Microsoft unveiled a strategy at its Ignite 2026 conference that centers on AI agents doing routine work and managing projects. It is expanding Copilot into a workplace hub where people and AI agents work together, and adding data, security, and management tools.
At Microsoft Ignite 2026, Microsoft unveiled new technologies and products to help organizations speed up AI adoption. It positions Copilot as a new work operating system where people and AI agents collaborate.
Microsoft unveiled Microsoft IQ, a platform that connects a company's internal work data and organizational knowledge to AI agents. It works with Microsoft Copilot as well as ChatGPT and Claude, and was announced at the Ignite 2026 conference in San Francisco.
Google has renamed its AI agent Gemini Spark to Tasks and made it free in six countries, including South Korea. Tasks can carry out multi-step jobs like planning trips from email and calendar data. It is a shift from a chatbot that only answers questions toward a personal assistant that does work.
Amazon AWS Machine Learning blog · Nov 18, 07:44 KST
Flightscape, an airline operations platform, built an AI agent that checks crew training schedules, and it handled 20 MB+ JSON files slowly and at high token cost. The team tried three ways to shrink what gets sent to the model each time, and their tests showed token use down by up to 71% and average latency down by up to 80%. The strategies also aim to keep the agent's reasoning working across long chats.
Microsoft updated its Foundry platform for businesses, which lets developers and companies build, test, and deploy AI apps and autonomous agents. The update lets agents check their own quality, safety, speed, and cost, and improve how they work. Microsoft also released new in-house AI models.
Meta's chief AI officer, Alexandr Wang, posted an AI-generated video showing OpenAI CEO Sam Altman and Meta CEO Mark Zuckerberg on a video call. The clip promotes Meta's AI agent Muse, and Meta says it is allowed because it is labeled as AI. Critics say the platform still allows many deepfake images of real people.
Amazon AWS Machine Learning blog · Nov 18, 05:25 KST
Amazon's AWS team wrote a guide on connecting AI agents to Snowflake's MCP server with secure login tokens. It compares three OAuth setups that control which user or service is allowed to call Snowflake's AI tools.
LangChain published a guide on building proactive agents that act on their own, using cron-style schedules. The post says these schedules power features like memory consolidation, proactive messages, and reminders. It also says many emerging proactive agents are built on this simple pattern.
This talk shows how OpenAI engineers Dominik Kundel and Charlie Guo use Codex, OpenAI's coding agent, in everyday work. It was given at the AI Engineer event, and the video is based on a CODE Summit 2026 session.
Amazon AWS Machine Learning blog · Nov 18, 04:40 KST
Amazon SageMaker AI now offers AI data preparation capabilities for building high-quality training and evaluation data. Generate synthetic examples, assess and improve datasets, auto-label at scale, and bring in expert review through Python SDK APIs and an agent skill, all from your own notebook or IDE.
Amazon Bedrock AgentCore Memory can now keep short-term memory events for up to 10 years, up from one year. This lets developers keep long records of agent conversations for compliance, audits, returning users, or looking back at past sessions. Retention is set per memory resource and applies to each new event, so existing events keep their old retention. Standard storage charges apply while data is kept.
Amazon Bedrock AgentCore Memory now keeps one full summary for each session. It gives a quick snapshot of how the conversation has gone so far. The summary updates as new events come in.
Google is renaming its Spark AI agent to Tasks and making it free for all users. Tasks in Gemini can handle jobs like meal plans, calendar management, and finding recurring expenses, and it runs on the new Gemini 4 Hydrogen model.
Amazon AWS Machine Learning blog · Nov 18, 03:48 KST
Amazon has released SMART, an open source Python toolkit and Terraform setup that handles the WhatsApp connection for AI agents built on Amazon Bedrock AgentCore. Developers keep control of their agent's logic while SMART takes care of messages, media, memory, and scheduling. The article says one command deploys the whole infrastructure to an AWS account.
Microsoft shared new features for its Edge for Business browser at its Ignite 2026 conference. Agentic browsing, which lets AI agents finish multi-step tasks across websites, is now part of Cowork and is generally available. A skills recorder for turning workflows into reusable steps is coming in preview, and faster agent performance is being tested.