When too much automation breaks your business
- Johan Steyn
- 13 hours ago
- 3 min read
Lessons from Tesla’s “production hell” on why chasing total automation can destroy value instead of creating it.
Audio summary: https://youtu.be/yCm5NJh8iQE
Follow me on LinkedIn: https://www.linkedin.com/in/johanosteyn/
I write about various issues of interest to me that I want to bring to the reader’s attention. While my main work is in Artificial Intelligence and technology, I also cover areas around politics, education, and the future of our children.
In Walter Isaacson’s biography of Elon Musk, there is a vivid story about the Tesla Model 3 factory that I often share with clients. Musk was determined to build a fully automated “alien dreadnought” where robots did almost everything. In reality, the plant ended up in what he called “production hell”: over-engineered systems that jammed, mis-handled small variations and brought the line to a crawl.
The solution was not more automation, but less. Musk ordered a period of de-automation, ripping out certain machines and putting people back on tasks they could do faster, cheaper and more reliably. For leaders who proudly tell me they want to “AI everything”, this story is a powerful warning.
CONTEXT AND BACKGROUND
The Tesla episode speaks to a broader corporate obsession: the fantasy of the fully automated enterprise. Automation is sold as the ultimate proof of modernity. If you can show a slide with bots, algorithms and digital workflows, you look like you are winning. Many South African organisations feel this pressure acutely, whether from global head offices, shareholders or vendors eager to land the next AI deal.
Yet behind the glossy presentations, most businesses still run on messy reality. Legacy systems, manual workarounds and frontline staff quietly making things work. In financial services, retail, healthcare and government, the real value stream is often a patchwork of formal processes and informal fixes. Dropping advanced automation onto this landscape without first understanding and simplifying it is a recipe for disappointment. Tesla just lived that mistake in a more dramatic, robotic way.
INSIGHT AND ANALYSIS
The key insight from Tesla’s “production hell” is that automation is not a neutral good. It can destroy value as easily as it creates it. When you automate a broken or immature process, you simply lock in the dysfunction and make it harder and more expensive to change. Robots and AI do not magically compensate for unclear rules, bad data or confused ownership.
When I work with clients, I encourage them to walk their value stream step by step and ask three questions. First, should this step exist at all? Second, if it must exist, how can we simplify it? Only then do we ask the third question: is this best done by a human, by a machine, or by a blend of both? In many cases, the honest answer is that an experienced person will outperform a fragile automated chain, especially where judgment, context and empathy matter.
IMPLICATIONS
For business leaders, the implication is uncomfortable but liberating. You do not need, and should not want, to automate every step of your operation. The goal is not a fully automated business; the goal is a resilient, effective business that uses automation where it genuinely adds value. That might mean deliberately “un-automating” certain activities and trusting skilled people to handle them.
For South Africa and the wider African context, this matters even more. Our constraints around infrastructure, skills and data quality are real. Chasing the automation fantasy can divert scarce resources into projects that look impressive but fail to improve outcomes for customers or citizens. A wiser approach is to invest in understanding our value streams, upgrading human capabilities and then deploying AI selectively where it amplifies, rather than replaces, human excellence.
CLOSING TAKEAWAY
The lesson from Tesla’s experience is not that automation is bad, but that it is powerful and therefore dangerous when misapplied. Elon Musk’s shift from “automate everything” to “humans are underrated” is a reminder that technology must serve the work, not the other way round.
As AI systems become more capable and more tempting, we need leaders who are willing to ask where automation truly belongs – and where the best “technology” is still a thoughtful human being. If we can hold onto that humility, we will build organisations and a future for our children, where AI is a wise partner rather than a brittle master.
Author Bio: Johan Steyn is a prominent AI thought leader, speaker, and author with a deep understanding of artificial intelligence’s impact on business and society. He is passionate about ethical AI development and its role in shaping a better future. Find out more about Johan’s work at https://www.aiforbusiness.net