Here are some words that any AI enthusiast should be familiar with. I see that people often confuse about the definitions of some of the words. So, taking some time to gather community-accepted definitions of these words.
- AGI (Artificial General Intelligence): This is a field of AI that tries to develop systems/models with ability to do a broad range of general tasks that a human can do, including learning, understanding, reasoning and applying the knowledge. AGI can perform any intellectual task that an average human can do.
- ANI (Artificial Narrow Intelligence): This is a field of AI that develops AI systems that are trained to do a specific task, for example, face recognition, playing Chess etc. ANI operates within the narrow domain they were designed for. They can outperform humans in their specific tasks. Examples including AlphaGo chess agent, Siri, Alexa, Self-driving cars etc.
- ASI (Artificial Super Intelligence): This refers to a level of AI that surpasses human intelligence across all fields, including creativity, general wisdom and problem-solving. So, ASI would be capable of surpassing the best human brains in every field, improving itself autonomously. This can lead to rapid advancements beyond human control or comprehension. Currently ASI is purely theoretical as this stage. It is often discussed in the context of future AI, ethics and potential existential risks to humanity.
- GenAI (Generative AI): A set of AI models that can generate new content (data or information) such as text, images, videos, music, voice, code and more. The generated content is similar but not identical to the data on which the model was trained. Examples of GenAI tool include ChatGPT, DALL-E and Midjourney etc.
- LLMs (Large Language Models): A set of language processing models that are trained on massive amounts of text datasets, hence the name “large”. They can understand and generate the text data.
- GPT models (Generative Pretrained Transformers): A series of LLMs developed by OpenAI. These models are trained to understand and generate text based on the input they receive. These inputs are…