2026-11-07 的 AI 资讯

OpenAI valued at $1.25 trillion, as rival Anthropic nears its IPO

OpenAI is now valued at $1.25 trillion in a new round of funding that has raised another $11 billion, according to Bloomberg. That would make it one of the world's most valuable companies, not long after raising $122 billion from investors in March at a valuation of $852 billion. Echoing FT and Wall Street Journal […]

China reportedly recruited AI exec to spy on Jensen Huang

China's intelligence agency recruited Megaspeed executive Alice Huang to spy on Nvidia CEO Jensen Huang, according to The New York Times. The Ministry of State Security (MSS) enlisted her to glean information about American AI chips and "potentially influence the company to push for its most advanced chips to be made available to China." Megaspeed […]

OpenAI reportedly raises an additional $11 billion at a $1.25 trillion valuation

OpenAI has reportedly raised an additional $11 billion as part of a funding round that hasn't closed yet but is expected to in the coming weeks.

ChatGPT silently passes data between connectors and the bounty wants an ID card

Hacker News 热榜讨论帖（69 分，27 条评论）

AI is dramatically accelerating AI research

Hacker News 热榜讨论帖（195 分，150 条评论）

All the latest updates on AI data centers

Massive new data centers are the physical foundation for tech companies’ hopes and dreams for AI. But the rush to expand warehouses full of energy-hungry servers has also kicked up fights across the world over their impact on power grids, utility bills, nearby communities, and the environment. Data centers were a major bipartisan issue ahead […]

Show HN: Agents infer Lego mechanisms from sources, then animate them

Hi there :) Around a month ago, I released LDraw Nova, an agentic toolset for assembling LEGO CAD models by generating LDraw source code files. For more details, here's the Show HN post: https://news.ycombinator.com/item?id=49937916 From the comments, I got the impression that there was quite an interest in seeing not only AI-generated LEGO models, but also AI-animated LEGO ones. I took a look at

Hundreds of Artex AI hacking tools found publicly exposed to the web

Security researchers found hundreds of servers hosting the penetration testing tool Artex exposed to the internet, allowing anyone access to the hacking tool.

Meta introduces Spaces, a new way to collaborate with your Muse agent

Meta's Muse agents continue to get new features and integrations. Spaces promises to be a new way to interact with the AI helpers.

Meta’s Muse AI can create custom ‘spaces’ for the version of the app you need right now

Meta will now allow you to build customized, app-like experiences within Muse. The new feature, called "spaces," allows its AI assistant to build things like a shopping hub, school dashboard, music tracker, trivia game, and more, which you can interact with directly in the Muse app. Meta says you can create a new space by […]

Memorilla: Latent Semantic Memory for LLMs

Hacker News 热榜讨论帖（86 分，7 条评论）

How a Blue Wave midterm might reshape the AI industry

Democrats’ historic victories in this week’s U.S. midterm elections could also transform the tech industry. The public voted on 41 data center-related ballot measures on Tuesday — in all but one, the majority voted to impose more restrictions on data center development. And even when data centers weren’t on the ballot directly, they were a […]

After the midterms, is the AI industry in for a rough two years?

Democrats’ historic victories in this week’s U.S. midterm elections could also transform the tech industry. The public voted on 41 data center-related ballot measures on Tuesday — in all but one, the majority voted to impose more restrictions on data center development. And even when data centers weren’t on the ballot directly, they were a […]

AI is the new dating dealbreaker

When Irene moved to San Francisco last year, she started using dating apps as a way to meet new people, get to know her new city, and connect with others who care about the things she does. After the 25-year-old analyst downloaded Hinge and started swiping, she noticed a trend. "Hinge has a prompt where […]

Anthropic IPO buyers see valuation as low as $1.1T

https://epyhia.com/anthropic-ipo-valuation-talks-drop-to-1-3...

