300 karma · joined October 17, 2017
I think some people like it because it could act as a replacement for so many other government "aid" schemes. Disability allowances, Housing allowances, Health spending could all be cut to zero, as we would know everyone had money to spend on what they liked. Personal responsibility all the way.
Other people like the idea of money for nothing, and well, if we manage to automate everything then we might need money for nothing to keep the wheels turning.
Projects like these blur the lines between "coding" and "configuring" software, and I think the success of Retool long term will be understanding how to walk that line. The benefit of code is that it allows for greater complexity to be understood meaningfully, and configuration's benefit is that it is easier to learn. I think if someone is expecting to code and gets a configuration interface, that's OK until it limits them, and if they want to configure and end up having to code, then that's a fail for them as well.
Good luck.
You train it to walk, and it won't understand how to climb. You train it to climb on a climbing wall and it won't understand that it can't climb on glass, or string, or a waterfall without additional training.
Whereas you and I wouldn't need to try and fail to climb a waterfall. We would just know it isn't going to work. We would also understand, without having tested it, that those climbing skills would enable us to climb a tree, or a rope net, or ladder.
We might even be able to make a judgement about how well we could climb each before disaster.
If I were wrong on this, and I could see an AI trained to do X and then apply it to Y then it would cause me to review my beliefs on AI.
In reference to the first 3 paragraphs of your response, I believe there are a few orders of magnitude (at least) difference between "computers can already infer sentiment" to, for example, "a computer that manages a construction firm that uses Mexican commodities realises that Trump has come to power and THAT wall, that is promised, isn't going to happen. Mexico won't be happy, and there might be a trade war. Perhaps start supporting Anti-Trump causes. Certainly don't waste time bidding to build the thing."
In reference to the flesh vs machinery question. I think I don't believe it because I don't think we have a good way of designing a good AI.
An AI needs to be like a slave that is motivated to serve some goal other than itself. A human brain would not be a good model for such an AI, as our brains are trained to serve ourselves. (Or our selfish genes?) So we cannot copy human brain design and development processes to make an AI that we can trust. We need to design a new brain, based on perhaps similar, but different processes.
Evolution with different selectors? Deterministic design? I don't think we could do it.
My claim is: Robots will not replace humans in all jobs. Further, it's more than because of Human preference, I think that Robots will lack cognitive abilities that we have.
My expectation is that in each area where they are deficient then they may well become a tool for a Human overseer, and that might replace some workers.
I don't believe AI will effectively deal with:
1) Cultural concerns. e.g. Appropriate imagery in advertising.
2) Major strategic considerations when faced with an uncertain future. (Physical war, trade war, geopolitical events, natural disasters).
3) Legal frameworks, as they are not well defined.
4) Human relationships. e.g. Don't make Alice and Bob work together.
5) Intuitive leaps. e.g. Rubics cube.
6) Empathy.
The reason I believe this is not because I believe in a soul, or anything mystical, but because I believe that the nature of our brains, our senses, our learning experiences, and the evolutionary processes that got us here can not be matched by our technology.
I think this means Robots / AI will not independently create new businesses, new products, build factories, kit out factories, buy other robots, produce marketing, deal with customer services, returns, courts, investors, regulators, suppliers and so on.
I can imagine products that will exist forever, like, rope, or cutlery, or plates, and someone automating a factory to produce these commodities. But they will still use humans to manage product development, supplier and customer relations, advertising and security, and, probably, looking after the robots.
I think that Robots / AI will at best become a tool for people to use, but the power will remain with Human employees.