Skip to content

AI NEWSSee every step of AI

THE JOURNAL

AI NEWS JOURNAL

AI, IN PERSPECTIVE

Search AI news

Search headlines, summaries and articles to connect the dots.

2026-10-21

NVIDIA Blog18:10

NVIDIA releases an extended version of the Agent Toolkit, empowering enterprises to build trustworthy agents based on private data.

NVIDIA’s open-source Agent Toolkit has added core components such as the Agent Knowledge Foundation, integrating the CUDA-X libraries, the Nemotron Retriever and Kumo series of open models, the OpenShell secure runtime, and the NemoClaw blueprint. This enables enterprises to build specialized agents that are auditable, controllable, and capable of deeply understanding business semantics—without migrating data or modifying existing workflows.

Timeline and related coverage

NVIDIA Blog18:00

NVIDIA releases a portfolio of open-source models across multiple domains, driving the development of sovereign AI in European industries.

NVIDIA announced the expansion of its open-source model family to encompass areas such as language, robotics, bioinformatics, and quantum computing, unveiling the Nemotron 3.5 series, the Kumo structured-data model, the Cosmos world model, the Alpamayo 2 Edge autonomous-driving model, the BioNeMo RFD4-Proteina protein-design model, and the Ising quantum-computing model. All these models are now available to developers and enterprises, with access to weights and tools, supporting localized customization and independent deployment.

Timeline and related coverage

NVIDIA Blog17:30

GTC Berlin 2026: Jensen Huang’s keynote comprehensively unveils the AI factory, security agents, and Europe’s AI ecosystem strategy.

At the GTC Berlin 2026 conference held at Berlin’s Tempodrom, NVIDIA CEO Jensen Huang unveiled several key advancements: the DSX POD modular AI factory, the Vera Rubin full-stack AI agent platform, the OpenShell security architecture, the Agent Knowledge Foundation knowledge layer, and the Agile ONE SN humanoid robot reference platform. He also announced a European AI investment commitment exceeding €10 billion, spanning more than 12,000 partners, while emphasizing that Europe is accelerating the development of an independent, controllable AI economic infrastructure.

Timeline and related coverage

Latent Space13:01

Tenex Builds an AI-Native SDLC: Agent-Driven Development Practices for Fortune 2000 Companies

Tenex, an AI-native software company founded 18 months ago, leverages output‑driven compensation, the MetaHarness framework, and Opus 5.5–powered AI agent workflows to serve roughly half of the Fortune 2000 companies. Its model focuses on process standardization, revaluing non‑coding capabilities, and addressing the last‑mile challenges of enterprise‑level AI deployment.

Timeline and related coverage

Latent Space13:01

Tenex builds an AI-native SDLC: outcome-based pricing, agent-driven development, and the last mile of enterprise AI deployment.

Tenex is an AI-native software delivery company established 18 months ago (as of 2026), with a team of 75 employees, serving roughly half of the Fortune 2000 companies and a comparable number of mid-sized enterprises. Its core practices include: a dynamic engineer compensation system anchored in story points; a quality‑assurance mechanism led by AI strategists; the MetaHarness framework, which embeds context directly within Git; and a capability realignment focused on “high‑value, non‑coding tasks.” Meanwhile, it has been observed that enterprise AI adoption remains concentrated on general‑purpose LLM deployments and engineering‑side agents, with backend office‑type production‑grade AI agents achieving only about 5% penetration.

Timeline and related coverage

Hugging Face 官方博客2026-10-21

NVIDIA launches SimReady Central: an open-source, simulation-ready 3D asset library for robotics and AI factories.

NVIDIA, in collaboration with Hugging Face, has launched SimReady Central, a free and open-source simulation-ready (SimReady) 3D asset library built on OpenUSD. The initial release includes over 2,000 validated industrial-grade assets, covering robot bodies, end-effectors, AI factory infrastructure, and more, enabling developers to rapidly build physics-based AI training environments and digital twin scenarios.

Timeline and related coverage

Hugging Face 官方博客2026-10-21

VocabX: Reduces the inference cost of domain-specific small language models by 77% through cost-weighted vocabulary specialization.

Dharma-AI has introduced the VocabX method, which reconfigures the vocabulary of language models for specific tasks—such as automated scoring of Portuguese essays—by optimizing for cost rather than word frequency. Without modifying the model architecture, parameters, or training pipeline, empirical results show a 77% reduction in inference costs, while the evaluation score (bench score) improved from 0.703 to 0.810.

