You may not have thought much about computers being able to understand what is good or bad — a moral concept. But researchers from the Technical University of Darmstadt recently found that a computer system can learn what is described as good and evil. The results could help with sentiment analysis, which seeks to decipher facts, opinions, and feelings around text and video content.
Why Natural Language Processing?
Natural Language Processing (NLP) is a subfield within Artificial Intelligence systems. It aims at the computational understanding of human language. The ultimate goal of NLP is to make computers understand and process human languages as humans do.
To understand why NLP is important, you must consider how computing has evolved. It started with binary symbols that carried no meaning for us. Then came the time for text, which machines read like a bunch of meaningless letters. Nowadays, computers can understand the text as a meaningful sequence of words and sentences, but they cannot still understand their meaning. That is where NLP comes into play. And that is precisely what will help artificial intelligence systems learn human concepts of ‘good’ and ‘bad’ in a way that makes sense for us humans.
The Moral Dimension in AI-Language
There is a moral dimension contained in the artificial language model. It is not just about the language but also about the worldview that undergirds it. Also, it is not just about individual words but about how these words are put together to form an ideological framework that guides our thinking and even our behavior. These models are pre-trained on a large corpus of text using a self-supervised learning framework. Consequently, they learn the distribution of the text and any patterns it might contain. The patterns often include biases that are present in the training data.
What Is Next for Artificial Intelligence Systems?
The market for AI language interpreters is steadily growing, and with continuous efforts of the AI industry, it is expected to grow more in the coming years. With some applications already on the market, Artificial Intelligence systems the language interpreting systems will present themselves as an efficient solution for most common problems faced in real-time conversation.
