Embark on a journey into the dynamic world of AI development with our blog series, where we explore the latest and most innovative AI Developer Tools. As we delve into the tools and technologies shaping the future of artificial intelligence, discover how these resources empower developers to create intelligent, efficient, and scalable solutions. Whether you’re a seasoned AI professional or just stepping into the realm of machine learning, these articles aim to provide insights, tips, and practical guidance to navigate the diverse landscape of AI Developer Tools. Join us in unraveling the potential and staying at the forefront of the ever-evolving field of AI development.
What is AI going to do to the traditional practice of writing software code?
What we’re already finding is that AI can do more and more types of coding as well, or almost as well, as human beings.
Digital transformation concept. Binary code. Programming. Quantum computer.
It makes sense, because writing code is a technical process as opposed to one that is more intuitive, like writing poetry.

However, there is intuition in certain kinds of code, so some people believe that it will take longer for those tasks to be fully automated.

First, experts are talking about how AI coding will democratize the process of creating software. If AI can write the code, less tech-savvy people can direct it to write certain kinds of code.

For more, I asked a group of MIT undergrads in a class I am connected with at a recent class this question:

“Will AI be writing the majority of the world’s code in 2024? What year will AI be writing 95% of the code? 1-2 years, 3-5 years, 6-10 years , 10+ years or never? (or no strong opinion)?”
MIT Deep Learning Class

Here’s what they came up with:

About 30% cast the likely time frame as 3-5 years, saying that the vast majority of code will be written by artificial intelligence entities within that time frame.

A few more of our respondents were on the longer side, positing that this event will happen anywhere from 5 to 10 or more years in the future.

A couple of them declined to give a specific prediction, and others talked about the process in a more detailed way, one of our respondents suggesting that the process is really already underway.

“(It’s) already happening,” said an MIT student in our group. “I practically used Github copilot to write code for the entirety of a Web App, …(and) to create an IoT Extension for a RFID Security System. It is (going to be like using) a calculator for math.”

Another of our respondents used the terminology “no-code” to describe how this will work.
MIT Deep Learning Class at my Office
“AI will generate the majority of generic and functional code within next 5 years,” he said. “I believe 80-90% of regular application code will be generated by AI. 10-20% of code will be written by humans specifically for complex integrations, legacy technologies and systems that are not developed with standard programming. In 2024 , less than 30% of code will be created by AI – (that) is primary because corporations are not ready to allow AI to generate code due to compliance and risk factors.”

“I think AI will be writing the majority of the code, but not all, because I think codes related to security would not be exposed to AI,” wrote an experienced consultant responding to our question. “But for other aspects (of coding), AI can be used to change the world, just as computer has changed the world, and the way we approach knowledge, etc. People need to be more creative … and contemplative in relationships … rather than just problem-solving with patterns.”

That’s a little bit of a survey of what informed and involved people are thinking when it comes to AI taking the driver’s seat and creating software code.

Obviously, there will be big efficiencies – but also some potential drawbacks, as the technology wrestles with that creative or intuitive side of programming that has always been the province of the human brain.

source