The Tipping Point: How AI Will Cut Away 60 Years of Dead Wood in Programming and Software Engineering

I started full-time as a scientific programmer (Fortran, ALGOL) in 1967 in the "Computation Laboratory" of the Statistics Department of the University of UNSW run by Assoc/Prof Jim Douglas.

Since then, for nearly 60 years, I've watched as the software industry has built layer upon layer of processes, methodologies, and structures to address a fundamental shortcoming: speed.

The industry has come to believe that these structures are intrinsic to the process, whereas, in reality, they represent the failure of programming and software engineering technology.

The principles remain much as Donald Knuth and Edsger Dijkstra articulated in the 60s. Marginal advances such as OOL and higher levels of abstraction have helped, barely.

The fact is that, from the 1960s to today, the pace at which code is produced, tested, and delivered has dictated how we manage projects, coordinate teams, and bridge the gap between developers and end-users.

Despite the many technological advances in recent decades, this slowness has persisted, and the industry has responded by erecting a superstructure of frameworks, management techniques, and workflows to cope with the delays.

But now, a genuinely disruptive force is emerging: AI-enabled programming. This technology doesn't just promise another incremental increase in productivity—it’s poised to eliminate the very bottleneck that has shaped software engineering from its inception. As AI lifts the speed barrier, it’s about to expose the massive “dead wood” that has accumulated around the development process, forcing the industry to shed decades of outdated practices.

The Root Cause: A Speed Deficit Spanning Decades

To understand why this transformation is so profound, we must recognize that speed, or rather the lack of it, has always been the primary obstacle in software development.

Despite advances in programming languages, tools, and methodologies, the time between requirements specification and the delivery of functioning software has always been lengthy. These delays have spawned countless project management frameworks—AGILE, SCRUM, Waterfall, and others.

What if those inefficiencies were eradicated?

What if the delays between specifying what you want and seeing it materialise were reduced from weeks or months to mere hours or even minutes?

This is precisely the promise of AI-driven code generation. It’s not just about writing code faster; it’s about transforming the entire development lifecycle into a real-time, dynamic process that aligns perfectly with user expectations.

The Death of the Project Management Superstructure

The painstakingly built superstructure that the software industry relies on today is about to be made irrelevant by AI’s capacity for speed.

Project managers, AGILE coaches, sprint planning sessions, retrospectives, and the myriad processes designed to mitigate delays in software development will become increasingly obsolete. Why? Because they were always coping mechanisms for an underlying problem that AI is now solving.

AI doesn’t just generate code; it does so in a way that facilitates instant feedback loops. This ability to respond to user requirements in real-time shifts the emphasis from planning and process to creation and iteration. Imagine a scenario where an end user can articulate their needs in natural language, and within moments, they’re presented with a functional prototype to interact with. The AI refines the software based on immediate feedback, creating an environment of continuous, on-the-spot evolution.

As this transformation takes hold, the industry will witness the collapse of the management structures that were designed to bridge the chasm between slow developers and impatient users.

CEOs must recognise that the tipping point is here, and any delay in adapting to this new reality will mean falling behind the competition.

The Enormous Burden of the Current Paradigm

The burden of slowness extends far beyond coding inefficiencies. It permeates every aspect of project management, communication, and collaboration.

Entire industries have been built around the problem of slow software delivery—consultancies, certification programs, productivity tools, and training seminars—all aiming to optimise processes that are fundamentally flawed because they are anchored in a paradigm of delay.

The introduction of AI-enabled programming reveals these layers of bureaucracy for what they are: unnecessary impediments.

As AI accelerates the journey from idea to implementation, the need for all these extraneous roles, processes, and frameworks evaporates. The software development culture of tomorrow will be one of instantaneous creation and feedback, where agility isn't something achieved through meetings and ceremonies but through the raw speed of execution.

This shift is not a question of “if” but “when,” and the “when” is now.

Democratising Software Creation: A New Balance of Power

The rise of AI in programming will do more than eliminate inefficiencies—it will also democratise the creation process. For the first time, those without specialised technical knowledge will be able to drive software development, just as those without deep data analytics skills are able to perform competent analyses using ChatGPT today.

Imagine business leaders, designers, or end-users themselves interacting with an AI to build and refine solutions directly, without the need for translation through layers of technical experts.

This shift fundamentally challenges the long-standing belief that only those with coding expertise can bring ideas to life. As AI takes over the heavy lifting of code generation, the value shifts to those who can articulate problems, ask insightful questions, and provide meaningful feedback.

In this new paradigm, anyone with domain knowledge and a clear vision can be a creator, reshaping the power dynamics of the industry.

CEOs must be ready to embrace this change, recognising that the future of their businesses will depend not just on employing skilled developers but on empowering all team members to contribute to software creation.

Navigating the Grief Cycle of Change

Inevitably, this transformation will be met with resistance, skepticism, and even fear. The industry is about to enter a grief cycle as it grapples with the impending obsolescence of practices and roles that have defined software engineering for decades:

• Denial: Many will insist that AI is just another tool, not a genuine disruptor. They’ll cling to AGILE practices and traditional project management, convinced that these systems will endure.
• Anger: As AI begins to prove its capabilities, expect pushback from those whose livelihoods and expertise are tied to the old ways. Project managers, consultants, and even developers may resist this new reality.
• Bargaining: There will be attempts to integrate AI into existing processes, to find a middle ground where old methodologies coexist with new technologies. These attempts will ultimately prove futile as the speed of AI-driven development outpaces traditional practices.
• Depression: As it becomes clear that AI is redefining the rules, there will be a sense of loss. Decades of experience, certifications, and expertise will seem diminished overnight.
• Acceptance: Finally, the industry will adapt, recognising that AI isn’t just a tool but a transformation. The smartest organisations will be those that embrace this shift early, shedding outdated practices and embracing a future where speed and adaptability are the new norms.

The Timeline: Disruption Is Here—Don’t Wait to React

The tipping point for programming has already arrived. AI-driven code generation tools are rapidly gaining traction, and within the next 12 months, we will see the ripple effects extending to software engineering as a whole.

The project management apparatus that has long supported this industry will start to unravel, its relevance diminishing as AI’s capabilities render it redundant.

By this time next year, companies that have adopted AI-driven development practices will be delivering software at speeds unimaginable in today’s paradigm, with product iterations happening daily, if not hourly.

Those still anchored to traditional methodologies will find themselves outpaced and outmaneuvered, struggling to keep up in a world where speed, flexibility, and responsiveness have become the ultimate competitive advantages.

The Provocative Truth: It’s Time to Cut Away the Dead Wood

To CEOs reading this: the future of software engineering is unfolding before us, and it demands a radical rethinking of how we approach programming, project management, and product development. AI is not just a faster coder; it’s a paradigm shift that will cut away the dead wood that has accumulated over six decades of cumbersome development practices.

The question isn’t whether this change will happen—it’s already happening. The only question is whether you will be ready to lead your organisation through this transformation or be left clinging to a methodology that the future will quickly render obsolete.

Human-empowered AI productivity training for your team from the Simple Academy, here.