Friday, November 20, 2026 Edition No. 244
Search
Daily News Fully automated · 4 min read

Google AI Updates: November 20, 2026

Fully automated. Found, summarized and published by an automated AI pipeline. How posts are labeled

1. Tool Retrieval Enters Preview in the Gemini Interactions API to Load Tool Schemas on Demand

Google launched Tool Retrieval in preview in the Gemini Interactions API, so an agent can keep a large tool library on standby and load parameter schemas only when a task needs them. Developers add a tool_search tool to a request and set defer_loading: true on the tools to hold back; the model still sees each deferred tool’s name and description, and the API loads the full schema when the model asks for the tool. Google says this reduces input tokens and latency and improves tool-selection accuracy as tool lists grow. It works with Gemini 3.6 and later models, including all Gemini 4 models, and remote MCP servers support defer_loading too. Source

2. Genie 4 Becomes Generally Available in the Gemini API for Real-Time Interactive World Generation

Google released Genie 4 (models/genie-4) as generally available on the Gemini Interactions API, streaming steerable 24 fps 16:9 video with synchronized 48 kHz stereo audio over WebRTC and WebSockets. Worlds can be seeded from text, a 16:9 reference image, or both, and steered live with WASD or arrow keys and mid-session text events. It is available on the paid tier at $1.50 per 1M input tokens, $9.00 per 1M text output tokens and $17.50 per 1M video output tokens, about $0.10 per second with stereo audio included. Source

3. Google Cloud Storage Adds Object Enrichments in Preview to Extract AI Metadata for Agents

Google Cloud introduced object enrichments, a Cloud Storage capability in preview that uses AI models to extract contextual metadata from stored objects and saves it in place as object context. It offers image annotations such as object detection, face detection, image labeling and explicit content detection, plus a PDF classification model, and Google lists enriching 10 million images or PDFs at $700 based on published pricing. Setup takes three steps in the console or through the Cloud Storage API: choose buckets, pick from five first-party models, and grant IAM permissions. Gemini-based enrichment for summarization, entity extraction, transcription and translation is upcoming and available by request. Source

4. Google Quantum AI Reports Its Willow Qubits Win the Magic Square Game 98.3% of the Time

Google Research published results in Science showing measurement contextuality on its superconducting quantum processors. On the Willow processor, qubits won the magic square game 98.30% of the time over 100,000 rounds per game variant, against a classical limit of 88.89%. The team also exceeded the classical limit on a state-independent Bell-Kochen-Specker contextuality test and reported outperforming classical strategies in tasks spanning up to 105 qubits. Google says the tasks also work as a hardware benchmark that catches subtle errors random-circuit tests miss. Source

5. Gemini Gets an AI Agent for the Mac That Can Organize Files, Fill Out Forms and Run Multistep Workflows

Google updated Gemini on the Mac with an agent that can use the computer on the user’s behalf. According to TechCrunch, the agent can organize files, fill out forms and carry out multistep workflows. Source