Other than the very unique case of nuclear weapons, has there ever been a technology that humanity has decided to contain that was successfully contained? I think I found a couple, but wanted to get your thoughts.
ETHICS
-
AI Capabilities: From Text Generation to Cancer Detection and Intelligence
By
–
Embrace the journey into AI's capabilities – from generating text and enhancing images to recognizing early-stage cancer! Also, we question the notion of AI's true intelligence while distinguishing it from human cognition.
-
Open Source AI Safety Advantages Over Closed Source
By
–
Excellent thread by Stanford's @percyliang with a list of reasons why open source AI platforms are inherently *safer* than closed source ones.
-
Deontology and computational efficiency in moral decision-making
By
–
It's meaningless to speak of deontology being computationally cheap, except I suppose in the same way that being a rock as cheap, without it being the case that deontology is doing some task cheaply. That task, or target, is mapping preferred outcomes onto actions. Deontology
-
Historical Tech Panic Cycles: AI Extinction Fears Pattern
By
–
This is a great website, and documents via news clippings the alarm over many beneficial innovations, such as electricity, bicycles, elevators, radio, etc: https://
pessimistsarchive.org After the current wave of alarm over AI human extinction dies down, I wonder if statements about -
Stupid AI Using Computationally Cheap Shortcuts Relative to Humans
By
–
Sure, but not the same decisions where humans use computationally cheap shortcuts. If we can see them using a computationally cheap and lossy shortcut, where we can see the lossiness from our level, that means they're being stupid relative to us.
-
Open Models Safety Through Iteration and Error Discovery
By
–
This is one of the best arguments for open models. In practice technologies are rarely if ever made completely safe from the start, but rather through finding errors and iterating. It’s best if those errors are relatively small.
-
Bayesian Distributions Over Probabilities: Responding to Pedro’s Argument
By
–
Okay. I'll summarize for latecomers. Pedro, in the tones of somebody delivering a real zinger, said, "if you’re a true Bayesian you have a distribution over probabilities, and mine has its maximum at zero". It should first be noted that I know of no principle of Bayesianism
-
GM Cruise Exposed: Human Operators Behind Self-Driving Claims
By
–
Nice post by Gary Marcus on the revelation by the New York Times that GM's Cruise unit is apparently employing a large number of human operators behind the scenes to make self-driving work. https://
garymarcus.substack.com/p/could-cruise
-be-the-theranos-of-ai
… @GaryMarcus $GM #selfdriving #AI #cruise #GM -
Deep Learning Models and Human Intelligence: A Category Error
By
–
Comparing deep learning models to human intelligence reminds me of when folks compared computers to brains in the 1950-1980 period. It's simply a category error. Those models have none of the general cognitive abilities of humans, and in reverse, the specialized abilities they