How AI-Powered programming may revolutionize all jobs in the future?
In recent years, the development of AI-powered coding assistants has garnered much attention and hype. These tools, which use advanced algorithms and natural language processing to assist programmers in writing code, have the potential to revolutionize the way we approach programming. In this article, I will present my opinion on the impact of these technologies on the future of jobs.
Text-based models and LLM large language models have been the real success story in AI for the last ten years, arguably way ahead of computer vision and other machine learning paradigms. This has led to the rise of AI-based programming tools like ChatGPT and Copilot. These tools, although impressive, still have a long way to go before completely replacing competent human developers.
What CoPilot and ChatGPT are good at?
One of the core strengths of these systems is identifying common computer science paradigms and patterns used in algorithms and applying these typical recipes/patterns to your own code. Although these models essentially predict the next word given the context distribution but somehow these implicitly are excellent in identifying the core structure of the code being written. Hence coding assistants are excellent at common problems such as writing HTML landing pages, writing machine learning everyday tasks such as data ingestion, pandas data wrangling, standard and generic React components, etc., but struggle in more nuanced subjects such as developing software architectures, writing complex classes and relationships based on business use cases.
Sub-Optimal Solutions?
Another concern is that these tools may not be able to provide the same level of quality and reliability as solutions offered by a diverse community of programmers. At StackOverflow, for example, the code provided by users is typically reviewed and verified by multiple people, which helps to ensure its accuracy and reliability. In contrast, solutions provided by AI-powered assistants may not be subject to the same level of scrutiny and oversight.
Impact on non-tech fields?
Looking at these tools in a historical context, these tools are simply following the overall s-curve trend in programming technology of increasing abstraction and automation. The impact of these tools on programmers will be a lot like the impact of spreadsheet tools in the 1980s on business and finance professionals, as these programs revolutionized the industry. Although far from perfect, Spreadsheet software made finance more accessible, cheaper, and easier. For example, businesses didn't need to hire a professional accountant to do trivial calculations, these could be done by an intern with spreadsheet training. Also, it made changing the calculation logic to changing business requirements much more manageable, professionals needed to just replug values, and the software will take care of internal cell calculations and produce the changed results. In fact, companies like Microsoft build their early successes on spreadsheet software. Hence similarly, AI-powered tools may take over non-tech industries.
End of Grunt/trival Programming work, impact on programming jobs
The first major impact in the short term of the use of AI-powered will lead to the displacement of thousands of programmers who currently perform “grunt” programming work. As these tools become more advanced and widespread, they may be able to automate many of the tasks that human programmers currently perform.
Rise of Citizen Data scientist and non-traditional programming
These tools will likely get integrated into many of the no-code/low-code data science tools currently available. These will greatly democratize software development and machine learning and eventually lead to greater penetration of machine learning solutions into other traditional non-tech domains. For example, people who do not have traditional computer science knowledge can now use state-of-art ML tools integrated into their familiar software like e.g., Ms. Excel and CRM, at no additional cost. Hence solutions like complex visual basic scripts in Excel may be redudant
Another Wall Street Moment?
Based on my earlier historical example of a similar trend, it is often said that the rise of spreadsheet software unleashed Wall St. on the world. Without the ability to test hundreds of scenarios and superior data-cruching, complex kinds of trading and investments/derivaties will not have been possible. The interaction of non-tech fields such as finance/investment with hard data crunching led to this brilliant marriage. It is impossible to predict the impact of these tools on the larger industry, but it has the potential to be world-changing, just like spreadsheets. For programmers, in the short-to-mid-term future, given the drawbacks of these tools, companies will opt for human developers to write code, although the nature of their jobs might change.
