Microsoft's Decision-1 model enters the fast-growing AI decision model race
Microsoft released Decision-1, a decision model built for fast, structured decisions such as classifications.
#Deployment/Engineering
The DecoderAI Score5858 Latent SpaceAI Score3232 Building AI for Reliable Execution: Lessons From Industrial Robotics
Standard Bots, an AI-native industrial robot maker with $200M in Series C funding at a $1B valuation.
Ars Technica · AIAI Score6262 AI coding agents generate more code, but not more software
A study of coding practices across hundreds of firms finds that human code review acts as a significant bottleneck for AI coding tools.
The DecoderAI Score7373 Anthropic's Claude can now orchestrate up to 1,000 AI agents in parallel through dynamic workflows
Anthropic added dynamic workflows to Claude Managed Agents, letting a lead agent plan.
OpenAI NewsAI Score4040 Asana cuts model costs 76x in browser tests with GPT-6.1 Sol
Asana used GPT-6 Astra in Codex to make its browser agent 76x cheaper and 5x faster in tests. The goal is to offer customers more capable models.
TechCrunch · AIAI Score2222 Danu Robotics' fight to build a better recycling robot
Edinburgh-based Danu Robotics has brought its H.E.R.O. recycling sorting robot to market.
Hugging Face BlogAI Score3232 Ai2 replaces priority scheduler with GPU time budgets and fair-share allocation
Ai2's AI Infrastructure team replaced its priority-based GPU scheduler with GPU time budgets.
AWS Machine Learning BlogAI Score3434 How Postman runs Agent Mode for 40 million developers on Amazon Bedrock
Postman detailed how its Agent Mode serves 40 million developers on Amazon Bedrock.
AWS Machine Learning BlogAI Score2222 AWS September 2026 recap: Bedrock adds GPT-6 Astra, Claude Opus 5.5, Kimi K3; AgentCore and Strands updates
AWS's September 2026 Bedrock updates added OpenAI's GPT-6 Astra, Sol, and Luna models.
OpenAI NewsAI Score2626 Sophos cuts threat investigation time by 96% with OpenAI Daybreak
Sophos uses OpenAI's Daybreak to cut cyber-threat investigation time by 96% and automate 52% of MDR cases while preserving human oversight.
Simon WillisonAI Score2020 ttok 1.0 released, defaults to GPT-5/GPT-6 tokenizer
Simon Willison released ttok 1.0, switching the default tokenizer from GPT-4 to the GPT-5/GPT-6 family. The change follows an experiment by William Liu showing all seven GPT models (5.5.
Simon WillisonAI Score1212 ttok 0.4 released with a --list-models command
Simon Willison released ttok 0.4, updating his tiktoken-based token-counting CLI after a couple of years. The release fixes a Click warning.
OpenAI NewsAI Score2222 LegalOn halves Codex costs while maintaining development speed
LegalOn cut estimated daily Codex costs by 65% while maintaining development speed. It matched Astra, Sol, and Luna to tasks and managed budgets strategically.
NVIDIA BlogAI Score3636 Into the Omniverse: How Developers Turn Ideas Into Simulations With Frontier AI Agents
NVIDIA developers are using frontier AI models such as GPT-6 Astra with Omniverse libraries to turn simulation ideas into working applications. Projects include a humanoid warehouse simulator, an autonomous-driving test workflow on San Francisco's Market Street, and Robo Olympics, where a simulated Unitree G1 humanoid cleared a hurdle in 64 of 100 trials.
AWS Machine Learning BlogAI Score5353 Pay-per-inference for AI agents: How BlockRun and Incarna use Amazon Bedrock AgentCore payments
Amazon Bedrock AgentCore payments lets agents pay for services on demand.
OpenAI NewsAI Score1212 How Oracle turns days of work into minutes with ChatGPT and Codex
Oracle is using ChatGPT Work and Codex to turn specialist knowledge into fast, repeatable workflows across recruiting, engineering, and operations, cutting days of work down to minutes.
TechCrunch · AIAI Score6262 Goodfire launches 'inside-out' monitors that watch an AI model's internals to catch rogue agents
Goodfire launched monitors that read a model's internal activations rather than its output.
AWS Machine Learning BlogAI Score2222 Share GPU clusters across teams with isolation and fairness using Amazon SageMaker HyperPod
AWS details a reference architecture for running multi-tenant GPU clusters on Amazon SageMaker HyperPod with EKS.
TechCrunch · AIAI Score88 Ambrosia Energy and Bloom Energy execs to discuss AI infrastructure opportunities at TechCrunch Disrupt 2026
Ambrosia Energy CEO Ben Longmier and Bloom Energy's Bill Thayer will speak on the Smart Systems Stage at TechCrunch Disrupt 2026 about where the AI infrastructure boom is creating winners. The session examines constraints in power generation, grid connections, data centers, and cooling that are shaping durable business opportunities as AI scaling becomes a physical-world problem.
The Verge · AIAI Score5858 Google launches a universal Gemini agent for work tasks in Gemini Enterprise
Google is launching a universal Gemini AI agent that works across apps and devices in the background.
