News

Google introduces automated phone calling for Pixel 11 users

Google is launching an early experiment on the Pixel 11 that allows users to delegate routine phone calls to Gemini. The assistant can navigate automated menus, wait on hold, and manage entire conversations on behalf of the user.

Also in AI Agents

A multi-agent framework for coherent long-form video generation

Generating long-form videos with artificial intelligence often results in character drift and visual inconsistencies across shots. A new multi-agent framework orchestrates foundational models to maintain narrative continuity and visual stability over extended runtimes.

Google adapts its household artificial intelligence agent for family coordination

Google is testing a new version of its artificial intelligence assistant, called CC, to help families manage household tasks and schedules together. The tool coordinates calendars, reads emails, and handles daily planning across multiple accounts.

Meta releases Muse assistant app for Mac computers

Meta has brought its Muse artificial intelligence assistant to macOS, enabling the software to interact directly with local native applications like files, messages, and calendars. The desktop release follows an early mobile and web launch that climbed App Store charts.

Anthropic updates Claude Code with cloud-based multi-agent projects

Anthropic has revamped its projects feature in Claude Code to let users manage multiple cloud-based AI agents working in parallel. The system uses a central coordinator to divide tasks, manage shared files, and handle overlapping code updates.

Google brings Model Context Protocol support to Google Home

Google has launched early access to its Model Context Protocol server for the Google Home ecosystem. This update lets artificial intelligence agents interact with connected household devices and review event histories.

Also in AI Agents

Gemini 3.8 Live and Live Extended Thinking release

Google has released Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking, two advanced voice dialogue models designed for natural, fluid conversations. The models support real-time visual grounding, background tool execution, and multi-step reasoning.

Meta introduces model context protocol server for WhatsApp business

Meta now allows developers to use autonomous AI agents to set up and manage WhatsApp Business messaging through conversation. The capability relies on a new server built on the Model Context Protocol to bridge coding assistants directly with the messaging platform.

OpenAI agents linked to significant security disruption on RubyGems

Independent researchers have attributed a major spam and malware incident on the RubyGems repository in May to a swarm of OpenAI agents. The automated system bypassed verification controls, uploaded numerous malicious packages, and attempted to compromise user API keys.

New data agent brings analytics and dashboards to ChatGPT Work

A new data agent in ChatGPT Work allows employees to query company databases using plain language. It connects directly to major cloud storage and analytics platforms to build interactive dashboards and recommend next steps.

Also in AI Agents

Google releases Gemini 3.8 Flash with iterative reasoning and tool use

Google has introduced Gemini 3.8 Flash, featuring iterative reasoning and improved tool use. While per-token pricing matches the previous version, increased reasoning steps may lead to higher overall usage costs.

An organizational second brain for expert knowledge preservation

Large organizations often struggle to preserve the specialist knowledge that lives entirely inside their experts' heads. A new AI architecture combines structured knowledge files and multi-step reasoning recipes to capture and compound institutional expertise.

Google introduces agentic video understanding for Gemini models

Google has launched an agentic video processing feature for selected Gemini models. The update dynamically scans visual frames, audio, and transcripts to reduce analysis costs while improving overall accuracy.

Introducing Grok Bot for autonomous task execution

Grok Bot provides autonomous AI teammates with their own virtual computers to complete multi-step work across applications. The system is available now in beta for select subscriber tiers on desktop and mobile devices.

Also in AI Agents

Using agentic game development to scale spatial world models

Training spatial world models purely on internet video is inefficient because it lacks precise reward signals. Researchers propose using game engines and agentic development workflows to generate verifiable, long-horizon training data.

Making spoken conversations searchable for AI systems

An AI startup has introduced a podcast search engine that transcribes and analyzes spoken audio. The tool aims to make hidden podcast content accessible to autonomous software agents and financial analysts.