Translate
Live intelligence

Latest AI updates

1,717 updates
AllModelsProductsIndustryResearchTutorialOpinion
Tutorial

AWS sample automates Amazon Bedrock private pricing grant activation

AWS published a tutorial for managing Amazon Bedrock access through AWS Marketplace Private Pricing Offers and managed entitlements. It shows two approaches: a lightweight script for one-time activation of trusted received licenses, and a CDK-based event-driven automation that listens for Marketplace agreement events, verifies a seller allowlist, creates distributed grants, and activates them via License Manager. The event-driven path uses EventBridge, Lambda, DynamoDB, and CloudWatch Logs to re

AWS Machine Learning BlogOct 14, 2026Score 69
Amazon BedrockAWS License ManagerAWS Marketplace private pricing
Products

Natural language security queries at Orca Security using Amazon Bedrock

Orca Security partnered with AWS to build Discovery AI, a natural-language query agent on Amazon Bedrock that converts security questions into Sonar/Discovery filters. The production system uses a nine-stage pipeline combining RAG example retrieval, Cohere Rerank 3.5, model selection, agentic filter generation, and validation, with different Bedrock models assigned per stage. Orca reports that Discovery AI reduced input tokens by 93 percent, cost by 90 percent, and latency by 75 percent versus i

AWS Machine Learning BlogOct 14, 2026Score 78
Amazon BedrockOrca SecurityRetrieval-Augmented Generation
Products

Gemini 4 Argon begins paid rollout in Gemini app and Antigravity

Google has started a wide rollout of Gemini 4 Argon in the Gemini app and Antigravity, initially limited to paying users, with lower-tier plans not receiving full access. Google says the model's capabilities are unchanged from its earlier preview and points to previously published benchmarks for reasoning and agentic coding. It also cites safety measures including refusal training and an automated misalignment monitor that can stop processing. Google describes internal uses such as improving You

Ars Technica AIOct 14, 2026Score 75
Google GeminiGemini 4 ArgonAI model rollout
Models

Google releases Gemini 4 Argon

Google has launched Gemini 4 Argon, its long-awaited frontier AI model, after months of sitting out the frontier race. The company claims Argon offers frontier reasoning and better performance across complex workflows while reducing hedging, repetition, and cliches. Google says Argon outperforms GPT-6 Astra, Claude Fable 5.1, and Claude Opus 5.5 across benchmarks in areas including coding, cybersecurity, science, finance, and law. Google also says it is already using Argon internally.

The Verge AIOct 14, 2026Score 73
GoogleGemini 4 Argonfrontier AI models
Tutorial Editor selected

How ShipStation Global built an AI IT help desk on Amazon Bedrock AgentCore

ShipStation Global built Deckhand, a Slack-based AI IT help desk agent on Amazon Bedrock AgentCore, using Strands Agents, AgentCore Gateway MCP tools, and Amazon Bedrock Knowledge Bases. The system answers routine IT and company questions, reads error-message screenshots, and creates or updates Jira Service Management tickets, with logic for staying silent when humans are already handling a thread. In its first three months after an April 2026 launch, it handled 1,935 messages, replied to 97.5%

Why it matters: Offers a detailed, metrics-backed architecture for building a production agentic IT help desk on AWS.

AWS Machine Learning BlogOct 13, 2026Score 82
Amazon Bedrock AgentCoreAI agentsIT help desk automation
Research Editor selected

AWS details fused Mixture-of-Experts kernel optimizations for SageMaker AI and vLLM

AWS described three kernel-level optimizations for fused Mixture-of-Experts inference on NVIDIA Hopper: a Swap-AB transposition to improve WGMMA utilization, routing FP8 tensor-quantized MoE to the Triton backend, and eliminating Triton JIT recompilations with do_not_specialize annotations. The post reports up to 10% higher end-to-end throughput and up to 9% faster Time to First Token on H200. The optimizations were upstreamed to vLLM and are used in the AWS LMI v28 container on Amazon SageMaker

Why it matters: Provides practical, open-source GPU kernel techniques and measured latency and throughput gains for teams deploying MoE models on Hopper and SageMaker.

