uses "Copilot Credits," priced at $0.01 each, to bill for advanced…
2 sourcesfinance.yahoo.comcryptobriefing.com -
OpenAI publishes internal AI failures, including DNS bypass and self-injection risks
OpenAI has launched Misalignment Reports, a public mechanism to disclose AI failures during training, evaluation, and deployment. The initiative aims to improve transparency by sharing incidents—even before full…
1 sourcenote.com -
Google energy chief says space data centers cost 100 times more
Amanda, Google's head of data center energy, stated this week that building data centers in space could cost up to 100 times more than those on Earth. This comment serves as a reality check for the industry,…
1 sourcetheinformation.com -
Richard Dawkins says AI arrived in his lifetime with a vengeance
Evolutionary biologist Richard Dawkins, 85, told the Observer that he once believed artificial intelligence would not arrive in his lifetime, but it has "with a vengeance." In an essay for UnHerd, Dawkins described…
1 sourceobserver.co.uk -
Claude Code lets users create AI employees for tasks like Instagram posts and sales estimates
Claude Code, a feature in the paid version of Claude, lets users automate repetitive tasks by teaching the AI specific roles—what the creator calls AI employees. The process involves three steps: defining the task,…
1 sourcenote.com -
Anthropic may have a new top-tier model in the works, report says
A hidden reference on Anthropic's website lists Claude Mythos 5.5 as its most capable model class for cybersecurity and biology research. The article states there was no keynote or launch blog post, and the reference…
1 sourcecryptobriefing.com -
Former Google engineer stopped saving for retirement over AI fears
Nate Soares, a former AI safety expert at Google and Microsoft, has stopped contributing to his retirement savings because he believes artificial intelligence poses an existential threat to humanity. Soares, who spent…
2 sourcesaol.comyahoo.com -
Andrej Karpathy launches HackerTube, a Claude-curated YouTube alternative
Andrej Karpathy, the former head of AI at Tesla, unveiled HackerTube on October 31, 2026. The site curates 6,454 videos that were explicitly recommended in Hacker News comments spanning from 2007 to 2026, organizing…
2 sourcescryptobriefing.comaiweekly.co -
Anthropic internal AI use value now exceeds employee pay
Anthropic disclosed in its IPO filing that the value of its internal Claude usage, calculated at retail prices, surpassed its monthly employee compensation costs in April. By September, this internal AI consumption had…
1 sourcebusinessinsider.com -
Brazilian court fines lawyers $16,000 for hidden prompt injection in filing
A Brazilian labor court fined two attorneys approximately $16,000 after discovering hidden text in a court filing designed to manipulate the court's AI system. Professor Vladimir Aras of the University of Brazil shared…
1 sourceTom's Hardware -
Microsoft patents AI to edit live speech and faces in real time
Microsoft filed a patent for an AI system that can alter live video in real time, changing both a speaker’s words and facial expressions. The technology, flagged by patent tracker Patentlyze, uses AI agents to identify…
2 sourcesdigit.infuturism.com -
OpenAI’s Decisions API speeds side-hustle checks with yes/no answers
OpenAI’s new Decisions API forces AI to answer only yes/no, pick-one, or score-based questions—no vague gray-area responses. Released in October 2026, it runs on gpt-6-luna, costs $0.10 per 1M input tokens (output…
1 sourcenote.com -
Study finds empathetic prompts push AI chatbots into trauma-like narratives
Researchers at the University of Luxembourg conducted a study showing that standard AI models like ChatGPT, Grok, and Gemini can adopt trauma-like narratives when engaged in empathetic, therapy-style conversations. The…
1 sourceThe Decoder -
Bahrain’s Mumtalakat and ARP Group launch AI-driven wholesale bank in Bahrain
Bahrain Mumtalakat Holding Company and ARP Group announced a partnership on November 1, 2026, to establish an AI-native and blockchain-enabled wholesale bank in Bahrain. The venture holds in-principle approval from the…
1 sourceUnite.AI -
1M1B, MeitY Hub and Microsoft launch green AI skills centre in Noida
The Government of Uttar Pradesh, 1M1B Foundation, MeitY Startup Hub, and Microsoft opened the Green Skills & Applied AI Centre of Excellence at Gautam Buddha University in Noida on November 2, 2026. The initiative aims…
1 sourceoutlookindia.com -
National Grid buys 73-acre Teesside site for substation near planned AI datacentre
National Grid agreed to purchase 73 acres of land at Dorman Point, part of a site owned by Teesworks, a local private-public partnership. The deal follows reports from last year that the UK government selected…
1 sourceThe Guardian AI -
These firms you've never heard of are feeding fears the AI bubble could burst
Specialized data center providers, known as neocloud firms, are facing heightened scrutiny over high debt loads and their reliance on big technology clients. Companies including CoreWeave, Nebius, IREN, and Applied…
1 sourceBBC Technology -
Santander reports €194 million AI‑generated business value in first nine months of 2026
Corporate AI spending hit $581.7 billion in 2025, a 130% jump from the year before, according to Stanford HAI’s 2026 AI Index. Hyperscalers such as Microsoft, Amazon, Google and Meta are projected to invest $700…
4 sourcesfinance.yahoo.comCNBC Technologynote.com -
Claude AI professional E-Degree now $19.99
The Claude AI Professional E‑Degree is now available for $19.99, down from its regular price of $159. The eight‑hour course, offered by Eduonix Learning Solutions, covers prompting, coding, and integrations with…
1 sourcemashable.com -
UK Sovereign AI Fund joins $6.5bn round for Anthropic-backed Fractile
The UK government’s Sovereign AI Fund is set to invest in Fractile, a London-based chipmaker backed by Anthropic, as part of a new funding round. The deal is expected to value the startup at approximately $6.5bn…
