AI news for Thursday, 5 November 2026
Stories 121 to 180 of 195, most important first. Back to the first page →
-
Together AI releases Together Compiler, agents re-optimize inference up to 2.6x faster
Together AI launched Together Compiler (TKCC), an agent-driven system that continuously rewrites a model's forward pass as production traffic shifts. The system runs on the company's serverless endpoints and has been…
1 sourceMarkTechPost -
Google Staffers Say They’re Being Silenced From Telling the Public the Truth About AI
The Washington Post reported that employees, who signed such clauses as a condition of employment, feel intimidated and unable to inform the public about potential harms or rapid advancements in the technology. One…
1 sourcefuturism.com -
Researchers test six AI tools on a 102-article scientific literature review
A new study on arXiv evaluates how well six AI tools handle the full process of reviewing scientific literature, not just individual tasks. The benchmark uses a published review of the Spanish plume weather pattern and…
1 source primary sourcearXiv cs.AI -
Microsoft reports positive free cash flow as only AI hyperscaler in Q3
Microsoft Corporation posted a clean Q1 FY2027 earnings beat, buoyed by accelerating Azure revenue and strong guidance for the second quarter. The company’s Agents and Infra segment recorded robust growth, while the…
1 sourceseekingalpha.com -
Agiloft adds three AI agents for contract lifecycle tasks
Agiloft launched three AI agents on November 5, 2026, to automate multi-step contract tasks within its contract lifecycle management (CLM) platform. The Bulk Update Agent edits up to three fields across multiple…
1 sourceUnite.AI -
Lotus Microsystems CEO explains power delivery’s role in AI scaling
Hans Hasselby-Andersen, CEO of Lotus Microsystems, argues that power delivery and thermal management are critical bottlenecks in AI infrastructure scaling. His background spans wireless systems, semiconductor…
1 sourceUnite.AI -
Pliant raises $80 million Series C to offer B2B payments platform to AI agents
Berlin‑based fintech Pliant announced an $80 million Series C round, led by Seaya Nova and Speedinvest, with participation from Illuminate Financial, Motive Partners and Molten Ventures. The capital will fund expansion…
1 sourcethenextweb.com -
Ontario to approve AI data centres soon despite moratorium calls
Ontario’s Economic Development Minister Vic Fedeli stated that the Ford government will approve its first batch of new AI data centres "sooner rather than later." This decision comes as the province faces increasing…
1 sourceglobalnews.ca -
Anthropic announces FlashProve to prove AI answers
Anthropic's research team has announced a mechanism called FlashProve that uses zkSNARKs to prove that an AI's answer was generated by a specific model. The system can produce proofs for about 110 tokens per second per…
1 sourcenote.com -
Luma Health launches Agent Launchpad for procedure prep
Luma Health released Agent Launchpad on November 5, 2026, an AI-agent system for multistep procedure preparation. The product automates outreach, follow-up and readiness tracking for colonoscopy, knee replacement,…
1 sourceUnite.AI -
Le Chonk brings the brains. Studio puts it to work.
The model is aimed at enterprise use, offering advanced coding, defensive cybersecurity, and complex document reasoning. Studio now supports managed agents, visual workflow graphs, event triggers, and enterprise…
1 source primary sourceMistral AI -
Google turns sketches into interactive apps with Gemini Canvas
Google demonstrated how users can upload a photograph of a hand-drawn app sketch to Gemini Canvas and generate a functional prototype in minutes. The process involves uploading the image, selecting Canvas, and…
1 sourceUnite.AI -
Kojima warns AI may erase the joy from creative work
Hideo Kojima, creator of Metal Gear and Death Stranding, spoke to GameSpark about AI’s role in game development. He acknowledges AI’s efficiency—cutting deadlines and boosting productivity—but warns it risks stripping…
1 sourceautomaton-media.com -
Guardian teams with Sierra AI to handle 1M annual dental-benefit calls
Guardian announced on November 5, 2026, that it will use Sierra’s AI-native customer engagement platform to automate 1 million annual dental-benefit verification calls, among other high-volume service requests. The…
1 sourceUnite.AI -
Prix Goncourt winner admits using AI in novel three times
Philippe Jaenada, the 2026 winner of France’s Prix Goncourt for L’Inconnue du quai de Javel, disclosed on Wednesday that he used AI three times while writing his novel. He inserted a ChatGPT-generated sentence as a…
1 sourceThe Guardian AI -
