Former Google CEO Eric Schmidt forecasts three major AI developments arriving soon, predicting their combined impact will be profound.
"In the next year, you're going to see very large context windows, agents, and text action," Schmidt stated. "When they are delivered at scale, it's gonna have an impact on the world at a scale that no one understands yet, much bigger than the horrific impact we've had on by social media, right in my view."
Here's a look at the three key areas he outlined:
1. Very Large Context Windows: AI's Expanding Memory
These large context windows will essentially function as AI's short-term memory, allowing them to process and recall much more information within a single interaction. Schmidt expressed surprise at their potential size, while also noting a current limitation that mirrors human cognition:
"In a context window, you can basically use that as short-term memory... the interesting thing about short-term memory is when you feed the... you ask it a question, read 20 books, you give it the text of the books is the query, and you say, tell me what they say. It forgets the middle, which is exactly how human brains work too, right? That's where we are."
2. Agents: AI That Learns and Acts Autonomously
The development of AI agents signifies a shift from passive response to active learning and task execution. These agents can independently explore subjects, test hypotheses, and refine their understanding:
"With respect to agents, there are people who are now building essentially LLM agents. And the way they do it is they read something like chemistry, they discover the principles of chemistry, and then they test it, and then they add that back into their understanding, right? That's extremely powerful."
3. Text to Action: Turning Words Directly into Deeds
This third advancement involves AI translating natural language commands directly into executed tasks, moving beyond information retrieval to task completion. Schmidt emphasized this capability, illustrating its power with the potential for complex, instant actions based on a simple request. Imagine telling an LLM:
Plan a detailed 7-day Costa Rica trip for two focusing on nature and adventure, budget $3,000 (excluding flights). Find 3 available eco-lodges near Arenal Volcano for August, suggest daily activities like zip-lining and hikes, list good local dinner spots, and put it all in a shareable document with maps and booking links. Do this in 5 minutes and revise if I don't like the first version.
The speed ("Boom, boom, boom, boom," as Schmidt put it) and complexity of tasks achievable through simple text commands are key aspects of this development.
These three advancements – AI with vastly expanded memory, AI agents capable of autonomous learning and action, and AI that turns text commands into immediate results – form the core of Schmidt's near-term forecast. He believes their combined arrival within the next year will reshape our world in ways we are only beginning to grasp.