At Adobe MAX 2026, Adobe announced support for the Model Context Protocol (MCP), an open standard that connects AI with outside apps. It lets external AI agents operate Adobe's apps such as Photoshop. A private beta is set to arrive in early 2027.
Google has released Gemini 4 Carbon, a model designed for AI agents that handle coding and security tasks on their own. The introduction does not describe its features in detail.
Amazon AWS Machine Learning blog · Nov 12, 07:26 KST
AWS summarized the AI models and features it released in October 2026. The focus is on choosing the right model for each task, keeping AI agents (programs that carry out work on their own) reliable, and making costs easier to predict.
In spec-driven development, a flawed spec leads a coding agent to build the same flaws. This video shows how AWS's Kiro finds spec errors before any design or code is generated.
Atuin, a tool for searching and syncing shell history, now also lets you search and resume coding agent sessions. For example, you can pick up Claude Code work from a dev server in Codex on your laptop. Session collection only starts when you turn it on.
This video shows how Duolingo used several AI agents to automatically raise test coverage (the share of code that tests run through) in its iOS app. The speaker is Duolingo engineer Kush Agrawal.
Amazon AWS Machine Learning blog · Nov 12, 05:45 KST
DXC Technology built a shared base on Amazon Bedrock, AWS's managed AI model service, to run several AI models through one interface. The company runs IT for banks and hospitals with strict rules. On top of this base it added DXC OASIS, a layer that oversees AI agents (programs that carry out tasks on their own).
OpenAI has released version rust-v0.167.2 of Codex, a lightweight coding agent that runs in the terminal. The introduction does not describe the specific changes.
ElevenLabs unveiled the ElevenAI workplace app and a partner program at its New York summit. The move expands the company from voice generation and transcription into customer service and staff workflow software. It also plans to tune some open models using data from real business use.
Amazon AWS Machine Learning blog · Nov 12, 01:05 KST
AWS Quick uses several AI agents to gather scattered customer signals and draft product planning documents. People approve each step, and the record of why decisions were made is kept from planning through the handoff to development.
This video shows how Peter Yang runs his one-person business with a personal AI system. It covers an AI chief of staff agent and several automated workflows.
Amazon AWS Machine Learning blog · Nov 12, 00:50 KST
AWS's machine learning blog introduced UAEF, an evaluation system for AI agents. It scores agents built on different frameworks with the same metrics and compares agent versions. Organizations in telecom, healthcare, and financial services used it before releasing their agents.
Google DeepMind's AI tool produced a mathematical proof that the encryption code in the Signal messaging app works as promised. An AI agent team finished the work in four weeks, and no new flaws were found.
Databricks has released an updated version of its AI agent Genie. It can carry out long, multi-step tasks without human supervision, using an ontology (a structured map of how data relates) built from customer data.
Google DeepMind researchers used agentic AI to mathematically prove key security guarantees of the Signal messaging protocol. The proof-of-concept project took just four weeks.
AI agents, which handle multi-step tasks on their own, are still considered too unreliable to run without close supervision. Startup Plurai says it will use the same kind of simulation that helped bring self-driving cars to roads to tackle this problem.
This video covers AI agents (programs that act on your behalf) connected to email, seller centers, and payment cards. It walks through five settings to change, based on five real incident cases, in about ten minutes.
Voice AI company ElevenLabs has formed an alliance with consulting firm McKinsey. The goal is to help companies roll out AI customer service agents across their organizations. They will first prove value in a few use cases, then expand step by step.