Decisions by AI.

Artificial Intelligence has been growing exponentially in recent years and the introduction of ChatGPT has just fueled it. Companies are racing to cut the jobs that could be automated by an AI or a person who can use AI efficiently.

We have been watching movies like Terminator 3 and ex-machina and putting forward the theories of AI taking over humanity. We are still asking that question and we have built a deeper understanding of AI, or are we just overestimating it?

If we go under the technical side of AI, we will only find a model taking in the 1 million decisions by humans, extracting a pattern, and using it for making other decisions. This learning behavior and pattern recognition is the basis of developments in AI and its applications. So, the understanding of decisions by AI starts here.

AI bases and defines its decisions on history.

Artificial Intelligence works on data that is collected in the past for making decisions in the present and forecasting a certain situation or pattern. This is where small doubts start arising for the rationale behind data points and the base of decisions (for a generic model), what kind of perspective the decisions are taking, and what the pattern will generate as per the quality and quantity of certain decisions. Do we look at our past patterns for making decisions in our daily lives? How often do you consider history, and how often do you just take the risk of not considering past decisions?

Critical thinking, awareness, empathy.

All three of these qualities are inserted as the function of specific prompts as they are observed to be present in human nature by default. Additionally, you might like to notice that a particular prompt assigned to intelligence is programmed to complete the assigned work, no matter what kind of decisions it faces. It is only through these decisions that we put stages of correction, molding of prompts, and insertion of elements in an intelligence that is then recorded in its database. An AI programmed to bring you a cup of coffee would do anything to get you one until you program it to ask every time a diversion is observed. So, having points of pause in the completion of a task reprograms the machine for options that again program it to adapt according to the needs and availability.

Who is accountable for the decisions made by AI

When we talk about the machines talking to us, we simply mean they are using a set of data to find what and how to speak in an understandable language. Where do we find this data?

The data comes from a lot of sources depending on the type of programming we have to provide to the machine. We use language data from research papers, interviews, etc., sorted and managed by language professionals like interpreters and terminologists professionals for developing the models. It is imperative to have the data right because this is the core intelligence that serves as the basis of all the decisions that the machine will make.

What do future developments hold?

There is a very high chance that an ethical authority will be established for tracing the developments in AI. There are many research labs already working on AI systems that automate many technical work and eliminate the need for human intervention and reduce error. We can look at this with both positive and negative perspective. If the bar for entry into the human workforce is increased, human minds will be employed for work that actually needs entrepreneurial nature and skills. The increased efficiency and reduced human errors make the work much optimized and faster. On the other side, the loss of jobs and the suffering population will bear the dawn of AI.

The technology powering AI.

All the technology in AI comes under machine learning and deep learning. Machine learning uses data and algorithms to learn about a behavior or concept, quite like humans and hence improving the accuracy of the model over time. Deep learning is simply a large neural network that is usually three or more. I have described these technologies in details in other articles in case you wanna have more insights into the AI tech :)

Where are we?

We are just getting started with the systems surrounding AI, and we still are waiting for a series of joint adoption and good data. Are we experience yet another dot com like bubble or is this just the time of AI revolution.

Would you like to learn something amazing every week?

Just Subscribe