Revolution on AI's Evolution

Revolution on AI's Evolution

turtle techies

Some cool stuff built with GPT-3

2020-09-01

Open AI

One of the best-known organizations in the AI field is Open AI.
Open AI’s mission is to ensure that artificial general intelligence AGI - which is highly autonomous systems that outperform humans at most economically valuable work - benefits all of humanity.
OpenAI consists of the for-profit corporation OpenAI LP and its parent company, the non-profit OpenAI Inc.
The organization was founded in San Francisco in late 2015 by Elon Musk, Sam Altman, and others, who collectively pledged US$1 billion.
You can read more about OpenAI at https://en.wikipedia.org/wiki/OpenAI .

GPT-3 (Generative Pre-trained Transformer 3)

The group of 31 engineers and researchers working at OpenAI introduced the project on May 28, 2020.
The full version of GPT3 has 175 billion parameters to process data.
This figure is twice the learning capacity of GPT2.
In the beta phase as of July 2020, GPT3 uses the natural language processing NLP system with pre-taught language samples.
Before the launch of the GPT3, the largest language model was the Turing NLG, which Microsoft introduced in February 2020 and has less than %10 capacity 17 billion parameters) of the GPT3.

Since GPT3 can produce news that are difficult to distinguish from articles written by humans, it is thought that GPT3 can be used in the development of both useful and harmful applications.
The researchers detailed the potentially harmful effects of GPT3 in their article dated May 28, 2020.
Misinformation production, spam, phishing, abuse of legal and government processes, fake academic essay writing, and social engineering are some of the potential harmful effects.
The authors highlight these dangers for research into reducing these risks.

What can be done with GPT-3?

Well, pretty much anything you can imagine.
However most examples are in the area of text creation and suggestion based tasks.

Even though it has been only a month since it’s release, beta testers found so many impressive things to do with GPT3.
Let’s see what people came up with.

Smart Ingredient Checker

Everybody wants to live a healthy life. Here is an example of GPT3 extracting ingredients of a product with computer vision.

Intellisense on Steroids

If you are a software developer, I bet you like Intellisense right? Guess what, completion tools are much more powerful with AI.

HTML Layout Generator

GPT3 is not only useful for backend development but for frontend development as well. It can design a layout according to a description in plain English.

Creating App Design From a description

GPT-3 can design a mobile app layout from a definition. It can also be told to check other applications and create the design to look similar to it.

Smart E-Mail Assistant

Answering e-mails can becomes a real pain. This app automatically write your emails for you in your own style.

Generating SQL Code

GPT-3 for lazy database administrators.

An Interesting Coding Interview

I wouldn’t hire it, but it sure is impressive.

Turtle Techies T shirt
Get your Turtle Techies swag