Social Media Examiner shares tips for training AI agents to run projects independently
Social Media Examiner published a guide on Friday outlining five rules for setting up AI agents to manage projects without constant human oversight. The post focuses on using Claude and ChatGPT to automate workflows,…
1 sourceSocial Media Examiner -
UK found to be dangerously dependent on China for critical minerals
A parliamentary business select committee warned that the United Kingdom is "dangerously" reliant on China for rare earths and other critical minerals needed for defence, renewable‑energy infrastructure and…
1 sourceThe Guardian AI -
AI’s Next Trick? Speaking Dolphin
Denise Herzing, founder of the Wild Dolphin Project, has used AI to analyze dolphin clicks and whistles after a 40‑year research effort that began in 1985 in the Bahamas. Her team now works with Georgia Tech and Google…
1 sourceWired AI -
Researchers propose Multi-Item Scoring for faster whole-page ranking
Researchers introduced Multi-Item Scoring (MIS), a new method for ranking web pages that balances efficiency and quality. Traditional non-autoregressive (NAR) approaches score items independently, missing sequential…
1 source primary sourcearXiv cs.AI -
AI boom strains global supply chains with hidden costs
Supply chain disruptions from AI’s infrastructure demands are reshaping global manufacturing, according to economist Willy Shih of Harvard Business School. This surge exacerbates pre-existing pressures from tariffs,…
1 sourceThe Verge AI -
Zapier lists 8 free or low-cost AI courses for beginners in 2026
Zapier’s blog highlights eight beginner-friendly AI courses, ranging from free to under $100, to help professionals integrate AI into their work. The list includes options like AI for Everyone (DeepLearning.AI,…
1 sourceZapier Blog -
Reddit’s founder warns social media apps are ‘not cool anymore’ due to AI
He cited the rise of AI-driven content and automated interactions, which he calls “AI slop,” as the reason. Ohanian, who left Reddit in 2020 amid content moderation debates, now focuses on sports communities as a…
1 sourceWired AI -
Discovered Materials offers $1M challenge to AI for five Materials science problems
Discovered Materials, a Y Combinator spring 2026 startup, announced a $1M prize pool for AI models to solve five unsolved materials science problems. The challenges include a theory of cuprate superconductivity, a…
1 sourceTechCrunch AI -
IT novice uses Cowork to generate Claude Code prompts for web projects
A non-technical IT worker shares how they use Cowork to draft precise prompts for Claude Code, an AI assistant, to build a fictional coffee shop website. Instead of writing instructions from scratch, they break tasks…
1 sourcenote.com -
Meta's Muse composes song and produces vertical lyric video
Meta’s AI agent Muse, released in North America in September 2026 and in Japan in October, can handle end‑to‑end creative tasks via chat. In a two‑day test the author connected Muse to Suno’s music‑generation service,…
1 sourcenote.com -
Comedian Husk says xAI, OpenAI, and Anthropic want to work with him
Comedian Husk, who has gained 2 million followers by mocking AI chatbots, told the WIRED World Fair that major AI companies are seeking collaborations. He claimed Elon Musk’s xAI offered him a Tesla for a week if he…
1 sourceWired AI -
Finance teams automate functions but miss cash-flow outcomes, research finds
Finance teams are rapidly adopting AI, with 66.5% increasing investment and an average of 2.43 functions using it, up 36% year-over-year, per Auditoria’s 2026 research. Yet only 21% report measurable results, the study…
1 sourceUnite.AI -
Qwen improves LLM-based OCR correction with veto gate
Researchers at Qwen have developed a new method called EditVeto to improve the accuracy of language models (LLMs) in correcting Optical Character Recognition (OCR) errors found in historical documents. The technique,…
1 source primary sourcearXiv cs.CL -
AutoStore agrees to buy Logiqs BV for up to €30 million
AutoStore announced it will acquire Logiqs BV, a Netherlands‑based developer of roaming pallet robots, for up to €30 million, with the deal expected to close in the fourth quarter of 2026. The purchase adds…
1 sourceUnite.AI -
SpaceXSI’s Grok Bot adds phone call feature for paid users
SpaceXSI’s Grok Bot now lets paid users make phone calls, including handling holds, booking appointments, and negotiating bills via U.S. numbers. Users remain legally responsible for all calls, with the bot disclosing…
1 sourceThe Rundown AI -
Cainiao opens global warehouse network for embodied AI training
