> ## Content Index > Fetch the complete content index at: https://www.bontehmagazine.com/llms.txt > Use this file to discover other available public pages before exploring further. # Uber Leaves Nigeria. And What This Means for Africa. - URL: https://www.bontehmagazine.com/uber-leaves-nigeria-tesla-launches-ro/ - Published: 2026-09-04T20:00:07.000Z - Updated: 2026-09-04T20:00:07.000Z - Author: Emmanuel Cobbi - Tags: World Within roughly forty-eight hours, three stories landed that, taken separately, look almost unrelated. Uber announced it was leaving Nigeria after twelve years. Tesla began putting its steering-wheel-free robo Cybercab into public robotaxi service in Austin and opened an interest form for people who want to purchase individual Cybercabs or entire fleets for commercial use. Then OpenAI released GPT-6 Astra, its most capable broadly deployed model yet, with major advances in software engineering, computer use, cybersecurity, research, and professional work. Read together, they tell a much more uncomfortable story. One part of the world is moving into an economy where cars can work without drivers and software can increasingly perform entire categories of professional work. Another part is still struggling to build the basic conditions that allow those technologies to take root. [Uber entered Nigeria in 2014](https://www.reuters.com/world/africa/uber-exit-nigeria-after-12-years-operations-2026-09-02/?ref=bontehmagazine.com) and helped reshape urban mobility in Lagos and Abuja. On September 2, 2026, however, the company ended operations there after reviewing its investments, while Nigeria's ride-hailing market continued dealing with expensive fuel, inflation, currency pressure and increasingly difficult operating economics. Uber remains a global technology company spanning mobility, food delivery, freight and logistics, but one of Africa's biggest markets could no longer justify its place in that network. Then look at what is happening somewhere else. Tesla has now launched its[ purpose-built Cybercab](https://www.tesla.com/robotaxi?ref=bontehmagazine.com), a two-seat autonomous vehicle without a steering wheel or pedals, into limited public service in Austin. More interestingly, this is no longer only the Elon Musk presentation we have heard for years. Tesla has opened an official [interest form](https://www.tesla.com/robotaxi/interest?ref=bontehmagazine.com) for people interested in purchasing Cybercabs individually or as commercial fleets. Imagine what that means when the model matures. You leave home in the morning and your car drives you to work. Instead of spending the next eight hours depreciating quietly in a parking lot, you release it into a robotaxi network. It carries passengers, earns money, and by evening it returns to pick you up. The thing you bought to consume income begins producing it. That future is still [constrained by regulation, safety questions and limited deploymen](https://www.reuters.com/business/autos-transportation/us-auto-safety-regulator-opens-probe-into-nearly-1000-tesla-cybercabs-2026-09-04/?utm%5Fsource=chatgpt.com)t, and Tesla is already facing scrutiny from American regulators over the Cybercab. But the important distinction is that the conversation has moved from *could this exist?* to *who wants to own a fleet?* There is something fascinating and uncomfortable inside that transition. Autonomous vehicles could eventually displace people who currently earn their living by driving, yet the same technology creates a completely different income opportunity around ownership, software and fleet operation. Progress does not always preserve yesterday's job. Sometimes it creates tomorrow's asset. Then, on September 3, [OpenAI released GPT-6 Astra](https://openai.com/index/gpt-6-astra/?ref=bontehmagazine.com). Astra can perform complex computer tasks, browse, write software, analyse data, conduct research and produce documents, spreadsheets and presentations with much less human guidance. [OpenAI also reports state-of-the-art performance ](https://openai.com/index/safety-overview-gpt-6-astra/?utm%5Fsource=chatgpt.com)across several professional and computer-use benchmarks, while Astra became its first broadly deployed model to reach the company's highest "Critical" cybersecurity capability threshold. And this is where the African part of the story starts bothering me. Somewhere, a software engineer is learning how to use AI agents to build in hours what previously took days. A designer is learning AI-assisted 3D workflows. An accountant is figuring out how to automate reporting. A cybersecurity professional is learning how to supervise systems that can investigate vulnerabilities almost independently. Meanwhile, somewhere else, a student is still being taught Microsoft Office as though knowing how to manually format a spreadsheet is the destination. Someone is memorising basic HTML syntax for an examination while modern AI can already build, inspect, debug and modify substantial software systems. None of this means HTML, Excel, accounting principles or programming fundamentals have become useless. You need foundations to understand what automation is doing. The problem is where the education stops. If one system teaches the foundations and immediately moves students into AI-assisted development, automation and higher-order problem solving, while another teaches those same foundations and sends students into the labour market as though technology froze in 2018, the certificates may look similar. The people holding them are entering different economies. Africa was already trying to close the previous digital divide before this one arrived. GSMA has repeatedly documented the continent's enormous mobile-internet usage gap: millions live within broadband coverage but remain offline because devices, data and digital skills are still too expensive or inaccessible. Now AI is building another floor on top of that inequality. And technology compounds. Businesses that digitised early now possess data AI can use. Workers who became digitally fluent early can use AI to multiply their productivity. Countries that built stronger digital infrastructure can adopt automation faster because the previous layer already exists. The person beginning with paper records must first digitise before they can automate anything. This is why Uber leaving Nigeria matters in the same conversation as Cybercab and Astra. Not because one company leaving proves Africa is failing, and certainly not because America has solved every technological problem. Tesla's robotaxis remain limited, GPT-6 Astra will not erase entire professions tomorrow, and African entrepreneurs continue building remarkable companies. The concern is the **direction and speed**. One economy is debating how people should own fleets of cars that work by themselves. Another is deploying software capable of taking on increasingly sophisticated professional work. Too many of our systems are still asking young people to become excellent at the manual version of work those systems are beginning to automate. Africa's response cannot be panic, and it cannot simply be another conference about "the future of work." Universities need to update curricula faster, companies need to expose employees to automation earlier, governments need to make technology easier to build and deploy, while individuals must become far more aggressive about learning beyond what formal education gives them. Because every technological revolution destroys some advantages while creating others. The tragedy would not be that a driver eventually loses work to an autonomous car. The greater tragedy would be that, when ownership of autonomous fleets becomes a new business, we arrive late to that too. The same applies to AI. The danger is not only that Astra can write code faster than someone who spent years learning to write it manually. The danger is that somebody else is already learning how to command Astra while we are still arguing about whether people should be allowed to use AI at all. Another part of the world is not waiting for us to catch up. And increasingly, neither is the future.