What happens when AI stops answering questions and starts taking action?

In this episode, I sit down with Peter Lee, President of Microsoft Research, to unpack what actually has to be true for AI to become a real coworker, not just a chatbot.

This conversation is grounded in Microsoft’s latest roadmap, What’s next in AI: 7 trends to watch in 2026, which lays out how AI is shifting from tools we use to systems that act inside real workflows.

We talk about:

Why AI that helps teams cannot live in a chat window
How multi-agent systems create trust through disagreement
What AI orchestration really means 
How AI could shrink global health gaps 
Why AI is becoming part of the scientific method itself
What developers need to know as AI starts understanding code in context, not snippets
Why the next leap in computing is closer than most people think

📄 Referenced article:
What’s Next in AI: 7 Trends to Watch in 2026
https://news.microsoft.com/source/features/ai/whats-next-in-ai-7-trends-to-watch-in-2026/

👤 Guest:
Peter Lee, President, Microsoft Research
https://www.linkedin.com/in/peterlee4/

Tiff In Tech

Microsoft Research identifies seven key AI trends for 2026, focusing on AI's evolution from a solo tool to a collaborative workforce partner. Key developments include multi-agent orchestration systems that enable AI to work alongside teams, built-in safeguards through challenger agents that verify outputs, applications in healthcare diagnosis through structured disagreement, participation in scientific research loops, smarter infrastructure beyond raw computing power, repository-level code understanding, and quantum computing integration. The shift emphasizes orchestration and action over individual model capabilities, with the most valuable skill becoming knowing what to delegate versus what to keep human.

The 7 Biggest Trends in AI for 2026