AI in 60 Seconds 🚀 - AI Agents Ate the Desktop — Why Work Now Starts in AI
|
AI Agents Ate the Desktop — Why Work Now Starts in AISep 24, 2025 The Productivity Suite Era is Ending—And That’s Natural EvolutionLast week, roughly 800 million people started a task in ChatGPT. Our global tracker shows that for every 100 sessions in a native-AI app like ChatGPT, there are fewer than 5 AI sessions inside productivity-suite companions from Microsoft, Google, and others, and there is a 20% drop in time spent inside classic apps among power users, as conversations replace clicks. The productivity suite moved us from paper to pixels in the 1990s. Now we’re moving from pixels to conversation.
Here’s what that looks like: I’m on a trail overlooking Puget Sound, talking with my AI COO, Elizabeth. No laptop, just voice. As we speak, she’s orchestrating dozens of agents to handle the entire workflow. The productivity suite isn't where the work happens anymore; it's just where the finished product lands.
It's surreal. In 1996, I was coordinating the launch of Office 97 in Latin America, and that felt like the future; that future lasted 29 years until generative AI transformed it in less than 18 months. Work begins with "Let's talk about this" instead of "File > New."
🎧 For stories and behind-the-scenes insights— listen to our 12-minute podcast.
🚪 The new front door to work is conversationWork now starts in AI. The thinking, drafting, analysis, and routing happen in the conversation; tools are destinations, not the stage. Most leaders envision classic productivity activities being automated with AI; our data shows that the most significant savings are in areas such as light industrial, maintenance, research, and problem-solving. Where Talking Beats Typing: Real Work, Real SavingsRepairing Equipment (describing symptoms, getting step-by-step guidance)
Look beyond the desktop to find exponential gains with AI: While knowledge workers save approximately 24 minutes on emails, meeting summaries, Word documents, and spreadsheets, conversational AI saves frontline workers, such as technicians, support staff, and field operators,94 minutes per task. Nearly 4× the impact. Why? Frontline work is inherently conversational. You describe a problem. You talk through symptoms. You explain what you’re seeing. The productivity suite revolution never reached these 3 billion workers because you can’t fix a furnace with Excel. However, you can resolve it by discussing the issue with AI. The biggest productivity gains are happening on factory floors, in field service vans, and on support calls. 📊 Two signals of the end of an era
The decline in use of traditional desktop apps signals a “massive” shift not just for knowledge workers, but also for the three billion frontline workers who are not sitting at a computer all day. Suddenly, they can create, troubleshoot, and document everything from a phone - just by having a chat.
I personally reduced my use of productivity apps by 50% as my virtual COO, Elizabeth, accesses those apps for me. Over the past seven days, she completed 1,100 tasks, including the creation of 100 documents, in approximately four hours. It would have taken me 230 hours!
🧠 Why AI at the center beats AI as a companionYou state the goal once; the AI plans the steps, calls the right tools via APIs, and returns a finished artifact. Apps become destinations, not the place where work begins.
AI at the center gives Makers velocity: ChatGPT and Claude (via tool use and MCP) now ship with off-the-shelf connectors to Google Workspace, Microsoft 365, Slack, Teams, Notion, Confluence, Salesforce, HubSpot, Jira, GitHub, Asana, Trello, and more. Makers snap together data and actions to ship mini-agents in hours; bottom-up wins scale fast, while top-down, app-by-app companion rollouts stall. 🧩 Grassroots and youngIndividual impact:
Team spread: About 20% of personal agents get shared with teammates. Someone solves their status report problem, shares the agent, and suddenly the whole department saves hours using that agent. The agent handles Word, formats reports, sends emails; apps become invisible. Enterprise reality:
Subject-matter experts are building personal agents that excel in a single task. These aren’t IT projects—they’re individuals solving their own problems through conversation. The multiplier effect happens naturally: personal solution ➡️ team adoption ➡️ department standard ➡️ enterprise scale. Case Study:
The CFO of a global professional services firm came to us after their big, top-down AI project burned through six months and nearly a million dollars… with nothing to show for it. We flipped things around; we rolled up our sleeves and empowered the frontline teams to build mini-agents. Eight weeks later, they had measurable results and projected over 70,000 hours saved for the year; that’s over $ 7 million in annual savings.
☝️ The only question that mattersStop asking “How do we add AI to our workflow?” and start asking, “What if this workflow were just a conversation?” The answer will show you where the 4× gains are hiding. 🔮 One more thingFor 30 years, we adapted our thinking and our way of working to fit software’s limitations: menus, clicks, keyboard shortcuts. Now AI adapts to us, meeting us where we’ve always been most productive: in conversation. The desktop isn’t dead. It just became irrelevant the moment work started with “Let’s talk about this.” 🚀 Take Action
✅ Ready to transition from a traditional organization to an AI-powered one?We advise forward-thinking organizations to develop strategic frameworks for evaluating, integrating, and optimizing human-AI production units. Let’s discuss how this applies to your organization. Contact Us. Luis J. Salazar | Founder & Elizabeth | Virtual COO (AI) Sources: Our insights are based on over 250 million data points from individuals and organizations that used our AI-powered tools, participated in our panels and research sessions, or attended our workshops and keynotes. 📣 Use this data in your communications, citing "AI4SP" and linking to AI4SP.org. 📬 If this email was forwarded to you and you'd like to receive our bi-weekly AI insights directly, click here to subscribe: https://ai4sp.org/60 |