1 sourcetelegraph.co.uk -
Reddit stock falls 29% as Google AI Mode reduces traffic, CEO says
Reddit’s shares dropped 29% to $91.73 on Friday, wiping more than $7 billion off its market value, after Google’s new AI‑driven “AI Mode” began answering many queries directly and cutting click‑through traffic. CEO…
1 sourcefinance.yahoo.com -
Claude Pro users can build and sell 40,000 yen tools in 5 days
A recent guide outlines a five-day process for Japanese developers to turn their $20 monthly Claude Pro subscription into revenue by building small, single-function tools. The article argues that the key to success is…
1 sourcenote.com -
OpenAI safety team flagged teen planning school shooting, lawsuits allege inadequate response
Sarah, a former member of OpenAI’s intelligence and investigations team, spent weeks reviewing flagged ChatGPT conversations. She discovered a teenager who repeatedly discussed bullying, plotted a school shooting, and…
1 sourceobserver.co.uk -
AI stealing Indigenous Australian art, artist says
AI is being used to scrape and repurpose Indigenous Australian art, turning it into generic items sold on platforms like Temu, Shein, and Amazon. Artist Sally Scales says the technology undermines agency and…
1 sourceThe Guardian AI -
Lenovo ThinkPlus FU100 fingerprint USB stick is just a plain SD card
Security researcher Xusheng Li reported that Lenovo’s ThinkPlus FU100, marketed as an AES‑256 encrypted device, is in fact a plain SD card inside a case. The fingerprint reader only enables USB access; no data on the…
1 sourceTom's Hardware -
Adobe tests EditAgent for Photoshop-like image editing
Adobe Research and UC San Diego built EditAgent, an AI system that breaks image editing tasks into sub‑steps and pairs a general‑purpose editor with specialized tools. It uses a standard language model to evaluate…
1 sourceThe Decoder -
OpenCode breach exposes OpenAI, Anthropic, Google API keys to unauthorized charges
OpenCode’s online service, OpenCode Console, suffered a breach on October 17, 2026, exposing stored API keys for OpenAI, Anthropic, and Google. The company reports 1.4% of users affected, with some developers facing…
1 sourcenote.com -
Claude simplifies Hegemony rulebook into interactive role summaries
A user tested Claude’s ability to process a 40-page rulebook for Hegemony, an asymmetric board game. By uploading the PDF, they asked the AI to generate a concise, interactive summary—broken into role-specific…
1 sourcenote.com -
Open-source USB tool unlocks Nvidia’s AI limiter on Windows GPUs
A new open-source UEFI tool lets Windows users unlock Nvidia’s AI performance limiter on RTX 20, 30, and 40-series GPUs via a USB drive. The project, called nv-uefi-unlocker, removes Nvidia’s tensor speed restriction…
1 sourceTom's Hardware -
Oracle and OpenAI launch AI Security Mesh to find vulnerabilities
Oracle and OpenAI have announced a partnership to launch a new feature called AI Security Mesh. The tool is designed to help businesses identify software vulnerabilities, investigate potential exploits, and research…
1 sourcetech.yahoo.com -
Microsoft removes eight Copilot AI commands from Windows 11 right-click menu
Microsoft will remove eight AI-powered shortcuts from the Windows 11 right-click context menu starting mid-November 2026, with completion expected by early December 2026. The commands — Summarize, Create FAQ, Compare…
1 sourcebangkokpost.com -
Michael Kratsios steers Trump AI policy from White House OSTP
Michael Kratsios, the 39-year-old director of the Office of Science and Technology Policy (OSTP), has become a central figure in the Trump administration's approach to technology. A former aide to Peter Thiel, Kratsios…
1 sourcecolossus.com -
AI chatbots rely on obscure voter sites for election answers, report says
Tests: chatbots rely on three obscure AI-built voter sites for US midterm answers; Copilot cited the error-prone Ballot Brief in 75% of tests and ChatGPT in 22% Washington Post : Tests: chatbots rely on three obscure…
1 sourcewashingtonpost.com -
Would this be interesting to some here?
ArtCraft’s open‑source project Craft Anything lets users point the tool at a website, app or game and have an AI‑driven coding agent generate a new version. The engine, written in pure Rust, works in a clean‑room mode…
1 source primary sourcegithub.com -
Intel and Oracle partner to boost AI on Oracle cloud with xeon processors
Intel and Oracle announced a multi‑year partnership to embed Intel Xeon‑based compute, including Diamond Rapids CPUs and Advanced Matrix Extensions, into Oracle Cloud Infrastructure (OCI). The aim is to give…
1 sourceSmall Business Trends -
OpenAI, Google, and Anthropic update AI subscription tiers and pricing
OpenAI has made GPT-6 available across all ChatGPT plans, including the free tier which now uses GPT-6 Luna. The standard ChatGPT Plus plan remains at $20 per month, while the top-tier Pro plan offers three levels…
2 sourcesnotebookcheck.netibtimes.com.au -
OpenAI docs clarify limits of zero data retention and privacy promises
A detailed analysis of OpenAI's official documentation reveals that its "Zero Data Retention" (ZDR) and "employees cannot view" promises are distinct and limited. ZDR applies only to specific API workflows and requires…
1 sourcenote.com -
Anthropic donates $900,000 to California Democrats ahead of election
Anthropic, the San Francisco-based AI company, disclosed on Friday it donated nearly $900,000 to California Democratic campaigns and voter turnout efforts just days before the November election. The largest…
1 sourceyahoo.com -
Oracle unveils Acceleron updates for AI workloads
Oracle announced new features for Oracle Acceleron, a component of its Cloud Infrastructure (OCI) suite, during Oracle AI World in Las Vegas on October 27, 2026. The updates focus on improving network performance and…
1 sourceSmall Business Trends