Timeline and related coverage

Hugging Face 官方博客2026-10-21

TorchNeuron System-Level Performance Analysis (Part 1): End-to-End Profiling Based on the AWS Trainium Chip

This article is the first in a series of technical blog posts jointly published by Hugging Face and Amazon, systematically introducing how to use TorchNeuron on AWS Trainium AI accelerator chips for PyTorch model inference performance analysis. It covers the principles of Neuron backend integration, methods for configuring torch.profiler, the use of NeuronConfig and NeuronProfiler tools, and visualized tracing with Neuron Explorer. Additionally, using RMSNorm as an example, it compares the device‑time differences among three implementations: plain PyTorch, the built-in torch.rms_norm fusion, and torch.compile’s automatic fusion.

Timeline and related coverage

Hugging Face 官方博客2026-10-21

NVIDIA releases Kumo Forecast: a multivariate forecasting model that integrates pre-trained time-series representations, historical pattern retrieval, and interpretability.

NVIDIA has launched the open-source time-series forecasting model Kumo Forecast (formerly NV-Tesseract Forecasting), which is built on the MOMENT-1-large Transformer architecture. It supports both direct prediction and dual-path inference via Domain-Aware Representation and Retrieval (DARR), and includes a built-in interpretability framework that requires no modifications to the model structure.

Timeline and related coverage

Hugging Face 官方博客2026-10-21

Microduck open-source bipedal robot supports both simulation training and real-world deployment: NVIDIA Isaac Lab 3.0 enables a Sim-to-Real closed loop.

Pollen Robotics and Hugging Face have jointly unveiled Microduck, a low-cost, open-source bipedal robot, integrating NVIDIA Isaac Lab 3.0 and the Newton physics engine. It supports training walking policies in GPU-accelerated parallel simulations, with domain randomization and high-fidelity actuator modeling—including backlash, latency, and friction—enhancing robustness during simulation-to-real transfer. The resulting ONNX model can be directly deployed on the physical robot.

Timeline and related coverage

Hugging Face 官方博客2026-10-21

NVIDIA releases Kumo Relational: the first foundation model for structured data that outperforms supervised learning without any training.

NVIDIA has launched Kumo Relational, a pre-trained relational foundation model that can perform predictive tasks—such as churn prediction, fraud detection, and demand forecasting—directly on enterprise‑grade relational databases, without the need for task fine-tuning, feature engineering, or table flattening. In 41 benchmark tests spanning 15 databases, it outperformed supervised models like XGBoost across the board, and on the SAP SALT enterprise benchmark, its MRR metric surpassed AutoGluon by 8%. The model supports databases with up to 500 billion rows and enables zero‑code predictive queries via Predictive Query Language (PQL).

Timeline and related coverage

Hugging Face 官方博客2026-10-21

IBM releases the Granite Speech 5.0 series of speech models: supporting multilingual transcription, translation, and speaker attribution.

IBM has released three open-source speech models in the Granite Speech 5.0 series—granite-speech-5.0-2b (high-precision multilingual ASR/MT), granite-speech-5.0-470m-turboctc (ultra-fast edge-side multilingual transcription), and the updated granite-speech-4.1-2b-plus (enhanced multi-speaker conversational modeling)—all licensed under Apache 2.0 and compatible with mainstream inference frameworks.

Timeline and related coverage

AWS Machine Learning Blog05:14

i14 releases the Laplace 3B linear attention model: inference with fixed-state memory, implemented on AWS Trainium2 and native PyTorch.

i14 has launched the Laplace 3B model with 3.028 billion parameters, which replaces Softmax attention with Gated Linear Attention (GLA). During inference, its per-sequence recursive state remains constant at 29.5 MB, independent of context length. The model was trained and validated on a single AWS trn2.48xlarge instance (equipped with 16 Trainium 2 chips), achieving a training throughput of 120,000 tokens per second and an inference throughput of 11,544 output tokens per second (with 1,280 concurrent sequences). All performance metrics are based on internal measurements conducted by i14 and the AWS GenAIIC team.

Official Tutorials

Timeline and related coverage

Ars Technica · AI04:07

Meta removes the term “training” from its commercial service terms to quell AI controversy, but refuses to comply with the robots.txt standard.