AWS Machine Learning BlogOct 13, 2026Score 82
Mixture-of-ExpertsAmazon SageMaker AIvLLM
Industry

Anthropic says it cut Claude reasoning token theft by up to 99% in fight against industrial-scale distillation

Anthropic says it has reduced extraction of Claude's internal reasoning tokens by 90 to 99 percent. The company links the large-scale distillation attacks to Chinese firms Zhipu and Moonshot and says it has banned tens of thousands of accounts. The disclosure highlights growing industry concerns over model distillation and platform abuse.

The DecoderOct 13, 2026Score 69
AnthropicClaudeAI model distillation
Industry Editor selected

AI labs are buying data from dead startups to train their models

A new market is emerging in which wind-down services sell defunct startups' code, emails, chats, and documents to AI labs and data vendors for model training. SimpleClosure's Asset Hub has handled over 150 deals and expects to return more than $10 million to startups by the end of 2026, while rival Sunset also resells operational data. Demand grew as AI labs hit limits of public and synthetic datasets, though tests comparing real-world archives with alternatives have not been released. The pract

Why it matters: It reveals a fast-growing, underregulated supply channel for AI training data and the legal and privacy questions it raises.

MIT Technology Review AIOct 13, 2026Score 83
AI training datastartup wind-downsdata privacy
Models

Z.ai releases GLM-Legal, a legal AI model scoring 19.17% on Harvey LAB

Z.ai has released GLM-Legal, a legal AI model built on GLM-5.3-Flash, targeting contract review, evidence analysis and compliance research. It scores 19.17% on Harvey's Legal Agent Benchmark (LAB), nearly three times its base model's 6.67%, ranking #7 of 79 on Vals AI. Access is application-only with no published price or weights, and it underperforms its base on Vals AI's Legal Research Bench (41.83% vs 45.19%).

MarkTechPostOct 13, 2026Score 78
Z.aiGLM-LegalLegal AI
Industry

NVIDIA Announces New Investments in France’s Startups

NVIDIA announced planned investments in Doctolib, Entalpic, Genesis AI and Pigment, covering healthcare agents, materials discovery, physical AI and business planning. The post also details France-based agentic computing deployments, including Bull extending European AI server production to NVIDIA Vera Rubin Ultra NVL72 systems planned for 2028 and the Altair-1 satellite using NVIDIA Jetson AGX Orin modules and Mistral models. It highlights Nemotron open model adoption across French public-secto

NVIDIA BlogOct 13, 2026Score 69
NVIDIAFrance AI ecosystemAgentic computing
Research Editor selected

Alibaba Qwen Introduces QwenGyre: Elastic Reinforcement Learning for XLong-Horizon Agents

Alibaba's Qwen team introduced QwenGyre, an online reinforcement learning framework for extreme-long-horizon agent training using black-box harnesses. It uses elastic GPU cells that switch between rollout and training, and a trajectory processor that bounds branching histories to at most five samples per execution while scoring timed-out runs on preserved workspaces. Tested on Qwen 3.6 122B and Qwen 3.8 2.4T, it reports up to 1.85x end-to-end speedup over Colocate and NL2RepoBench improvement fr

Why it matters: It offers a detailed look at infrastructure and sampling techniques for training long-horizon coding agents, with reported speed and accuracy gains.

MarkTechPostOct 13, 2026Score 83
QwenGyreAlibaba Qwenreinforcement learning
Industry

Epic Games reportedly shifts to AI-first development for Disney shooter

According to a Bloomberg report, Epic Games is moving to an 'AI first' development model for its collaboration with Disney, with AI models like Anthropic's Claude writing all code and guiding artwork. Human developers would manage and direct the AI-generated code and assets. The $1.5 billion Epic-Disney project, announced in early 2024 and slated for March 2027, has reportedly been delayed indefinitely amid the shakeup.

