Responsive and Responsible AI
This article has been moved to its new home here: https://benperk.github.io/msdn/2017/2017-01-responsive-and-responsible-ai.html
#artificialintelligence #AI #artificialGeneralIntelligence #AGI
I was watching this video where the CEO of Microsoft Satya Nadella was talking about AI.
He touched on a few interesting points and I really love the way he articulates his thoughts and the choice of his words.
He mentioned we need to teach machines to learn so they can do things humans can do that can in turn help humans and that AI should be everywhere and not just owned and controled by a few large companies. He went on to mention a term I had never heard of, which is ‘artificial general intelligence’ looking it up on Wikipedia it describes that as;
“Artificial general intelligence is the intelligence of a machine that could successfully perform any intellectual task that a human being can. It is a primary goal of artificial intelligence research and a common topic in science fiction and futurism. Artificial general intelligence is also referred to as "strong AI", "full AI" or as the ability of a machine to perform "general intelligent action"
As I have been writting many articlce about #chatbot, #LUIS and other #CognitiveServices here, I started thinking about or had a reinforced feeling that I was moving along the right path. Reading many articles about the new #chatbots #chatbot and what they are doing, I see mostly that they are nothing but a search engine that answers a query, there is nothing more than simply a database that is searched with keywords and response with some information. This is not what a #chatbot is about, IMO. My #chatbot is designed and does use AI and has a personality.
Other comments in the video were in regards to the reduction of jobs or the removal of jobs from humans as the AI and technology get better at doing the work done by humans now. For example, what I just said above, creating a #chatbot that answers questions that have ‘known answers’. Statements like ‘there is no growth, we need technological breakthroughs that drive growth beyond us and that societies must help people get skilled for the jobs of the future.’ Its the ‘although some human jobs may get replaced by AI, new ones will certainlly be created to replace them. The mentions ‘technological breakthroughs’ must happen in order to create those new jobs.
Lastly, one other comments I found interesting were ‘’'design decisions, tasteful AI, create more trust in technology, security and privacy, how to open the block box for inspection’ and ‘before the ethics, get focus on thing we can control, make design decision’. I remember reading about some code in AI that was written about the dilemma in self driving cars. Take for example a situation came up where a decision needed to be made that resulted in the driving getting hurt or a group of individuals standing innocently on the sidewalk? I was of the opinion that the AI needs to be designed in a way that a situation like that would never come up at all, but that has lots of ifs. Which car would you buy if you knew the AI in the car would decide to hurt you versus the others?
Regardless of that, before we get to those ethical questions, we need to make sure the AI we create is secure, provide privacy and is open to all for the usage into thier own application and personal life, if desired. This is cool stuff.