Big Tech lost big in the 2026 election

On Tuesday, millions of American voters went to the polls and voted overwhelmingly for Democrats and against AI. The fight against Big Tech is about to come roaring back, and the fight over data centers is looking even harder for the folks trying to build them. On this episode of The Vergecast, Nilay and David […]

How Akai and Moog are embracing vibe-coded mods

inMusic, the parent company of Akai, Moog, Native Instruments, and more, has decided to embrace the modding community. In an interview with The Verge, inMusic CEO Jack O'Donnell said, "We're going to provide a toolbox to the community, we're going to provide the inner workings of, say, an MPC or Sample so that people can […]

Rune: A native GPU-rendered Rust UI framework for macOS, Windows, Linux and web

Hacker News 热榜讨论帖（153 分，109 条评论）

Reading the Anthropic S-1, with Zvi Mowshowitz

Hacker News 热榜讨论帖（25 分，0 条评论）

Ramp data shows AI spending stalling just as Anthropic nears its IPO

AI usage keeps climbing, but spending has stalled. Price cuts and smarter cost controls among the biggest spenders could challenge investor assumptions about AI hypergrowth.

After blocking Muse, Amazon now lets users text Alexa to help them shop

Amazon said that the feature will roll out to millions of users in coming weeks.

Amazon will let you text Alexa your shopping list

Amazon is adding a new feature to Alexa for Shopping that allows customers to purchase items by texting the AI shopping assistant. Users can describe the products they're looking to buy and/or provide a photograph of the item, and Alexa for Shopping will search for matching results in Amazon listings, text back with recommendations, and […]

Gemini Omni 1.2 Flash

Hacker News 热榜讨论帖（294 分，220 条评论）

Underdog shows off a 32MB speech model in new demo

Despite its small size, Underdog's new model can generate both speech and music.

Physicl raises $10M to make the digital world real enough for robots

Physicl is trying to solve the sim-to-real gap with a pipeline originally built to market goods online.

Protesters break into Microsoft data center site in Amsterdam

Protesters have broken into a construction site of a new Microsoft data center in Amsterdam this morning. Members of the Geef Tegengas group helped organize the protest against the construction of three 85-meter data center towers in the outskirts of Amsterdam. Reuters reports that around 100 people made it into the construction site, with some […]

AI for scientific discovery: How researchers use Claude

Hacker News 热榜讨论帖（87 分，37 条评论）

Partnering with Hacktron to find vulnerabilities in OpenAI’s systems

OpenAI is partnering with Hacktron to find vulnerabilities, test connections with third-party services, and strengthen defenses with frontier AI.

Palo Alto Networks in talks to buy AI agent security startup Clutch for $500M

Clutch, backed by Lightspeed and SignalFire, was last valued at $85 million, according to a source.

I struggle to believe that AI R&D is ingenious

Hacker News 热榜讨论帖（178 分，184 条评论）

OpenAI is stepping back from NDAs for data center projects

Following this week's midterms where voters resoundingly voiced their opposition to data center expansion, OpenAI today announced new commitments to communities where it is working on data center projects, including that it won't use non-disclosure agreements (NDAs) with public officials. Many data center projects have come under scrutiny for relying on NDAs to hide discussions […]

Measuring how AI agents handle everyday wet-lab science

OpenAI introduces BioBasicsBench, a 100-task benchmark measuring how well AI agents interpret raw experimental data and handle everyday wet-lab science.

Airbnb says it’s launching an AI travel agent in the first half of 2027

Airbnb says its upcoming AI agent will help travelers plan trips, compare options, manage bookings, and get assistance throughout their stay.

Claude Code sends your email address to websites

Hacker News 热榜讨论帖（1553 分，492 条评论）

Cult.fit builds operational muscle with Codex

Cult.fit is using Codex to help teams across functions build tools, move faster, and change how work gets done.

Meta says US users can now video chat with their Muse AI avatar

Meta's AI agent can now take calls as a customizable animated avatar, ahead of the launch of a Tamagotchi-like wearable designed to bring Muse to life.

Meta warns that election backlash against data centers could make AI expansion harder

Meta warns that growing political opposition to data centers could threaten future AI investments, as communities push back against massive infrastructure projects.

OpenAI released a new cybersecurity model. Is that enough?

OpenAI's newest cybersecurity model comes out as AI labs face scrutiny of their work to protect digital infrastructure.

