You are viewing a single comment's thread from:

RE: 𝑰𝒇 π’šπ’π’– 𝒂𝒓𝒆 𝒂 π’‰π’–π’Žπ’‚π’, you may be affected by the 𝒇𝒖𝒕𝒖𝒓𝒆 𝒐𝒇 𝑨𝑰

in #artificialintelligence β€’ 4 years ago

I don't agree on the slope.

  1. We are probably more already in this slope
  2. While we have technological progress, we often have regress on a society level. So your slope fits in technology but not for general human progress imo.
Sort: Β 

The main point is as technology advances, it starts to accelerate rapidly (nearly exponentially).

If we developed a super AI that is 1x our intelligence, within a short time it will be 2x, then 10x, then 50x, then 1000x, and so on.

The chart represents the entire timeline of the human species. If you think about it, most of the real technology we have developed has been in the last couple hundred years.

As a species, we have been around for 315,000 years, but only the last 50,000 years represent what you would consider modern man.

The amount of technology and advancement in the next 100 years will dwarf the first 50,000 just like the next 5 years will likely dwarf the last 20-50. The relationship between those numbers over time will get even more ridiculous.

That's why I meant that we are probably already in the slope and not before it.
But we can't forget that as it goes up, it easily stagnates as well.
Especially AI, AI is stagnating for a long time already. The AI we are building at the moment are the same we're building for a while just with more processing power.

Similarly, quantum computers might be ready for commercial shipping relatively soon already, but until we understood for what they're good a long time might pass still.

And, the more the society becomes resistant to change, they can slow this down strongly and might bring us to a more modern type of "medieval area" similar to the technological expansion we saw in the antique which was then stopped for a long time by religion.

AI is still moving along pretty fast, but not at exponential levels and it is still limited to very focused tasks (which is a big problem for real progress).

New techniques are coming out daily but they mostly are coming out solving the same problems, just faster and more accurately.

Society is eating up AI, it is being integrated into almost every device developed these days and businesses are focusing on AI to solve tons of problems and reduce costs dramatically. I don't buy society is resisting the change, quite the opposite.

Oh, not yet. But eventually, when even more people don't have jobs because AIs take over and we move in the same direction with the "fuck them" attitude, they will.

Meh, I read quite a lot into AI research. What is changing are the applications they come up with the techniques exist already for a long time (at least in the scientific community).

Loading...
Loading...