The MIT researchers created a computer program that can predict human emotions.

The new computational model successfully predicts human emotions. That kind of system might be prohibited in the EU if the new AI limitation or AI-control laws are ratified. The systems that predict human emotions are one of the most risky things in the world. Sometimes people ask what kinds of risks are hiding in the systems that can predict or follow what humans are feeling. 

The AI that can uncover feelings can use the computer's web camera. So if somebody uses that kind of tool during a web meeting. That thing can uncover things like if somebody on the table is somehow stressed. That AI-based software can also find out if somebody is telling lies. And this is one of the things that breaks privacy. When we think about this kind of software there should be a blinking warning that this kind of algorithm is in use. 

"MIT neuroscientists have created a computational model that successfully predicts human emotions in social scenarios, using the prisoner’s dilemma game as a base. The model considers individuals’ desires, expectations, and the influence of observers, deducing motivations, comparing outcomes with expectations, and predicting emotions based on these factors. This model, mimicking human social intelligence, outperformed other emotion prediction models, and researchers aim to adapt it for broader applications". (ScitechDaily.com/Decoding Emotional Intelligence: MIT’s Computational Model Excels in Predicting Emotions)

"While a great deal of research has gone into training computer models to infer someone’s emotional state based on their facial expression, that is not the most important aspect of human emotional intelligence, says MIT Professor Rebecca Saxe. Much more important is the ability to predict someone’s emotional response to events before they occur. Credit: Christine Daniloff, MIT". (ScitechDaily.com/Decoding Emotional Intelligence: MIT’s Computational Model Excels in Predicting Emotions)

The model-making base is in machine learning. The system can learn how people react to things while they are feeling something. So the system might follow the movements on the face or changes in volume and way how person talks. 

Or those algorithms should be reserved only for official use. The thing is that those algorithms that predict emotions are suitable for things like employment interviews. This thing can use to make profiles about employees. But the AI-based algorithms that can follow emotions can use in marketing tools that are using feelings to make people buy things. 

If that kind of AI is connected with marketing pages, it can follow the emotions that people feel, when they are looking at the homepages. Then the AI can adjust images and colors including the sound that people feel more comfortable. 

This is a new kind of tool for emotional marketing that is prohibited in the EU. The use of emotions as a marketing tool is one of the things that are causing situations in which some people spend all their money on new clothes and other things.

https://scitechdaily.com/decoding-emotional-intelligence-mits-computational-model-excels-in-predicting-emotions/