Daily Times

Do you know that 1.25 million people are killed annually in traffic accidents? It is twice the number killed in war, crime and terrorism combined. More than 90 per cent of these accidents are caused by human errors. The reasons may differ. This number can be minimized to the lowest level. Self-driving vehicles will never do any of these things. Switching to autonomous vehicles is likely to save the lives of millions. All this is possible through Artificial Intelligence (AI).

AI has revolutionized the way of living, whether it is the healthcare system or finance, automobile or job sector etc. through this piece of writing. I have attempted to bring forth the basic AI tools for better understanding.

Artificial Intelligence (AI) has three main components: text, images and sound. Each of the components can be further divided and combined. The core of everything is the text because it is the closest to thinking. If you want to chat with an AI, ask it questions, create a relationship with it, or create realistic characters, the core is a text-based AI.

The future of this is the perfect personal assistant, coach, friend or partner for every one of us.

The GPT-3 is the most famous, from Open AI which is the most advanced Generative AI company today. the other advanced AI models like whisper DALL-E PaLM and LaMDA are among the popular.

Everyone is overestimating how humanlike AI and GPT-3 are. The GPT-3 is being dealt with and understood differently by different people according to the capacity of their thoughts. They somewhat label the GPT-3 as a decision maker or sometimes they mingle with that is has some parts of emotions too. By believing so they can be termed to be grievously misunderstood. For better understanding, it can be plainly said that GPT-3 synthesizes and rewrites to top pages, but it is not making decisions or having emotions. It is simply regurgitating what is in the data set i.e. books or articles etc.

A few years back, there was a frenzy of AI Chatbots but they remained unsuccessful because they were dumb due to the reason that the GPT-3 was not there. Now AI is getting good enough. Now you can play with GPT-3 today. It has dozens of uses it can do the following:

i) Movie to Emoji (Converts movie titles to emoji)

ii) Explain code (explain complicated code)

iii) Grammar Correction (Corrects sentences into standard English)

iv) Product name generator

v) Keywords (extract keywords from a block text)

vi) Q and A (Answer questions based on existing knowledge)

vii) Calculate time complexity

viii) Summarize difficult text into simpler

The aforementioned are a few examples. There are other examples of companies working in the space. They write text for you i.e. emails, ads, articles etc. The famous companies doing these include Jasper, Copy.ai, Writer, Anyword, Hypotenuse, Copysmith, Lex etc. Mem can write and rewrite text for you. Sudowrite is focused on literature and helps authors express themselves more poignantly. Summari summarizes the text. Otter and Fireflies transcribe the meeting on top of summarizing. Character.ai created a marketplace for bots.

It is not that hard to see where this is going. The future of this is the perfect personal assistant, coach, friend or partner for every one of us. Coding, Transcription and translation are also noticing other interesting use cases of text which have been tried and can be seen as just translating between a human language, can do for coding. All sorts of coding can be done with AI, from auto-completing code to asking it to explain a code you don’t understand. Co-Pilot and GIT Hub are the best helpers for AI-enhanced coding.

Open AI has an AI specifically focused on translation i.e. Whisper. It is (Whisper) not only good at translation between the languages but also amazing at transcription (from speech to text). Open AI’s whisper is by far the best transcription software which is usually 99+% accurate even with background noise or fast speakers. Whisper can understand speech much better than Siri or even humans whether there is background noise or not. it can also be translated virtually and into language.

You can create a PDF transcript of YouTube videos or create subtitles in other languages. It can record a meeting automatically, transcribe it and summarize it so as every meeting can be digested easily by anybody.

Text to images is the most famous one. You would be pretty surprised to know that an AI won the top prize in the painting competition. Indeed, the images that Ai-3 can generate are incredible. It is amazing how AI image is generated. It works by giving an AI a text prompt, and the AI creates an image that corresponds to that. DALL-E was the first AI to do this for the first time. Now new models have been introduced i.e Midjourney and Stable Diffusion, or Hugging face.

If you ask AI/Midjourney-3 through text to show you the image of Roman soldiers harvesting, it will show you the image as desired. Tools like Stable Diffusion and Midjourney allow you to go from text to image. Likewise, Lexica and Promptomania do this from image to text.

Besides this, one can experience the same from image to image. Companies like Snapchat have already been using AI to edit pictures. Image Enlarger is the tool which increases the definition of a picture. Remove is another tool which eliminates the background from the image. Some companies try to combine several of these tools, like ClipDrop which not only allows you to remove backgrounds but also persons and text change the light. It is mindblowing tech.

(To be continued)

The writer is a freelance columnist.