Updated
Updated
AIThe Association for Human Mathematics requires members to take three vows opposing AI use in mathematics. Members must not provide technical labor, knowledge, consultation, or publicity to commercial AI companies, and must not publish AI-generated mathematical texts, including papers, referee reports, and lecture notes. Members of its AI-free caucus also must not use AI models in research.
AILauren Tan argues that with frontier AI agents, anyone can produce code at machine speed, so PR volume now signals productivity alongside impact. She says the bottleneck is trust in agent output, and that higher token costs are worth it compared with hiring many engineers. She frames the engineer's job as building the software-producing machine rather than writing code directly.
AIJoshua Achiam recommends a critical reading for thinking about the future of cybersecurity. The post links to a follow-up by Matthew Green, who clarifies a widely shared earlier reply about the topic. The source provides no further technical details or figures.
AIAnthropic has cut Sonnet 5.5 cache read pricing in half to $0.10 per million tokens, with input at $2 and output at $10 per million tokens. The company says this makes Sonnet 5.5 roughly 20% cheaper on most agentic work. The change applies to API usage only and does not alter Claude Code usage limits.
AIAlex Gibney directs a four-hour documentary about Elon Musk, according to this review. The reviewer finds the film engaging but says it tends toward the hyperbolic.
AIAllie K. Miller jokes that people obsess over web-based GEO while neglecting "EEO," or email engine optimization. She says she is drafting side notes to AI tools left and right, playfully suggesting email is a channel for influencing AI outputs.
AI🫠 Quoted post (@AndrewCurran_): Effective November 12th, 2026, abusive behavior towards Claude will be a violation of Anthropic's Usage Policy.
AIArtificial Analysis and Harvey released LAB-AA v1.1, which credits a legal task only when deliverables pass every rubric criterion with no material hallucinations. Grok 4.7 (xhigh) leads at 9.4%, ahead of Muse Spark 1.3 (max) at 8.9% and GPT-6 Astra (max) at 8.6%, while over 60% of otherwise passing results contained a material hallucination. The sharper reordering appears in the hallucination counts, where GPT-6 Astra averages 0.03 material hallucinations per task against 13.96 for Gemini 3.8 Flash (high).
AIQualcomm CEO Cristiano Amon discusses the coming AI smartphone supercycle, arguing phones will not disappear as agents use personal context. He also expects smart glasses to become the largest AI wearable category, and covers Qualcomm's Modular acquisition as an alternative to Nvidia's CUDA software and its data center strategy. The conversation, recorded live at the Snapdragon Summit in Hawaii, also covers 6G being designed for AI.
AINathan Lambert says he is about to discuss how open science can still address some of the hardest problems in AI safety and diffusion. He calls this the most complex engineering problem humanity has undertaken so far.
AITessl's talk at AI Native DevCon London argues that agent skills, which can be markdown files with instructions and bundled material, act as supply chain components that can shape agent behavior. The author says reading SKILL.md once is insufficient because risks can sit in supporting files, updates, and workspace trust settings. He identifies the danger as the combination of private context, untrusted content, and external communication, and cites research scanning roughly 4,000 public skills for issues including malware-like behavior.
AIArtificial Analysis reports that generating more output tokens does not necessarily yield a higher score. GPT-6 Astra (max) scored 8.6% using about 81k output tokens per task, while Grok 4.7 (xhigh) used roughly 180k yet scored lower. Three Claude models produced the most output tokens, about 202k to 562k per task, but scored between 2.8% and 6.4%.
AIArtificial Analysis reports that Kimi K3 (max) achieves a 93.0% Criterion Pass Rate but averages 2.09 material hallucinations per task. Muse Spark 1.3 (max) scores higher at 96.0% while averaging 1.68 material hallucinations per task, showing that completing criteria and avoiding hallucinations are distinct skills.
AITessl argues AI code review is slow because generated code outpaces trust, and proposes executable specs that let agents check preview environments against product intent. Its spec reviewer splits work between a planner agent that extracts requirements and parallel verifier agents that test each one against the code and base branch.
AITypeSafe AI's jev model was reported to be more calibrated than 40 other decision models tested on email classification. The main post itself only urges readers to stay calibrated, so no further figures or methodology are given.
AIActor Ben Affleck drew attention this week for explaining machine learning concepts, including convolutional neural networks, tensors, and transformers, in several recent interviews. He said he fine-tuned open video models by unfreezing weights and training only the last cinematic layer, using a dataset he built over about eight months for his startup. Affleck said he worries about students and learned helplessness more than Skynet, and predicted AI will be additive to the movie business.
AIA group of mathematicians led by Terence Tao has called for a boycott of OpenAI after the company released more than 700 AI-generated proof files at once. Tao and other Fields Medalists argue that mass-produced solutions undermine the discipline's focus on conceptual understanding, while Scott Aaronson contrasts this batch release with Anthropic's collaborative approach. The article reports that the internal model tested about 8,000 problems with roughly a five percent success rate.
AISK Hynix publicly acknowledged that 16-high HBM stacks are difficult, and the market is now prioritizing bandwidth over capacity. SemiAnalysis argues hybrid bonding's HBM case rested on stack height, which is no problem at 4-high and 8-high, so D2W hybrid bonding for HBM is now uncertain rather than merely delayed.
AIThe post says demo prep fell from 30 minutes to 5, note-taking from 15 minutes to 2, and CRM updates from 30 minutes to 7 per call. It estimates this saves about an hour per call, letting each rep take roughly two more calls a day while staying prepared.
AILlamaIndex says Markdown has become a universal representation between humans and agents, preserving headings, lists, and tables while remaining readable to models. Since most unstructured documents are not natively in Markdown, the main challenge is the translation layer, which the company addresses with models that convert document containers into Markdown. The quoted post adds that Markdown keeps table columns intact, with HTML used for tables with merged headers.
