Search Engine for a Procedural Simulation of the Web with GPT-3

Web 4.0 could be the propitious evolution for the procedural web.

Goopt is an experiment in what the "procedural web" could be. This new web will use procedural content generation to create varied content, completely synthetic, since these are generated by algorithms and artificial intelligence. For now, the content is only text that is being generated automatically with GPT-3, the recent OpenAI model.

Goopt works as if it were a search engine, allowing us to search for any term, obtain related results and access their content. Simulating in this way the experience of browsing the web.

Goopt Demo

Procedural Web

The procedural web will be the future of the web. It will offer us infinite content, since it will not be necessary that someone has written or created it before. All the content will be synthetic and generated at the moment, with infinite possibilities. From informative text, articles, images, videos, to games, applications and services with interfaces and functionality that are automatically generated. All this adapted to our queries and needs, and increasingly personalized to our preferences.

Web 4.0 could be the propitious evolution for the procedural web. The automation that this new version of the web poses about applications, services, interfaces, APIs, devices and others, could be exploited by connecting them with the procedural web. These reality data interfaces would help empower and enhance their generative capabilities, as well as connect their functionality to the real world, having services and devices on which to execute actions. This makes the procedural web even more interesting, because it endows it with cybernetic capabilities.

The procedural web is based on natural language processing (NLP) and procedural content generation (PCG). Advances in these fields, as well as in the field of computational creativity, will allow us to generate increasingly better synthetic multimedia, thus nurturing the procedural web with more and better content. It will be interesting to see if this content comes to satisfy us more than the traditional web and human creation. Maybe one day we won't ever be able to distinguish.


Being an experimental project there is no public access online version, you have to replicate the project yourself as a developer and use your own OpenAI API Key.


Clone the project:

$ git clone
$ cd Goopt

Make a file called .env.local in the root of the project and write a line with your OpenAI API Key to it:


And finally install the project and its dependencies and start it:

$ yarn install
$ yarn start         # visit http://localhost:3000


Licensed under the MIT License.

Copyright (c) 2022 Kenny Cruz.


View Github