Here Are 10 Terms In AI Technology That Need To Be Known

JAKARTA - The term artificial intelligence or AI, has been used in computer science since the 1950s. However, most people outside the technology industry only started talking about it at the end of 2022.

For that, Microsoft provides some keywords and explanations that need to be understood, so that we can better recognize the AI term and be part of global conversations.

Artificial intelligence (AI)

AI is a very intelligent computer system, which can mimic humans in several ways. Its nature is artificial because intelligence is made by humans using technology.

Microsoft explains that AI is not a physical machine or robot, but AI is a computer-running program, which works by including a huge data collection through algorithms, which is a series of instructions to create models that can automate tasks that usually require intelligence and human time.

Machine learning (ML)

If AI is the goal, then machine learning is how we can achieve that goal. Machine learning is a field of computer science under the AI umbrella, where humans teach computer systems how to do something, by training them to identify patterns and make predictions based on these patterns.

The data is run through algorithms over and over again, by providing different inputs and feedback at each time, to help the machine learn and improve performance during the training process.

Big language model (Lange language model / LLM)

Large language models, or LLMs, use machine learning techniques to help process languages, so that they can mimic the way humans communicate, developed based on neural networks or NNs, which are computational systems inspired by the human brain.

Models are trained using large amounts of text to study patterns and relationships in languages, to help models use human words. Their problem-solving ability can be used to translate languages, answer questions in the form of chatbots, summarize text, and even write stories, poetry, and computer code.

Generative AI (Generative AI)

Generative AI utilizes large management models to create new things, not just repeat or provide existing information. Generative AI learns patterns and structures, and then produces something similar but new. Generative AI can create things like images, music, text, videos, and codes.

Hallucination

AI's generating system can create stories, poetry and songs, but sometimes humans want the results of a genrative AI based on truth. Because an AI system cannot distinguish between real and fake, a generative AI can provide inaccurate responses. Well, this phenomenon is called a developer as a hallucinations, or a more accurate term, fabrication.

responsible AI (Responsible AI)

AI's responsive guides humans when trying to design a safe and fair system at every level, including machine learning models, software, user interfaces, as well as rules and restrictions imposed to access apps.

The Responsible AI practice is an important element because AI systems are often tasked with helping make important decisions concerning humans. However, because AI is made by humans and trained using data from imperfect worlds, AI can produce biases. Therefore, one of the keys to AI's Responsible practices is understanding the data used to train the system and looking for ways to overcome its drawbacks.

Multimodal model (multimodal models)

Multimodal models can work with various types or data modes simultaneously. It can view images, listen to sounds, and read words. In other words, the multimodal model is a true multitasker. This model can combine all information to perform tasks such as answering questions about images.

Prompts

Prompt is an instruction incorporated into the system using language, images, or codes to assign AI assignments. Engineers and all of us interacting with the AI system must carefully design a prompt to get the desired results.

Copilots

pilot is like a personal assistant who works with you on all kinds of digital apps, helping to do tasks such as writing, coding, summarizing, and looking for information. Copilot can also help you make decisions and understand a lot of data.

The recent development of LLM allows the presence of Copilot who understands human everyday language and provides answers, creates content, or takes action, while humans work on different computer programs.

It's what allows Copilot to interact with software and other services. This can help AI systems access new information, perform complicated mathematical calculations, or connect with other programs. How to make AI systems more sophisticated by connecting them throughout the digital world.