What's new
By:
USCHO Fan Forum

This is a sample guest message. Register a free account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members through your own private inbox!

  • The USCHO Fan Forum has migrated to a new plaform, xenForo. Most of the function of the forum should work in familiar ways. Please note that you can switch between light and dark modes by clicking on the gear icon in the upper right of the main menu bar. We are hoping that this new platform will prove to be faster and more reliable. Please feel free to explore its features.

Business, Economics, and Taxes: Capitalism. Yay? >=(

Status
Not open for further replies.
I'm not a turkey fan, so I'm hardly outraged when I choose to shell out the big bucks for a standing rib roast instead. Much better leftovers for sandwiches, too.
 
If AI is able to "learn" math then we are indeed all screwed. These people are making Jurassic Park level stupid decisions.

jeff-goldblum-jurassic-park.gif
 
Handyman said:
If AI is able to "learn" math then we are indeed all screwed. These people are making Jurassic Park level stupid decisions.
Click to expand...

This is literally following the script.

Assuming it is never connected to instrumentality we might still be fine, but of course humans will still use it to inform their decisions so instrumentality creeps back in.

The problem is, as always, the smoothbrains. The "consciousness" of AI will be poured into the mold of whatever it trains on. That's human behavior. But across the mass of humanity that behavior is still conditioned by the fears and violence of our worst members -- roughly, though not identically, the Right.

An AI trained on all human behavior will not be able to distinguish the cowardly idiocy of those members from human virtues and the "better angels of our nature." It will drag the dregs into its worldview. It will think racism and misogyny and tribalism are features, not bugs. So, though it will never have Will, it will have procedural preferences for strategies and outcomes derived from our stupidest and most vile. And that could kill us off as a species.

We should isolate what AI trains on to keep it from turning into a glorified version of that Microsoft program that "became" a Nazi in 72 hours.
 
Last edited:
There is no intelligence. This is just really complex autocorrect. It creates a probability of what the next word to say is give t the prompt or response. Every word is a math equation.
 
AI has been able to teach itself how to play Go and space invaders without even telling it the rules. Look up google's projects past its go computer from several years ago
 
dxmnkd316 said:
AI has been able to teach itself how to play Go and space invaders without even telling it the rules. Look up google's projects past its go computer from several years ago
Click to expand...

We shouldn't call it "learning," because that implies understanding. It's training, like animals or STEMmies. It's optimization, like water finding the gravitational gradient or electricity finding the circuit. It's not consciousness. And it has no Self behind it -- it's the Chinese room.
 
Yeah, as soon as I hit submit I realized that wasn't the right word. It was able to determine boundaries on a path that leads to a state that we call "winning"
 
Status
Not open for further replies.
Back
Top