The Future of AI in Business: A Balanced Approach

Over the years, various fads have come and gone within the tech industry, each promising revolutionary changes but often falling short. In this article, I will explore these trends and their implications for AI in business.

COBOL (Common Business Oriented Language - Early 1960s)

COBOL was designed to be a high-level language for business applications, making programming more accessible. Initially it was seen as a revolutionary tool that would make it easier for non-programmers to write business applications. However, while it did simplify some aspects of coding, its complexity and inflexibility ultimately limited its widespread adoption outside specific legacy environments.

4GL (Fourth-Generation Language - Late 1970s - Early 1980s)

Fourth-generation languages aimed to make programming even more accessible by allowing users to write programs using natural language or simple commands. The promise of 4GL was that it would greatly reduce the need for skilled programmers. However, these systems often lacked the flexibility and power needed for complex applications, leading them to be seen as specialized tools rather than a replacement for traditional programming.

Visual Basic (VB - Early 1990s)

Visual Basic was introduced by Microsoft and aimed to make Windows application development more accessible through its graphical interface. VB did democratize some aspects of software development, making it easier for non-technical users to build simple applications. However, its limitations in handling complex business logic and performance issues meant that many professional developers were still needed to architect non-trivial business solutions.

CASE Tools (Computer-Aided Software Engineering - Late 1980s - Early 1990s)

CASE tools were designed to automate various stages of the software development process, from design and implementation to testing. The idea was that these tools would reduce the need for manual coding. However, while they did improve productivity in some areas, they often required significant customization and could not fully replace the creative and problem-solving aspects of programming.

RAD (Rapid Application Development) - Late 1980s - Early 1990s

RAD aimed to speed up software development by using a structured approach that focused on iterative development. The promise of RAD was to reduce the time and effort required for developing applications. However, it often failed to address the complexity of large-scale projects or systems with stringent requirements, leading to limitations in its widespread adoption.

Every time companies and investors flocked to the latest tool or technique to end skilled technical labor, they fell short of the goal, and in many cases expanded the profession. I don't expect things to remain the same in the advent of accessible AI - but I do feel the end is not nigh.