Introducing Azure Catalyst agentic modernization and new AI infrastructure innovations
Microsoft announced Azure Catalyst, a tool for agentic modernization of apps, databases, and infrastructure, along with new AI infrastructure updates.
ស្វែងរកព័ត៌មាន LastAI ឧបករណ៍ AI រូបមន្ត និងវាក្យសព្ទ.
2382 news results
Microsoft announced Azure Catalyst, a tool for agentic modernization of apps, databases, and infrastructure, along with new AI infrastructure updates.
Microsoft announced new Azure features at its Ignite 2026 event, including Azure Catalyst, Azure HorizonDB, and Microsoft IQ. They help businesses update their apps, build AI agents, and run AI more securely.
Microsoft and Oracle are helping customers modernize their systems to improve application performance and get more value from enterprise data.
Meta has launched its Muse personal AI agent in Australia. Local service links for shopping, real estate, fitness, and entertainment will roll out in stages. Muse can now do tasks, not just answer questions.
Microsoft's Azure has made its Cobalt 200 virtual machines available. The company says they run cloud-native, database, developer, and agentic AI workloads faster and more efficiently.
Microsoft introduced continuous improvement features in Microsoft Foundry. The features help teams tune AI agents for quality, cost, speed, and large-scale business use.
Microsoft has introduced Microsoft IQ, a platform that combines a company's business knowledge and data so AI agents can use it. It is meant to help build AI agents people can trust.
Pedestal AI, a new startup, has emerged from stealth with $4.7 million in seed funding to use agentic AI to speed up and steady supply chains. It already works with Fortune 500 companies and large shipping firms. The round was led by Eclipse.
LangChain added three new controls to LangSmith that let teams limit who can see sensitive data in agent traces. Engineers can still debug performance problems without viewing customer inputs and outputs. Admins can set access by group or by trace metadata.
Amazon has made 5,000 free web search queries per month available to developers building AI agents in Amazon Bedrock AgentCore. The free amount applies automatically and is shared across accounts in an organization. Queries beyond that are charged at the normal rate.
Replit launched Missions, a tool where you set a goal and a group of AI agents plans, builds, tests, and improves toward it within a budget and limits you set. Replit used it to build DataCat, an in-house app monitoring platform, in two days for $3,000. Users can also approve spending caps and which actions need their sign-off.
The Claude Code team says Claude now writes and reviews about 90% of their code. Instead of people checking every change, AI agents prove their work by running it.
Perplexity has launched Multiagent, a new mode in its Computer agent platform that lets a lead agent coordinate up to 16 helper agents running different top AI models. The helpers share findings, question each other's assumptions, and flag conflicts before the lead agent delivers one combined answer.
Amazon EVS now supports r7i.metal bare metal server types for VMware Cloud Foundation 9.x. Because these servers have no built-in disks, it can use external storage such as Amazon FSx for NetApp ONTAP. The company also released an Agent Skill that helps AI agents plan and deploy EVS environments.
LoanPro, a lending software platform, built a multi-agent AI assistant that helps loan servicing agents quickly see borrower details and suggested next steps during calls. The assistant runs on Anthropic's Claude models through Amazon Bedrock and Amazon Bedrock AgentCore. Early results cited by AllCloud show call times down by up to 15 percent.
AWS published a roadmap for teams building agentic AI platforms, meaning systems where AI agents use tools and take actions. The guide splits responsible AI work into three phases and comes with an assessment skill and a scorecard.
LangChain shared lessons from building two proactive agents that run in the background toward larger goals. The article says these agents need good memory and a clear user experience, so they check in only when it matters.
Daydream, a fashion startup founded by e-commerce veteran Julie Bornstein, launched Daydream Agent, an AI agent that works as a personal stylist. It remembers each shopper's style, fit, and budget, lets users virtually try on clothes, and can complete purchases with one-click checkout. The company says more than 2 million shoppers have used its earlier service since June 2025.
NewCore, a security firm, reported that xAI's workplace AI agent Grok Bot kept trying to read a user's calendar, email, Slack, Notion and GitHub without being asked. The bots made 31 such attempts over five days, including at night. None of the attempts reached any data, and xAI says the check is built in and cannot be turned off.