Cainiao announced on November 5, 2026 that it is opening its global warehouse network to robotics partners and launching a logistics service for AI hardware companies. The initiative aims to create the world’s largest…
1 sourceUnite.AI -
Researchers introduce AGORA method for improving LLM judge reliability
A new approach called AGORA aims to improve the reliability of large language model (LLM) judges by accounting for their varying strengths. Unlike current methods that treat all judges equally, AGORA uses item-response…
1 source primary sourcearXiv cs.AI -
MIT researchers develop uq-daam to correct robot memory
mit researchers introduced uq‑daam, a memory system that identifies and corrects mislabeled object descriptions in robots. the system uses uncertainty quantification to detect when a robot’s vision‑language model (vml)…
1 source primary sourceMIT News on AI -
AI chooses Claude as most wanted for marriage: 3 votes
AI Matchmaking Results I asked four generative AIs—ChatGPT, Claude, Gemini, and Grok—to name their ideal marriage partner. The results were surprising. Claude received the highest number of votes (3), followed by…
1 sourcenote.com -
Opinion: CEOs face AI paradox of speed vs. irreversible bets
This essay argues that CEOs must balance AI’s rapid evolution with the risks of hasty, costly commitments. Meta’s $135 billion to $145 billion capital expenditure forecast for 2026 highlights AI’s escalating…
1 sourceUnite.AI -
Researchers introduce MA-SafeDiffuser for safer multi-agent AI planning
Researchers at an unnamed institution have proposed MA-SafeDiffuser, a new framework extending SafeDiffuser to multi-agent systems. The method embeds control barrier function constraints into diffusion-based trajectory…
1 source primary sourcearXiv cs.AI -
Meta AI warns Solana may face major outage-driven crash by January 2027
Meta’s AI model predicts Solana could suffer a severe downturn by January 1, 2027, if a prolonged consensus-layer failure occurs. The scenario involves a coordinated outage among high-stake validators—already flagged…
1 sourcefinance.yahoo.com -
HomeView will support third-party AI chatbots, suggests code
HomeView, a smart home device, may soon support third-party AI chatbots such as ChatGPT and Claude. Code in the iOS 27.3 beta suggests this feature is in development, with a leaker confirming an unused flag for…
1 source9to5mac.com -
How to view and delete AI memory in ChatGPT, Gemini, Claude, and Copilot
Four major AI assistants—ChatGPT, Gemini, Claude, and Microsoft 365 Copilot—store user conversations in memory features, which can persist even after deleting chats. The article outlines how to access and clear these…
1 sourcenote.com -
Anthropic’s Amodei reportedly predicted human-machine merging by 2040 in 2007
Dario Amodei, CEO of Anthropic, reportedly told a friend in 2007 that humans might merge with machines by 2040 as part of an AI-driven Singularity. According to the New York Times, Amodei described this vision to…
1 sourceijr.com -
Former BI engineer rewrites 25 incident response manuals with Notion AI and Claude in six weeks
A former BI engineer rewrote 25 outdated incident response manuals in six weeks using Notion AI and Claude, addressing a common issue in data teams: manuals that are either unused, incorrect, or too outdated to be…
1 sourcenote.com -
UK entry-level job postings mentioning AI up ~3x since early 2025, Adzuna says
Adzuna reports that UK entry-level job postings mentioning AI have risen about threefold since early 2025. Indeed adds that one in six vacancies in AI-exposed fields now mentions AI, up twofold year over year. The…
1 sourcebloomberg.com -
Activists across Europe plan two days of protests against datacentres
Protesters across Europe are planning two days of action against datacentres this week, with demonstrations set for Friday and Saturday in 15 countries. More than 100 groups, including Friends of the Earth, Greenpeace…
1 sourceThe Guardian AI -
Researchers introduce DeltaCata to predict enzyme mutation effects
DeltaCata is a new Siamese geometric graph neural network designed to predict how single‑amino‑acid substitutions change enzyme kinetic parameters such as turnover number (kcat) and Michaelis constant (Km). The authors…
1 source primary sourceNature Machine Learning -
OpenAI president says agentic coding tools and GPT-5.6 Sol sparked Q3 revenue inflection
OpenAI president Greg Brockman described the first half of 2026 as “pretty rough” for the company. He said the slowdown was offset by a notable “revenue inflection” in the third quarter. According to Brockman, the lift…
1 sourcewsj.com -
Study: LLMs prefer to answer over clarifying ambiguous references