User Embeddings are Superpositions of Interpretable Behavioral Modes

User embeddings are central components of modern recommendation, personalization, and behavioral modeling systems. Though these embeddings can compactly encode rich behavioral information, they are inherently dense and opaque, limiting interpretability. To address this gap, we argue that user embeddings can be understood as sparse superpositions of interpretable behavioral modes. This structure is

SwarmEvolve: Latent Agent Swarms for Evolutionary Discovery and Optimization

LLM-guided evolutionary search uses AI agents to improve programs for mathematical and computational problems by iteratively refining solutions and combining high-scoring methods. However, the process is often slow because the LLM generates a long textual chain of thought for each round and must repeat this generation many times to maintain population diversity. We present SwarmEvolve, a framework

Scientist World: Grow a Research World for AI-Scientists

AI-Scientists arrived, but a world for them to grow in did not. Human scientists take theirs for granted $-$ centuries of open problems and shared ideas, waiting in the scientific literature. To AI-Scientists, that world is readable but not runnable, so we build Scientist World for them: every research paper becomes an executable environment. Our autonomous pipeline extracts each paper's open-ende

XStack: A Benchmark for ML Repo Migration

ML code should run where researchers choose, not where it was first written. Today, machine learning code mainly lives in two ecosystems, PyTorch on NVIDIA GPUs and JAX on Google TPUs. Migrating repositories between them means building and testing across frameworks and hardware. This migration is (1) a practical need, since it unlocks different hardware for researchers, and (2) a natural benchmark

Same Behavior, Different Mechanism: What Finite Experiments Can Certify About Integrated Information

Interpretability, AI auditing and assessments of machine consciousness increasingly rest on mechanism-level quantities: functionals of a system's causal transition law that describe how it computes, not what it outputs. Experiments see that law only through finite, partial and noisy interventions. We ask when such a quantity is identifiable from an experiment's access, continuous in the mechanism

Towards interpreting any residual stream direction by scaling activation inversion

To interpret a direction in a language model's residual stream, we can look for text that strongly activates it. But when such text is found by scanning a corpus, it is limited to what that corpus contains. We develop Maximally-Activating Example Metamodels (MAEMs), language models which map residual stream directions from a subject model to text exemplars that strongly activate those directions.

Adversarial Training with Large Step Sizes: Implicit Bias, Optimization, and Sharpness

Adversarial Training (AT) is an important method for improving model robustness. We study the dynamics of deterministic Gradient Descent (GD) in AT. Most existing works in this direction focus on the small or vanishing step sizes regime, where GD resembles a continuous-time flow. However, such step sizes can make training inefficient and are rarely used in practice. In this paper, we study the AT

Unpacking the Compute Efficiency of Strong-to-Weak OPD against GRPO

On-policy distillation (OPD) trains a student on its own rollouts under a frozen teacher's per-token supervision. Does this dense signal reduce the training FLOPs needed to reach a target accuracy compared with reinforcement learning? We systematically compare vanilla reverse-KL OPD with Group Relative Policy Optimization (GRPO) across task difficulty, student and teacher size, domain, and optimiz

ReproBridge: Building Reproducibility into ML Research Artifacts

Reproducing machine learning experiments requires reconstructing procedures from information scattered across papers, released code, and supporting artifacts. For coding agents, reproduction success can therefore depend on their ability to fill in missing details, rather than on the released artifacts alone. We ask whether preserving recovered procedures and decisions improves subsequent independe

A Hidden Markov Preference Model for Online Prediction of Adversary Actions

A defender anticipating an attacker needs a probability distribution over the attacker's next action from information available before that action occurs. Two structures complicate this: the attacker's behavioral regime persists and can switch within an episode, and each action combines an operation with state-dependent arguments whose feasible values change over time. We capture both with a hidde

Label-Wording Augmentation in a Typed-Decision Encoder: Gains on Unfamiliar Words, Not on Abbreviations or Codes

Typed-decision models score options that the caller supplies at request time, so their accuracy depends on label wording that the model's developer does not control. We compare two releases of a 307M-parameter multilingual cross-encoder, Ru-Jev, trained on thirteen Russian-language tasks; the later release adds an augmentation that presents labels in several surface forms. Holding questions fixed

Understanding AI from the Inside: The State of Mechanistic Interpretability for AI Safety

