The words we use to name things change our perception of those things independently of their actual nature. A lot of the hype and mystique around AI is due to our use of terminology from the fields of neuroscience and psychology.
Show this thread
-
Take any paper that make heavy use of such metaphors and replace the neuroscience vocabulary with the corresponding terms from mathematics or signal processing (e.g. "fully-connected neural layer" -> "affine transform"). It suddenly sounds very different. It cuts through the hype
Show this thread
This is true at the level of scientific paper, and even truer in public discourse. The term "AI" encourages people to think that our goal is to create artificial humans. In fact, automating cognition is not about replacing ourselves, it is about augmenting our capabilities.
-
-
Ouside of media and company communication, you don't see the term AI in scientific paper.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
The goal was, and has been, to create artificial human brain. I find it hard to believe that the goal would ever change. Sincerely. If you gave up, the next generation would start again.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
This Tweet is unavailable.
-
They wrote in blogpost that separate model has to be trained for each specific usecase. I understand that lots of recent ML breakthroughs look very impressive (AlphaGo, Dota 1v1 etc) but it's premature to assume that these easily extrapolate to something like AGI.
End of conversation
-
-
-
We should be seeing ads for: How will GDPR affect your AI augmentation?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
augmented intelligence
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
You may like. AI overhype & reality. https://www.linkedin.com/pulse/artificial-intelligence-suffering-over-hype-tornado-bill-schmarzo/?published=t …
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I definitely approach it this way. But many mainstream AI research orgs explicitly target "general intelligence" (eg DeepMind) so the field itself is encouraging the "artificial human" mindset that is in the public discourse.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
There is no common goal in AI Research that I am aware of.
#automatinginequality shows that the application of „affine transforms“ follows different goals than you do.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.