The most popular and comprehensive Open Source ECM platform
Let us know more
Artificial Intelligence: Is AI-Authored Code Any Good?
Researchers have taught AI to write software. OpenAI’s Codex has been the first AI model which has exceled in being able to write software based on a natural language prompt or problem description. Codex is able to generate code in 12 different software languages and can translate between different software languages.
Codex isn’t the only option. Google has project called Bayou. Microsoft has another called DeepCoder. Is AI coming after softwar developer jobs? Maybe just not yet. It’s enough to have developers pretty worried though. A survey by Evans Data Corporation found that the top worry of developers is being replaced by AI.
But software requires interpretation of user requirements. That’s something much harder for AI to capture.
Artur Hebda, developer at Railsware, said that “creating software is primarily about converting vague requirements into a strict specification that can be run deterministically. And that’s not easy even for humans.”
Tom Smith, software engineer at Gado Images, told the New York Times that “we thought these tools were going to completely remove the need for humans, but what we learned after many years was that this wasn’t really possible — you still needed a skilled human to review the output. The technology gets things wrong. And it can be biased. You still need a person to review what it has done and decide what is good and what is not.”
Greg Brockman, the chief technology officer of OpenAI, said that “AI is not playing out like anyone expected. It felt like it was going to do this job and that job, and everyone was trying to figure out which one would go first. Instead, it is replacing no jobs. But it is taking away the drudge work from all of them at once.”
So, is AI-Authored code any good. Yes, but it will only take you so far.
