AI Prototype Shocks Developers By Turning Drawings Into Working Software
A new artificial intelligence (AI) prototype is wowing developers by turning drawings into working software. The "Make It Real" feature was released via a collaborative whiteboard app maker called "tldraw."
What is the AI feature about?
The feature allows users to draw an image of software and breathe life into it using AI. The feature uses OpenAI's GPT-4V API to visually interpret a vector drawing into a functioning web code (Tailwind CSS or JavaScript). In addition, the code is able to replicate user interfaces and even create simple versions of games like Breakout, ArsTechnica reported.
In a viral thread, designer Kevin Cannon explains how the feature was able to create functioning sliders that can rotate objects on screen, an interface for changing object colours, and a functional game of tic-tac-toe.
Additional demonstrations showed a clone of Breakout with a working dial clock that can tick, a snake game, a Pong game, and more. If you want to experiment with a live demo of Make It Real online, follow this link.
Also read:?ChatGPT's Getting Smarter: 4 New Useful Features Coming To OpenAI's Chatbot
How can you use it?
To be able to run it, though, you need an API key from OpenAI, which could pose a security risk if others intercept your API key. As ArsTechnica explained, your API key may be used to run the code locally and shoot up your bill.
Also read:?Newly Launched 'Humane AI Pin' Powered By GPT-4 Will Replace Your Smartphone
Who developed the tool?
Tldraw, developed by Steve Ruiz, is an open source collaborative whiteboard tool that has a basic infinite canvas for drawing, text, and media without needing a login.
It was launched in 2021 and has received $2.7 million in seed funding and is also supported by GitHub sponsors. After the recent launch of the GPT-4V API, Ruiz integrated a design prototype called "draw-a-ui" created by Sawyer Hood to insert AI functionality into tldraw.
How does GPT-4V come into play?
GPT-4V is a large language model (LLM) develop by OpenAI that can interpret visual images and use them as prompts. Make it Real works by "generating a base64 encoded PNG of the drawn components, then passing that to GPT-4 Vision" through a system prompt and instructions that turn the image into a file using Tailwind, as AI expert Simon Willison explained on X.
What do you think about this fascinating tool? Let us know in the comments below.?For more in the world of?technology?and?science, keep reading?Indiatimes.com?and?click here?for our how-to guides.??