Neural networks are likely to be among the most consequential artefacts in human history. As these systems become more capable and autonomous, their potential to cause societal harm also grows, making AI safety one of the defining challenges of our time. Yet we continue to build increasingly powerful systems without fully understanding how they work, how they will behave in unfamiliar situations,

MedGraph: Maintaining Patient State with Graph Memory for Long-Term Medical Dialogue

LLM-based medical assistants that follow a patient across visits need a memory that keeps the patient's clinical state correct as it changes. We examine the errors of five memory systems on longitudinal medical dialogue and find three recurring causes: the retrieved evidence gives an outdated state, includes another person's health facts, or lacks a safety fact the answer depends on. We present Me

The Power of Meta Access in the Software of the Future --- Demonstrated on LLM Chat UIs and Chat AI

Chat remains the default surface for many large language model (LLM) tasks, even as agentic tools such as OpenCode excel at coding and automation. Yet the model sees the current dialogue but not the workspace around it. We explore meta access---giving the assistant a structured, permissioned view of the interface itself and propose a ladder of levels: feature tutoring; metadata, settings, and pref

Failure Prediction and Classification in Vision-Language-Action Models Via Calibrated Latent Probing

Vision-Language-Action models (VLAs) enable general-purpose embodied intelligence, but their performance is brittle under distribution shift, where unanticipated failures can damage equipment or endanger people. Deploying VLAs in real-world systems therefore requires a runtime monitor that is timely, calibrated, and informative while being efficient enough to run on the edge. Existing methods fail

Data-driven discovery of intrinsic Lie point symmetries of differential equation systems

Discovering continuous symmetries directly from differential-equation data can reveal geometric structure for model reduction and scientific machine learning without symbolic determining equations or a closed-form representation of the governing system. We formulate intrinsic Lie point-symmetry discovery as an operator-theoretic problem on sampled jet-space data. Admissible generators are identifi

On the limitations of molecular representations

Molecular representations such as fingerprints and learned embeddings play a central role in cheminformatics and small-molecule machine learning. They serve as the primary input for a diverse array of tasks, including clustering, dataset splitting, and virtual screening, yet their evaluation is often limited to downstream model performance on narrow benchmark tasks. We carried out a systematic inv

Benchmarks as Software: Terminal-Bench and the Case for Continuous Benchmarks

Coding agents are among the most economically consequential deployments of LLMs to date, making reliable evaluation increasingly important. Yet even benchmarks that are carefully validated at release can become stale. As models improve, they can discover valid solutions that existing verifiers reject, while changes to evaluation infrastructure can invalidate assumptions embedded in tasks and tests

Spectral and Subspace Dynamics of LoRA in Supervised Fine-Tuning and Reinforcement Learning

Low-rank adaptation (LoRA) is commonly used in both supervised fine-tuning (SFT) and reinforcement learning (RL), often under similar optimizer settings. Yet, how the different dynamics of the two stages manifest in the low-rank increment that LoRA optimizes remains unclear. Tracking the singular spectrum and subspaces of LoRA updates on mathematical reasoning tasks, we find that RL concentrates t

The Gold Rush in AI4Math, II: What are signatures of AI-assisted Proofs?

The growing use of AI in mathematics raises a question: can we detect AI involvement in proofs, separately from AI assistance with writing? Historical arXiv papers, recent papers disclosing substantive AI use, and manuscripts released by OpenAI provide a rich empirical basis for studying this question. However, several challenges remain: lack of matched case-control pairs; potential incompleteness

Conformal Prediction for Time-Dependent PDEs

Uncertainty quantification is crucial in scientific machine learning, where models inform safety-critical tasks such as flood forecasting, financial risk management, and thermal control in machines. Conformal prediction provides distribution-free coverage guarantees, but in time-dependent settings common to physics and engineering, these guarantees can break down, leading to systematic undercovera

Social Humanoid: Socially Responsive Human-Humanoid Interaction

Humanoid robots are typically trained as solitary agents, leaving them unable to coordinate physically with human partners who naturally anticipate and adjust to one another. To bridge this gap, we present Social Humanoid, a hierarchical system that generates and executes socially responsive whole-body motion on a 29-DoF Unitree G1 conditioned on a human partner's motion. Its high-level policy per