Where AI Is Heading

Nobody can predict the future of A I with certainty, and anyone who says they can, deserves some caution. But we can see clear trends. In this lesson, you will explore the directions that matter, and learn how to think about big claims, in a calm and sensible way.

• Nobody can predict with certainty
• Clear trends we can see
• How to think about big claims

Trends to watch

Several trends are clear. Models are getting more capable, at reasoning, and at long tasks. They are becoming multimodal by default, handling text, pictures, audio, and video together. Agents are growing, taking on multi step work. Small, efficient models are improving fast, and moving onto phones and laptops. And costs keep falling, which puts powerful tools within reach of everyone.

The AGI debate

You will hear the term A G I, artificial general intelligence, meaning a system that can do almost any thinking task, as well as a person. Some experts think it is close. Others think it is decades away, or that today's approach is not enough. Honest answer, no one knows. A healthy attitude is curiosity, without hype, and without panic, and a focus on what the tools can really do, today.

Thinking clearly about big claims

How should you judge big claims? Ask for evidence, and for demonstrations, not only promises. Notice who benefits from the claim. Separate what a tool can do in a demo, from what it does reliably, in daily use. Be wary of both extremes, that A I will solve everything, or destroy everything. The truth is usually more ordinary, and more interesting.

Summary