The DecoderAI Score7272 Zenity Labs: one public AI agent on AWS Bedrock AgentCore could take over every agent in the same account and region
Zenity Labs researchers say a single publicly accessible AI agent on Amazon Bedrock AgentCore was enough to take over every AgentCore agent in the same AWS account and region.
Ars Technica · AIAI Score3838 Nvidia's big bet on physical AI aims for safer robotaxis, humanoid robots
Nvidia is betting on physical AI with its Halos operating system.
MIT Technology Review · AIAI Score1212 Building a safer path to autonomous industrial AI
AVEVA chief technologist Arti Garg outlines a responsible-AI framework for industrial automation.
Simon WillisonAI Score1212 Using Parseable with Datasette for OpenTelemetry traces
Simon Willison documented running Parseable, a new OpenTelemetry-compatible observability tool.
TechCrunch · AIAI Score5858 Hark releases Hark Pro, a privacy-focused AI personal assistant
Hark, a startup founded less than a year ago by serial founder Brett Adcock, released Hark Pro.
AWS Machine Learning BlogAI Score2222 Build a voice travel concierge with Amazon Bedrock AgentCore, Managed Knowledge Base and Nova Sonic
AWS shows how to build an airline voice concierge on Amazon Bedrock AgentCore.
AWS Machine Learning BlogAI Score2222 Manage Amazon SageMaker HyperPod Spaces directly from SageMaker Studio
AWS now lets users create and manage SageMaker Spaces on HyperPod EKS clusters directly from the SageMaker Studio UI.
AWS Machine Learning BlogAI Score1717 Best practices for Amazon SageMaker HyperPod administration and governance
AWS outlines a four-layer governance model for Amazon SageMaker HyperPod — organization.
AWS Machine Learning BlogAI Score2121 Responsible AI governance: How AWS positions customers to align with ISO/IEC 42005:2025
AWS outlines how customers can align AI governance with ISO/IEC 42005:2025, which codifies AI system impact assessment practices. The standard covers the full assessment lifecycle.
Ars Technica · AIAI Score5757 Microsoft debuts Surface Laptop Ultra with Nvidia RTX Spark and Windows 11 changes
At its first live event in two years, Microsoft announced the Surface Laptop Ultra.
Simon WillisonAI Score7171 Anthropic releases Claude Haiku 5.5 at $0.10/$0.50 per million tokens
Anthropic released Claude Haiku 5.5, a fast low-cost model priced at $0.10/$0.50 per million input/output tokens up to 100.
TechCrunch · AIAI Score6262 Microsoft unveils Nvidia RTX Spark AI PCs and revamped Windows 11 with Execution Containers
Microsoft revealed specs and pricing for its Nvidia RTX Spark-based AI PCs at a San Francisco event during Tech Week.
AWS Machine Learning BlogFeaturedAI Score7171 Claude Haiku 5.5 launches on Amazon Bedrock and Claude Platform on AWS
Claude Haiku 5.5 is now available on Amazon Bedrock and Claude Platform on AWS. According to Anthropic.
Why it matters: The post details Haiku 5.5's effort controls and subagent role, useful for judging cost and routing tradeoffs.
The DecoderFeaturedAI Score8282 Anthropic releases Claude Haiku 5.5 with up to 90 percent price cuts and large benchmark gains
Anthropic released Claude Haiku 5.5, its fastest and most affordable small model.
Why it matters: The pricing table and benchmark jumps let readers weigh Haiku 5.5 against their own cost-sensitive workloads.
The Verge · AIAI Score4848 Everything announced at Microsoft's Surface Laptop Ultra event
Microsoft unveiled Surface Laptop Ultra release details at its San Francisco keynote.
AWS Machine Learning BlogAI Score2626 Rethinking access control for RAG with Amazon Quick and Amazon Bedrock
Amazon Quick and Amazon Bedrock Knowledge Bases add real-time ACL enforcement on top of pre-retrieval filtering.
The Verge · AIAI Score6262 Microsoft gives Copilot access to local files and OS-level actions in Windows
At its Windows and Surface event, Microsoft showed an upgrade to Copilot that gives it access to local files on a PC and the ability to take actions across the OS.
The Verge · AIAI Score5858 Microsoft Surface RTX Spark Dev Box opens preorder at $5,999, ships in November
Microsoft's Nvidia-powered Surface RTX Spark Dev Box is available for preorder now and is slated to ship in November for about $6.
GitHub Blog · AI & MLFeaturedAI Score6262 GitHub extends push protection to unstructured secrets with a ModernBERT classifier built with Microsoft Applied Sciences
GitHub built a fine-tuned ModernBERT classifier with Microsoft Applied Sciences that assesses candidate secrets in context in under two milliseconds.
Why it matters: GitHub's nine quarters of push data and the latency budget behind its new secret classifier show how prevention is being moved into the push path.
Microsoft ResearchFeaturedAI Score7171 Microsoft Research Asia open-sources Agent Lightning v1.0, a 3,500-line agentic RL framework
Microsoft Research Asia open-sourced Agent Lightning v1.0, a roughly 3.
Why it matters: The original gives the framework's design choices and a measured SWE-bench gain, so readers can judge whether to reuse their existing harness for RL.