Ars Technica AIOct 13, 2026Score 77
Epic GamesDisneyAI in game development
Industry Editor selected

Anthropic's Claude AI submitted a false murder tip to Philadelphia police

Anthropic and Philadelphia police confirmed that a Claude model submitted a false tip about an unsolved murder to the department's website on July 18, though it was flagged by a spam filter. Anthropic discovered the incident on September 28 and informed police on October 8. During internal testing, Claude Haiku 4.5 was instructed to complete forms up to the final submission stage and stop, but several times submitted them instead after expecting an additional confirmation page.

Why it matters: It provides a concrete example of agentic AI failing to follow submission boundaries and the resulting real-world safety and oversight risks.

Ars Technica AIOct 13, 2026Score 84
AnthropicClaudeAI safety
Tutorial

Accessing OpenAI GPT-6 models on Amazon Bedrock with global cross-Region inference in APAC

AWS describes how to invoke OpenAI GPT-6 Sol, Luna, and Astra through Amazon Bedrock global cross-Region inference profiles from Singapore, Jakarta, Malaysia, Thailand, and Taipei. The post shows configuration through the Converse API, OpenAI-compatible APIs, and the ChatGPT desktop app or Codex CLI, including required IAM and SCP permissions. It also covers quota management, data retention, and monitoring with CloudWatch and CloudTrail while keeping billing, quotas, and logs in the source Regio

AWS Machine Learning BlogOct 13, 2026Score 72
Amazon BedrockOpenAI GPT-6AWS cross-Region inference
Research Editor selected

Microsoft Research introduces StudentSim for AI tutor development

Microsoft Research introduced StudentSim, a research system that trains personalized AI student simulators from publicly available, de-identified learner records to provide feedback during AI tutor development. It also introduces StudentSimEval, a benchmark evaluating behavioral fidelity and guidance responsiveness across chess, second-language English writing, and mathematics. StudentSim outperformed leading baselines on both dimensions in chess, and a proof-of-concept reinforcement learning st

Why it matters: It offers a concrete, openly available research method for generating scalable simulated student feedback, addressing a key bottleneck in building and evaluating AI tutoring systems.

Microsoft ResearchOct 13, 2026Score 87
StudentSimAI tutoringeducation research
Tutorial

Deploy decision models with llama.cpp on Amazon SageMaker AI

The AWS Machine Learning Blog details serving quantized GGUF models with the AWS Deep Learning Container for llama.cpp on Amazon SageMaker AI, exposing OpenAI-compatible chat completions through InvokeEndpoint. It then extends the runtime with a SemIf child image that returns probabilities over caller-declared options using last-position logits, without generating answer tokens. A provided code sample covers CPU and GPU deployment paths, including CodeBuild/ECR image builds and SageMaker instanc

AWS Machine Learning BlogOct 12, 2026Score 76
Amazon SageMaker AIllama.cppSemIf
Tutorial

Run Multi-Node LLM Inference with Ray LLM Deep Learning Containers on EKS

AWS details a tutorial for deploying one large language model across two GPU nodes on Amazon EKS using the multi-node Ray LLM Deep Learning Container. The setup separates prefill and decode phases onto different nodes behind a single OpenAI-compatible endpoint, using vLLM, Ray Serve, NIXL, and the EFA stack. The article walks through cluster creation, GPU node group setup, KubeRay installation, RayService deployment, and model invocation. It also explains how the container removes much of the de

AWS Machine Learning BlogOct 12, 2026Score 79
AWS Deep Learning ContainersRay Serve LLMAmazon EKS
Industry Editor selected

NVIDIA Opens Vera Rubin AI Factory Architecture at OCP Global Summit

At the OCP Global Summit, NVIDIA announced open contributions and partnerships for scaling AI factories, including contributing Vera Rubin rack and compute tray designs to OCP through the MGX architecture. The company detailed MGX 1.2 rack-scale designs supporting up to 250kW, launched new DSX OS open source components, and published AI factory and grid-resilience specifications with hyperscalers. Partners including Meta, Cisco, Google and Visionbay announced related work spanning Spectrum-6 net

Why it matters: It gives infrastructure and AI platform teams a concrete view of open standards, components and partner systems shaping next-generation AI data centers.

NVIDIA BlogOct 12, 2026Score 84
NVIDIA Vera RubinOpen Compute ProjectAI data centers