The era of the passive chatbot is ending; the era of the active agent has arrived. Google officially launched Workspace Studio this week, a new no-code platform that allows any employee, regardless of technical skill, to build custom AI agents powered by Gemini 3.

For years, the promise of AI has been limited by the blank cursor problem: users didn't know how to prompt. Workspace Studio solves this by integrating deeply into the apps companies already run on. Instead of chatting in a void, users can now click-and-drag to create agents that monitor a shared Drive folder for new invoices, triage incoming Gmails based on sentiment, or autofill Sheets with data extracted from PDFs.

Instead of just simple tasks, Studio allows for complex multi-agent workflows. A prime example comes from cleaning giant Kärcher, an early adopter. They deployed a virtual team to handle new feature ideas: a Brainstorming Gem assesses merit, a Technical Gem checks feasibility, and a UX Gem maps user flows. The result? A 90% reduction in drafting time, turning hours of manual consolidation into a ready-to-review plan in two minutes.

For IT directors sweating over "shadow AI," Google is rolling out centralized visibility, but it’s still early days. While employees can build agents now, the robust admin controls to audit and manage these agents are currently launching via the Gemini Alpha program. The goal is a walled garden where innovation happens fast, but IT eventually gains the keys to lock down sensitive data and enforce policy guardrails.

Apparently, Google is doing something right, as the U.S. General Services Administration recently announced it would be adopting Google Workspace.

Google isn't just selling a tool; they are trying to solve the adoption gap. By moving AI from a chat interface to a workflow interface, they are betting that the killer app for AI isn't writing poetry, it's doing the boring work that humans hate.

Early metrics from their alpha testing show over 20 million tasks already automated, suggesting that when you make agent-building boringly simple, people actually use it.

In a reversal of Biden-era policies, President Trump announced he will allow Nvidia to export its powerful H200 chips to China, subject to a 25% tariff. The move, confirmed via Truth Social, signals a shift from "total denial" to "tax and monitor."

The decision settles a fierce internal debate in Washington. While security hawks argued for a total blockade to starve China's AI development, industry lobbyists warned that China was simply catching up by building domestic alternatives like Huawei's Ascend series. Trump's "thoughtful balance" allows US companies to capture Chinese revenue (and pay a hefty cut to the US Treasury) while theoretically keeping the absolute cutting-edge chips (Blackwell and Rubin) out of Beijing's hands.

Wall Street immediately voted with its wallet, pushing Nvidia shares up 2% in after-hours trading as investors calculated the potential windfall. Analysts estimate that reopening the Chinese market, historically responsible for 20-25% of Nvidia’s data center revenue, could add $2–5 billion annually to the company’s top line, even after the government takes its 25% cut.

The logic is brutal but simple: the H200 is roughly 6x more powerful than the restricted H20 chips China was previously allowed to buy. By flooding the market with "second-best" American silicon, the administration hopes to hook Chinese AI labs on Nvidia’s CUDA ecosystem once again, stalling their migration to domestic Huawei hardware before it becomes irreversible.

Critics call it a colossal failure of national security, fearing the H200 will accelerate Chinese military AI. Proponents argue that if China is going to get chips anyway via smuggling rings, one of which was just busted moving $160M in hardware, the US might as well control the supply chain and collect the cash.

🎧 Yahoo launches personalized AI audio digests for morning commuters

🩺 Doctors in Jersey turn to AI for rapid skin cancer screening

📈 Anthropic reportedly targets huge valuation for 2026 IPO listing

🛎️ Virgin Atlantic selects OpenAI to power new travel concierge

💰 Microsoft invests billions to expand Canadian AI cloud infrastructure

🔍 EU investigates Google over AI-generated search result summaries

Here’s what to do:

📂 Step 1: Locate 3-5 related documents in your Google Drive (e.g., a project plan, a budget sheet, and a meeting notes doc).

👀 Step 2: Open a blank Google Doc and summon Gemini (click the sparkle icon).

✍️ Step 3: Type the following prompt (using the @ symbol to tag your specific files):

"Review @ProjectPlan, @Budget, and @MeetingNotes. Write a purely factual executive briefing that highlights the top 3 risks, the current budget burn rate vs. projection, and the immediate next steps for this week. Format as a bulleted memo."

🧠 Step 4: Fact-check the output. Did it catch the nuance in the meeting notes?

⏱️ Step 5: Save this prompt. You just saved yourself 45 minutes of reading.

From agents that do the dishes (metaphorically) to tariffs that change the chip war, the landscape is shifting fast, and we’re here to help ensure you don’t fall behind. See you next time! 🚀

Stay informed, stay curious, and stay ahead with Jumble!

Zoe from Jumble