Microsoft's GitHub has used its high-tech annual convention in San Francisco to launch a new generative-AI tool called Spark as it aims for one billion users.
Already used by almost two million Australian open-source software developers, big-budget customers including some of Australia's leading banks have turned to GitHub to get a jump on competitors in the age of AI.
But GitHub CEO Thomas Dohmke said the vast majority of the world's population had been prevented from building software for too long.
"This can change with GitHub Spark, our new AI-native tool to build applications entirely in natural language," he said at Fort Mason in San Francisco.
GitHub CEO Thomas Dohmke said many people had been prevented from building software for too long. (HANDOUT/Supplied by PR Sefiani)
Usually, learning how to code involves knowledge of syntax and semi-colons that makes sense to a machine - and a narrow group of people.
However, the natural language aspect of generative AI means machines can be programmed to understand and respond to human language.
"With Spark, we will enable over one billion personal computer and mobile phone users to build and share their own micro apps directly on GitHub - the creator network for the Age of AI," he said.
Experienced developers can directly make changes to the underlying code while consumers or novice developers can work entirely in their own language, according to GitHub.
Once a user is happy with their "spark", they can automatically run it on their desktop computer, tablet, or mobile device.
They can share their sparks, with access controls, and allow others to remix their spark and build on their creations.
GitHub has also responded to calls for software developers to be able to build with the models that work best for them, as the rapid boom has spawned models that excel at different programming tasks.
Thomas Kurian, CEO at Google Cloud, said the new access to their Gemini models would mean millions of developers globally could benefit from "trusted, enterprise-grade AI".
Jared Kaplan, co-founder and chief scientist at Anthropic, said being on GitHub Copilot put the most advanced AI capabilities into developers' hands "wherever they're needed and wherever they work".
Microsoft bought GitHub six years ago in a $US7.5 billion deal to tap into the coding platform's community of open-source developers and bring them into the tech giant's orbit.
The company's analysis of the 5.2 billion contributions made on GitHub in the past year suggests AI is attracting and helping more people become developers.
There has been a surge in generative AI activity in India, Germany, Japan and Singapore.
Python is the most used language on GitHub as global open source activity continues to extend beyond traditional software development and as the language of AI takes root with developers.
"There is no one model to rule every scenario," Mr Dohmke said.
"It is clear the next phase of AI code generation will not only be defined by multi-model functionality, but by multi-model choice. Today, we deliver just that."
* AAP travelled with the assistance of GitHub.
12°C