In its updated Commercial Terms, effective October 30, 2026, Meta removed the term “train” in response to intense criticism of its AI data‑scraping policies; however, the company explicitly stated that it will not include language in the terms to respect the robots.txt protocol.

Timeline and related coverage

AWS Machine Learning Blog01:52

AWS releases a technical preview: Amazon Quick and Bedrock AgentCore enable the construction of governable, conversational supply chain intelligence systems.

AWS has launched a modular agent architecture based on the Model Context Protocol (MCP), leveraging AgentCore to standardize enterprise‑level tool interfaces. Amazon Quick provides a natural‑language interaction interface for business users, enabling real‑time, approval‑driven supply chain decision support across ERP, procurement, monitoring, and other systems. This solution is currently in technical preview and is not guaranteed to be production‑ready or available for customer deployment.

Official Tutorials

Timeline and related coverage

AWS Machine Learning Blog01:48

GPT-OSS-20B Multilingual Reasoning Fine-Tuning Experiment: SFT Enables Zero-Base Behavior Acquisition, while GRPO Only Effectively Improves Performance When a Probabilistic Headspace Exists

The AWS team conducted multilingual inference fine-tuning experiments on Amazon SageMaker HyperPod using GPT‑OSS‑20B, employing LoRA for parameter-efficient fine-tuning. SFT boosted the correctness rate of inference in five languages (English, French, German, Spanish, and Italian) from 22% to 100%, though the correctness rate for the answer language remained at only 62%. GRPO further improved the answer-language correctness rate to 70% without compromising inference-language performance; however, the gains varied across languages—for example, Spanish saw a +30 percentage-point increase, while Italian experienced a -20 percentage-point decline. The experiments clearly demonstrated that GRPO cannot bootstrap new capabilities from scratch and can only optimize behaviors that already occur sporadically.

Official Tutorials

Timeline and related coverage

AWS Machine Learning Blog01:34

Octave builds a secure, multi-tenant AI infrastructure based on Amazon Bedrock, supporting semantic search across 2 billion regulatory documents.

Octave Reliance has built a rigorously tenant-isolated AI infrastructure using Amazon Bedrock, integrating Cohere Embed v4 with the AgentCore micro‑VM runtime to deliver a Complaint Triage Assistant across the medical device, pharmaceutical, and food manufacturing industries. This solution supports semantic search across more than 2 billion regulated documents—such as complaints, CAPAs, and audit findings—while completely preventing cross-tenant data leakage throughout the process.

Official Tutorials

Timeline and related coverage

AWS Machine Learning Blog00:42

AWS Bedrock has launched cross-region inference in the Canada region, supporting calls to the nominal GPT-5.6/GPT-6/GPT-6.1 models.

Amazon Web Services announced that Canadian customers can, via Amazon Bedrock, invoke models branded as OpenAI GPT-5.6, GPT-6, and GPT-6.1 in the ca-central-1 (Canada Central) and ca-west-1 (Canada West/Calgary) regions using cross-region inference (CRIS); the actual inference takes place in North American or global AWS commercial regions, while data remains at rest in the Canadian source region.

Official Tutorials

Timeline and related coverage

AWS Machine Learning Blog00:22

AWS Agent Registry officially released: supports runtime semantic discovery and dynamic orchestration of MCP servers and A2A agents.

On October 26, 2026, AWS announced the general availability of its Agent Registry service, a managed offering that leverages hybrid semantic-plus-keyword search to dynamically discover, instantiate, and execute approved Model Context Protocol (MCP) servers and Agent-to-Agent (A2A) agents at runtime, replacing hard-coded tool integrations and addressing bottlenecks in the large-scale orchestration of enterprise-grade AI agents.

Official Tutorials

Timeline and related coverage

2026-10-20

MIT Technology Review · AI23:20

The New Economics of Enterprise AI: From “Can It Be Used?” to “How to Achieve Efficient Output”

The article uses the widespread adoption of electricity as an analogy, noting that enterprise AI is transitioning from a phase of isolated experiments to one of infrastructure‑level deployment. The central question has shifted from “Can we deploy AI?” to “How can we deliver measurable business outcomes at the lowest effective cost?” It introduces “the effective cost per useful output” as a key metric and analyzes the economic differences among three AI operating models—building an in‑house stack, leveraging on‑demand cloud services, and integrating a private AI platform—while emphasizing the decisive impact of operational practices such as utilization rates, model selection, and cache management on ROI.

Timeline and related coverage