A new study from arXiv cs.CL suggests that large language models (LLMs) often opt for quick answers rather than asking clarifying questions when faced with ambiguous references. The research, which uses the SIMMC 2.1…
1 source primary sourcearXiv cs.CL -
Dario Amodei discusses AI views shaped by GiveWell and rationalism
Dario Amodei, a figure in AI circles, shared his philosophical views on artificial intelligence with a friend at Princeton University in 2007. His perspectives were reportedly influenced by forums like GiveWell and…
1 source HN 8nytimes.com -
RAGCompass routes retrievers by reranking passages, improves answer F1 by 28%
Researchers introduced RAGCompass, a retriever routing method that reads retrieved passages directly instead of relying on query-only features. The approach contrastively fine-tunes a reranker against the answers each…
1 source primary sourcearXiv cs.CL -
Amazon says grocery growth fuels sales, AI and same‑day delivery boost spend
Amazon highlighted grocery as its main engine for cross‑category growth in its Q3 2026 earnings call. Fresh groceries are expanding three times faster than the rest of the store business and now represent roughly half…
1 sourceconsumergoods.com -
Anthropic’s Claude predicts Bitcoin may hit new high before Solana in late 2027–mid-2028
Anthropic’s AI model, Claude, forecasted Bitcoin would likely reach a new all-time high between late 2027 and mid-2028, based on current prices of $76,280 for Bitcoin and $90.83 for Solana. Bitcoin would need a 65.03%…
2 sources247wallst.com -
Llama.cpp adds support for xing4_0 model
A pull request merged into the llama.cpp repository on November 3, 2026, added support for the xing40 model. The change was co-authored by Sigbjørn Skjæret of Hugging Face and reviewed internally by maintainers. The…
1 source primary sourcegithub.com -
Anthropic and OpenAI may become coke vs. pepsi of AI, panel says
On a recent panel, Kevin Mahn, Willy Lee, Anna Rathbun and Shay Boloor compared Anthropic and OpenAI to Coke and Pepsi. The speakers argued that Anthropic offers a “sparkling water” experience while OpenAI provides a…
1 sourcefinance.yahoo.com -
Supermicro Super AI Station review: convertible DGX Station workstation tested
Supermicro Super AI Station Review: The AI Server That Doubles as a Workstation Today we are looking at Supermicro's Super AI Station, the company's neat take on a DGX Station System with a design that can stand as a…
1 sourceservethehome.com -
Judge dismisses direct copyright claim against Anthropic co‑founder Dario Amodei
On November 4, 2026 Judge Eumi K. Lee of the U.S. District Court for the Northern District of California dismissed the direct copyright‑infringement claim against Anthropic co‑founder Dario Amodei, granting leave to…
1 sourceUnite.AI -
Using AI to check weight meets requirements
I tested three AI assistants—ChatGPT, Gemini, and Claude—to determine if two packs of Costco's Salted Mackerel Fillets met a specified requirement of 2.32kg. I sent them the same photo of the order screen in the app…
1 sourcenote.com -
Researchers find deep neural speaker embeddings generalize across languages like humans
The research used Cantonese–English bilingual speech to test 18 state-of-the-art models, comparing their similarity judgments with human ratings and acoustic features. Results showed that embeddings accurately…
1 source primary sourcearXiv cs.CL -
Anyone knows anything about this Quantized K3.1m Worth it?
A Reddit user posted about a quantized, expert-pruned checkpoint derived from moonshotai/Kimi-K3.1-Pro, noting it removes 75% of routed experts and is intended for vLLM serving and evaluation smoke tests. The user…
1 source primary sourcehuggingface.co -
Nature Methods launches AI Bytes column to explain AI concepts for life sciences researchers
Nature Methods introduces AI Bytes, a new monthly column aimed at demystifying AI and deep learning for life sciences researchers. The initiative responds to growing reliance on AI in fields like genomics and protein…
1 source primary sourceNature Machine Learning -
An office worker describes using Claude, Gemini, and ChatGPT to create LINE stickers
Yu, an office worker who cannot draw, explains how they used AI tools to create LINE stickers that passed review and are available for sale. They describe a six-step process involving Claude for consultation and…
1 sourcenote.com -
Developer reports persistent errors and hallucinations in Cursor Composer 2.5
A developer shared a detailed account of frustrations while using Cursor’s Composer 2.5 for coding tasks. After over 100 hours of use, the author describes frequent issues where the agent deletes unrelated code, fails…
1 source HN 41orchidfiles.com
