GPT-3 is used to create computer code from scratch. It’s not yet available for free to the general public, but it may never be. Microsoft has already released Microsoft Power Fx, which merges natural language with the formula language used in Excel. Technology is already powering some useful tools – and they’re probably just the start of the process. So far, the technology that makes it work is the open-source result of their work, which is the result of a group banding together to try and replicate it.
In 2019, AI research and development company OpenAI completed work on GPT-2, its second AI natural language generator. At the time, the company refused to release it, saying there were too many ways it could be misused.
But before the end of the year, they had changed their minds. What followed was a torrent of wacky – often hilarious – uses of the technology, mostly as a tool for amusement. It seemed that OpenAI’s worst fears about its release had not come true.
Things went so well that OpenAI moved on to its next version, aptly named GPT-3, which is said to be even more powerful than its predecessor. And while it’s not yet freely available to the general public (and it may never be), there are already some great examples of the types of text it can create.
It turns out, however, that GPT-3 is good for some other important tasks. One of them is creating computer code from scratch. Here’s a look at some of the ways it is (and will be) used to improve the lives of programmers around the world.
Enable Low-Code Development for Everyone
In September 2020, OpenAI announced that it had sold exclusive rights to the technology behind GPT-3 to global software giant Microsoft (this is one of the reasons it may never see a free public version ). And hence, the software giant has taken a step ahead of everyone else in the use of new technology. And it didn’t take long for them to do so.
In May 2021, they announced the availability of Microsoft Power Fx. It is a low-code programming language that merges the natural language processing capabilities of GPT-3 with the pre-existing formula language built into programs like Excel for decades. The idea is to allow users already familiar with this formula language to use it to create custom interfaces without learning anything new.
The concept is at the heart of a platform called the Microsoft Power Apps canvas. It is an ecosystem of products designed to enable enterprises to create custom applications that interact with and manipulate the data they have stored in the cloud. Microsoft envisions the platform as a one-stop-shop for businesses to build low-code apps that eliminate the need to hire outside developers or send employees to programming courses.
A function auto-completion utility
Another easy way that GPT-3 already helps programmers is to give them code suggestions as they work. The GitHub Copilot is a great example of this idea in action. Its users receive helpful code suggestions as they work, which the AI engine pulls from GitHub’s vast library of existing code.
Additionally, Copilot can also allow a programmer to preview multiple approaches to the same function. This opens up the possibility for programmers to use Copilot as a learning tool – something akin to asking for coding help from an assignment expert – but without waiting. And because it can understand and work in multiple programming languages, Copilot can help programmers transfer their work into unfamiliar languages in no time.
Creation of complete web applications thanks to NLP
Although few people or companies have yet had access to GPT-3, some have already created products that use its new features. One of them, called Debuild, is testing a system it claims can build fully functional web applications with no input other than a plain English description of what you want.
The implications of such a tool are enormous. If it’s even a marginal success, it would be nothing short of groundbreaking. Companies could use it to generate custom software on demand. And web developers could use it to create quick, working prototypes, freeing them up to focus on higher-level work.
Even though GPT-3 was originally designed to understand and process natural human speech, it may have found its true calling in the world of computer programming. As you can see, it already powers some useful tools – and they’re probably just the start. But the biggest obstacle preventing even more progress is the fact that OpenAI has so far refused to make the technology available for public use.
But that may not matter for long. A group of AI researchers have banded together to try and replicate the technology that powers GPT-3. The result of their work (so far) is the open-source GPT-J, which has similar capabilities to the original. And according to many people who already use it to tackle programming tasks, it outperforms the original in several ways.
This means that it may not be long before the examples detailed here are joined by countless other amazing and useful programming tools. And because they’ll be built on open-source technology, they’ll likely be available to the public to use and modify as they see fit. So keep an eye out, because there are going to be some amazing new developments to come.