I've been curious to learn more about ChatGPT
Question from Prerna - what is AI and how does someone create that?
As usual, I did not give a very concise or clear answer so one of the goals of this post is to help me to improve my answer. My working answer to this is to explain something about how people learn new things; you start with a goal and measure the outcome that you want. For example, I'm hungry and I want something salty. I would eat things and get some subjective level of satisfaction that I like food A versus food B. We don't verbalize in language exactly why we like A v. B but we feel that we A was better in some gestalt way of comparison.
AI is similar to that in which you create a set of goals and let machines learn how the reference set of judges view the desired outcome of a good solution to the problem (from a point of view). Then you load a data set to let the system train itself how "people" take the input data and how they value the output against the reference goals.
The system then does this over a massive number of iterations to build its own version of taking input to create a version of output that optimizes satisfaction against the measure of good v. bad solution.
Artificial General Intelligence - (from What We Owe The Future - MacAskill)
Artificial Intelligence (AI) is a branch of computer science that aims to design machines that can mimic or replicate human intelligence. Because of the success of machine learning as a paradigm, we've made enormous progress in AI over the last ten years. Machine Learning is a method of creating useful algorithms that does not require explicitly programming them; instead, it relies on learning from data, such as images, the results of computer games, or patterns of mouse clicks.
…
The ultimate achievement of AI research would be to create artificial general intelligence, or AGI: a single system, or collection of systems working together that is capable of learning as wide an array of tasks as human beings can and performing them to at least the same level as human beings. Once we develop AGI, we will have created artificial agents - beings (not necessarily conscious) that are capable of forming plans and executing them in just the way that human beings can. An AGI could learn not only to play board games but also to drive, to have conversations, to do mathematics, and countless other tasks.
How has it been coming up in channels -
- What we owe the future, AGI to drive values autonomously and over generations of human lifetimes
- The idea of sock puppets for influencing the masses on the internet in the form of social media, comments, Reddit, etc...
- the low-tech example from fiction from Season 6 of Homeland that a private agency was working underground to create social media manipulation to control a massive number of fake accounts in order to push a political agenda
- Hearing things about ChapGPT in the news that
- Thinking about how to invest in the megatrend of AGI

Comments