Elon Musk has spent years warning that artificial intelligence could overpower human civilisation, yet he is now building the very tools that may one day allow the human mind to break free from the limits of biology. Through xAI and Neuralink, he is creating the foundations for a world in which consciousness might be stored, transferred, or even improved. What once belonged to the realm of science fiction is moving quietly but steadily into scientific possibility.
I follow this moment closely. My work in responsible artificial intelligence has taught me that progress in one field often reshapes the moral questions in another. Here, the intersection of super scale models and invasive neurotechnology forces us to ask whether humanity is ready for the consequences of its own ambition.
xAI is the first pillar of Musk’s project. When he left OpenAI in 2023, he said he wanted to build intelligence that seeks truth without commercial or political constraints. The result is the Grok series of models. Grok three is already in wide circulation. Grok five, now delayed to 2026, is expected to advance long context reasoning and planning to a level Musk describes as “crushingly superior” to anything else in development.
Much of this acceleration comes from the scale of data and compute now available to xAI. Musk’s takeover of X earlier this year placed hundreds of millions of daily posts in his hands. It created a real time stream of human expression unmatched by any other platform. On top of that, xAI is building enormous computing clusters, including the rumoured Memphis facility stocked with more than one hundred thousand H100 chips. These systems are designed not only to train ever larger models but to prepare for neural simulation at far greater resolution.
The second pillar is Neuralink. Its pace should give anyone pause. As of November 2025, seven patients carry the N one implant. The early data is startling. The first recipient, Noland Arbaugh, now controls digital interfaces with remarkable fluency. Others have begun to recover movement by routing signals around spinal injuries. The company plans to move from dozens of implants to hundreds by the end of the year and to scale sharply through the next decade.
Neuralink’s latest threads contain more than four thousand electrodes per array. This is still a tiny fraction of the human brain but enough to convert thought into text at speeds above one hundred bits per second. Each new generation promises more density, less scarring, better longevity, and finer stimulation. Musk’s team believes that within a decade the device will record and replay neural patterns with fidelity that was once unimaginable.
This brings us to what Musk calls the real goal: digital immortality. He has repeated the same central claim all year. If we can map the full connectome of a human brain and run it on a suitable substrate, the person continues to exist. The pathway he describes is straightforward. Neuralink achieves whole brain read and write. xAI develops an intelligence capable of simulating biological circuits. The mind is transferred from a dying body to a machine, a cloud system, or, in Musk’s preferred version, a humanoid robot that does not age.
Whether this arrives in fifteen years, as Musk privately claims, or fifty, as more cautious scientists argue, it is no longer a theoretical argument. The pieces are being actively assembled.
Some say this matters less than building colonies on Mars. Yet colonies only delay the biological clock. A digital mind could exist indefinitely. It could accelerate its subjective experience, learn at impossible speeds, or live across multiple embodiments. The leap between a world of short lived biological beings and a world of persistent digital minds would be the largest transformation in the history of the species.
None of this removes the risks. A flawed upload could trap a mind in a state of pain or confusion. Unequal access could create a permanent divide between those who have biological limits and those who have none. And if the models controlling these systems outrun our ability to govern them, we would be placing humanity’s entire future in the hands of entities we do not understand.
Yet we must be honest. The alternative is to surrender to the inevitability of decay. For thousands of years death has been inescapable. For the first time in our history, serious people with serious resources are working to challenge that assumption. The United Nations pushes for ethical artificial intelligence. Musk’s team pushes for technological possibility. Somewhere between these two forces the next definition of humanity will emerge.
This decade will decide whether we remain bound to biology or step into an age where consciousness outlives the body that first carried it. The choice is no longer distant. It is being built in laboratories today, wire by wire and model by model.
The views expressed in this article are those of the author and do not represent the editorial position of The Southern African Times.
Sonny Iroche is CEO of GenAI Learning Concepts. He was a Senior Academic Fellow at the African Studies Centre at the University of Oxford from 2022 to 2023. He holds postgraduate training in Artificial Intelligence from Saïd Business School and serves on the National Artificial Intelligence Strategy Committee of Nigeria and on the UNESCO Technical Working Group on AI Readiness Assessment.
