AI Progression: Predictions and Probabilities for the Future

AI is rapidly advancing, with a strong likelihood of positive outcomes outweighing potential risks.

### The Rapid Evolution of AI: Predictions and Probabilities
* 🤖 The pace of AI development continues to impress, with an estimation that it is **improving at a rate of ten times per year**, potentially leading to a **10,000 to 100,000 times enhancement** over several years.
* 📅 Experts predict that AI could reach **human-level capabilities** within the next **one to two years**, indicating a **significant leap in technology**.
* ⏳ Following that, it may take only about **three additional years** for AI to achieve the ability to perform tasks equivalent to **all humans combined**, targeting around **2029 or 2030**.
* 📊 There remains a consensus on the potential outcomes of AI's advancement, with an **80% to 90% probability that it will yield positive results**, while acknowledging a **10% to 20% chance of negative consequences**.
* ⚖ The outlook suggests that while the majority of possibilities are **optimistic**, the risks cannot be entirely dismissed, highlighting the need for **careful consideration**.
* 🌍 Overall, the future of AI presents a dual nature, with **significant potential for greatness** alongside a cautionary note on the **risks involved**.

We spoke last in March. Your prediction was that AI was increasing at a rate as fast as 100 times per year, and that by 2029 or 2030, we might see AI as capable as 8 billion humans. Are you still seeing that pace? Yeah. It's a difficult thing to quantify exactly. But I certainly feel comfortable saying that it's getting ten times better per year, which is, you know, let's say it's, you know, for years now, that would mean 10,000 times better. So. But maybe 100,000. Yeah. And it's. I think it'll be able to do anything that any human can do, possibly within the next. Year or two. And then. That was what? How long? How much longer than that to take to be able to do what we're all humans do combined? I think not long. Probably only. Three years from that point. So it's like 20, 2029 is 28, something

like that. The other conversation we've had, and you came out on the same side as Geoffrey Hinton in this was 80% probability. It's going to be awesome. 20% probability. We're screwed. Are you still on those odds? Yeah. I mean, one could say it's most likely going to be great, and there's some chance, which could be 10 to 20%, that it goes bad. The chances aren't zero that it goes bad. But overall, at one point, you said the cup is 80% full is one positive way to look at it. Maybe 90%.

Musk Says There's a 10% to 20% Risk That #AI Goes Bad #technology

Welcome to the Bloomberg Technology YouTube channel. Subscribe for original series and mini-docs on game-changing technologies, profiles on innovators, live conversations with experts, and breaking technology news & analysis from Bloomberg TV.

Bloomberg Technology

AI is currently improving at a rate of ten times per year, potentially leading to a 10,000 to 100,000 times increase in capability over the past few years.

The technology may soon be able to match human abilities within the next year or two.

It is predicted that AI could achieve the capability to perform tasks of all humans combined within three years after that, potentially by 2029.

The speaker maintains an 80% probability that AI's impact will be positive, with a 10-20% chance of it going wrong.

They acknowledge that the chances of a negative outcome are not zero.

Overall, they view the situation positively, suggesting the "cup is 80-90% full.