We have spent a lot of time talking about AI and its impact on programming over the past year, but the new year will confirm that it's a game changer or just another passing fad. It is the one big issue of 2025.

This isn't the first time I have considered the suggested demise of my job. There was a time when it was thought that code generators and expert systems like "The Last One" promised or threatened to automate code to the point where human programmers wouldn't have to be particularly skilled or even be needed at all. We all know how that turned out. The Last One was far from being the last one and sank without trace.

After struggling with AI for a long time, we have finally reached the point where it might be good for something very general. It is now a widely held belief that it is good enough to write programs or at the very least it might be good enough to help make code generation more efficient. So much so that you hear claims that programming is a dying occupation and whatever you do don't take it up as it has no future.

At a lesser extreme you will also hear claims that even if programmers are going to be needed for a while longer, we will need fewer of them and certainly no "grunt" programmers at the lower-levels of job competency.

If you scan the news and opinion pieces, you will find that opinions really do differ and in some interesting ways. Some programmers and non-programmers claim that LLMs make programming easier. They use them to generate code to get the job done or they use them to perform a code review and spot errors very early in the process. Others say that, while this is possible, the LLMs simply waste time by providing code that has to be checked and get you chasing bugs that aren't really bugs in the wider system context.

The differences of opinion seem to be stark and some of them  could be put down to using different LLMs. Sometimes it is suggested that it is all a matter of asking the right questions and that prompt engineering might well be what replaces programming. It also could be that the differences are down to what the user of the LLM actually wants to be true.

These are the opinions of the normal mortals amongst us. When we turn to the big bosses then the opinions are far less split. , Sundar Pichai recently said that Google relies on Artificial Intelligence to generate a quarter of its new code. This is an astounding claim, but then it isn't clear what it means. It almost certainly doesn't mean that 25% of Google's new code is never seen by a human.

Microsoft boss Satya Nadella isn't quite claiming that AI is revolutionizing Microsoft, but he does state that:  

“The era of traditional software is coming to an end. We’re moving toward a world where intelligent agents handle complex tasks, enabling people to focus on creativity and strategic thinking.”

This is a slightly different point of view, but just as damaging to programming. He seems to be arguing that we won't need  programs and hence programmers in the future, because AI will do what we ask and so replace all of the apps we currently use to do the same tasks. This isn't so much the death of programming as the invention of the universal app that just does what the user asks.

Whatever you think about AI and programming it is fairly certain that this year, 2025, is crunch time. Either things will go on much as before, with us using AI as a programming aid or not as the mood takes us, or we will be in a brave new world, with AI doing most of the work and junior programmers an extinct species.

• Mike James, Founder and Chief Editor of I Programmer is also a prolific author. In addition to his books on specific languages, he is also the author of The Trick Of The Mind: Programming and Computational Thought, aimed at programmers and non-programmers alike, which examines the nature of programming and reveals why it is a very special skill.

More Information

Microsoft’s Satya Nadella Is Betting Everything on AI

Related Articles

OpenAI Enriched By Stack Overflow

Copilot Chat Improves Confidence and Enjoyment

GitHub Copilot Provides Productivity Boost

The Impact of Copilot Chat

AI Helps Generate Buggy Code

Developers Like Code Assistants Even When They Are Incorrect

It Matters What Language AI Thinks In

Copilot Improves Code Quality

Prompt Engineering Techniques To Make You An Expert

Peter Norvig - As We May Program

Magic Prompts For LLMs?

Tell A Chatbot "Take a deep breath ..." For Better Answers

Runaway Success Of ChatGPT

Microsoft Goes All Out On Generative AI

To be informed about new articles on I Programmer, sign up for our weekly newsletter, subscribe to the RSS feed and follow us on Twitter, Facebook or Linkedin.

Comments

Make a Comment or View Existing Comments Using Disqus

or email your comment to: comments@i-programmer.info

<ASIN:1871962722>