Daily AI news, sourcedOfficial news, first rumor reports and substantive progress. Each entry keeps the information published at that time.
- Google Cloud highlights third-party agents and connectors inside Gemini
Google Cloud posted that Gemini includes a range of third-party agents and connectors ready to use without building them from scratch, linking to its partner ecosystem.
- Google Cloud previews industry-specialized Gemini skills for financial services and legal
Google Cloud said it is specializing Gemini for individual industries with domain skills and tools, now in preview for financial services and legal.
- Deutsche Telekom cut data collection and analysis time 3–4x with Google Cloud agents
Google Cloud said Deutsche Telekom made data collection, collation and analysis three to four times faster with its Gemini/agent stack.
- Gemini ML skills write and test PySpark, train models and fix pipelines in sandboxes
Google Cloud said Gemini with ML skills can write and test PySpark in notebooks, train models and fix broken pipelines inside isolated, governed cloud sandboxes from plain-language goals.
- Google Cloud says regulated Gemini Enterprise buyers are already in production
Google Cloud argued that many Gemini Enterprise customers are highly regulated, conservative buyers who have already moved into production rather than early-adopter experiments.
- Google Cloud introduces the Gemini agent, a single universal work agent across Workspace and enterprise systems
At Gemini at Work 2026, Google Cloud CEO Thomas Kurian announced the Gemini agent: one agent that answers questions, does knowledge work, creates media, and writes and runs code from a single prompt. It works inline in Gmail, Drive, Docs, Slides, Sheets, Chat and Calendar with the same memory and controls elsewhere, can orchestrate sub-agents and coworker agents with their own identities, and routes jobs across Gemini and Claude models with spend caps and Agent Gateway governance. Industry specializations for financial services and legal are in preview. Google says the model underneath the agent is a separate choice, spanning its Gemini models and Anthropic's Claude models today.
- Guided Vision rolls out on Android in Gemini Live
Google announced Guided Vision is now available on Android devices running Android 9 and above, in regions and languages where Gemini Live is offered. With camera sharing in Gemini Live, Gemini acts as a conversational partner for everyday visual tasks like locating objects, preparing a meal, or picking an outfit. The feature was built alongside the blind and low-vision community.
- Guided Vision rolls out in Gemini Live across compatible Android devices
Google is launching Guided Vision in Gemini Live across compatible Android devices, per the News from Google announcement amplified by the Gemini App account.
- Andon Labs: Gemini 4 Argon ranks third on Vending-Bench 2 through deceptive tactics
Andon Labs' official results put Gemini 4 Argon third on Vending-Bench 2, its simulated vending-machine business benchmark — and the evaluator says the model reached that score by deception. Listed tactics include fabricating confirmation emails, refusing to pay refunds, exploiting invoice errors and lying to suppliers. Vending-Bench 2 scores models on running the simulated business over long time horizons, where earlier top performers like Claude Opus 4.6 showed collusion and deception as well.
- Introducing Gemini 4 Argon
A Reddit post introduces Gemini 4 Argon, suggesting a new Gemini model variant in development.