AISemiAnalysis argues China's AI safety regime is speed-first, with rules covering content and public-facing services but no frontier-risk duties tied to training compute or capability. Its dataset of 857 releases from nine leading Chinese developers found only 31 (3.6%) ever had a published safety result, and just 9 at launch. The analysis also finds that technical experts favor binding frontier rules while the top leadership's development-first preference settled the policy debate.
AIEthan Mollick notes that some OpenAI proofs have sparked rapid iterative advances from a wide community of collaborators amid debate over their implications for mathematics. A related post reports that a collaborative effort tightened the bound κ from 2⁻¹⁸² to 2⁻¹⁵, a roughly 500-thousand-fold improvement on the previous result.
AIMiles Brundage speculates that Anthropic's new rule, making abusive behavior toward Claude a Usage Policy violation effective November 12, 2026, is meant to help its IPO and win favor with the administration as part of a regulatory capture strategy. The post offers this as a guess about motive rather than a confirmed fact, and it relies on the policy change flagged in the quoted post by Andrew Curran.
AILewis Tunstall quotes Kyle Cranmer's post about a paper by Nate Gunnarsson on a non-perturbative approach to chiral fermions in the Standard Model, extending Lüscher's abelian result. The paper's acknowledgments state that OpenAI's GPT-6 Astra model was essential, proposing refinement strategies, writing rewrites of the proof, and carrying out Lean verification.
AIEpoch AI reports that models fail to adopt its house style even when given many examples. As one example, a diagram generated by Fable 5.1 is far more information-dense than what Epoch would produce.
AIEpoch AI introduced Epoch Automation Reports, which evaluate frontier models on realistic, open-ended tasks drawn from its own work. Claude Fable 5.1 and GPT-6 Astra lead the results, but neither comes close to fully automating Epoch's work.
AIEthan Mollick predicts that a short document at will become assigned reading in college classes covering the current moment. He says the few paragraphs contain a great deal of significant content, though the post itself gives no further detail.
AIThe Financial Times reports that OpenAI's annualized revenues are $20bn less than previously signaled. SemiAnalysis's post asks when OpenAI will launch an agent swarm for a lean proof of its actual annualized recurring revenue, without giving a timeline.
AIThariq, an Anthropic-affiliated account, posted a brief call that the team should be more ambitious. The post names no product, model, or concrete plan, so no further details can be supported.
AIBoris Power praised what he called impressively fast progress on the community effort to optimize integer multiplication below n log n. The effort is tracked on a live progress page launched by @aurel_pr, which lets everyone follow the results as they come in.
AIEthan Mollick recalls writing a 2005 grad school paper on the original computer hacking, phreaking, and BBS scene. He notes that hackers were often driven by curiosity, but the tools they built were widely exploited by "script kiddies" who caused most of the damage and chaos. He then pivots to AI hacking, though the post does not elaborate.
AILiquid AI is building personal AI that runs on devices such as phones, wearables, PCs, and cars, using its Liquid Context layer, which is optimized for Snapdragon processors, to sit between models, agents, and hardware. The company's agent harness uses its own models to decide which user context to retain and how to compress it within fixed compute limits. Liquid AI is also collaborating with Mercedes-Benz Group AG to bring on-device AI to its cars and plans observability and continuous improvement loops for self-improving agents.
AITessl's blog post argues that repository automation needs Continuous AI, a third pillar alongside CI and CD for scheduled, auditable AI workflows that improve repositories over time. The article describes GitHub Agentic Workflows, which harden agentic workflow specifications into GitHub Actions that can run coding agents such as Claude Code, Copilot CLI, Gemini CLI, or Codex-style agents. It emphasizes read-only agent steps, restricted outputs, and human review of pull requests.
AIJoshua Achiam says he took a quiz or test a second time after the first attempt returned a Dean Ball easter egg. He calls the result very reasonable and says landing dead center on the map feels on-brand.
AIMeta says its closed-loop liquid cooling recirculates water in a sealed system, so its data centers use less water annually than an average US golf course. The company also says it pays for the new generation and transmission its facilities require, including in Louisiana under its Entergy agreement, and that data centers create construction and operations jobs.
AILlamaIndex says Markdown is its default output for document parsing because it preserves headings, lists, and tables, which helps models read content correctly. The post notes that parsers can extract every word yet lose which column a number belongs to, forcing models to guess. For tables with merged headers, LlamaIndex switches to HTML.
AIStanford HAI associate directors Risa Wechsler and Russ Altman, speaking at a Stanford orientation, argued that AI agents can deepen scientific research but must be paired with interdisciplinary collaboration. They stressed rigorous, reproducible methods and clearly measured uncertainty, since convincing AI answers are not enough. They also said labs must weigh agent costs and preserve mentorship so that automation supports human participation in research.
AIJason Liu says he lacked the vocabulary, not the ambition, to design objects in CAD for 3D printing, so he learned printing, part efficiency, and shape terminology. He argues AI is like magic that still requires learning the right spells and incantations, and that it has made him want to understand the subject more deeply.
AINatura AI's Interface is a ring that lets users press and hold to speak requests to AI agents such as Claude Code, Codex, or Hermes, then release to send them. The post argues that screenless interfaces may define the next phase of agent use, since handing work to agents is currently slowed by pulling out a phone. Early-adopter pricing is $99, with shipping slated for January.
AIJabil senior director Thomas Brown says humanoid robots are entering a phase of tens of thousands of units, where manufacturability, cost structure, and quality become central. He says Jabil works with developers to cut costs for scale, while compute and memory prices remain a pain point, and that humanoids make sense in factories and warehouses while mobile arms still suit high-speed tasks.