We recently blogged about a supposed innovation in which Elon Musk’s company Neuralink successfully implanted a chip into a patient’s cortex. The company’s vision is that the device would process the electrical impulses generated by thoughts in the brain that are specific to movement, and then transmit them to various devices. If the method proves to be viable in the long term, the technology could have many applications.

Of course, possibilities immediately arise in the medical field, such as restoring mobility to patients who have suffered spinal cord injuries. A similar experiment has been done before. But there are also cases where the device can be connected to an external device to simply control it by your thoughts. The company plans to have an external app process the data. Based on the company’s previous objectives, it is reasonable to assume that the process will eventually include Artificial Intelligence (AI), for example in the processing of signals. This would create a more direct link between ‘man and machine’ than ever before.

Although the solution may seem utopian at first glance, it is worth considering the potential problems also (apart from the obvious medical risks, of course).

Making of a Cyborg

From the 90’s to the present day, the representation of transhumanism in pop culture as a possible path of human evolution has been a very popular idea. From animated films such as Ghost in the Shell to recent blockbuster films such as The Creator, the topos of a human being continuing to live in a robotic body has been seen in many approaches, although it has not yet become a reality.

Developments like the current one can represent a big step towards the augmentation of human existence, in which some of the previous limitations will be removed, and recovery from previously incurable conditions will become possible.

As Mónika Mercz asked, am I even considered to be fully human after enhancing my abilities with AI technology? Perhaps we are still far from having to think about whether we really remain human just because we can give instructions to the devices around us in a new way.

The other question she asked seems to be much more pressing soon: could my memories, thoughts and actions be influenced, and if so, who would be responsible for possible criminal behavior? Given that there are already many methods and stakeholders involved in manipulation and influence, research (whether ethical or illegal) that takes this problem to a new level may become more important.

Data protection

When we see that AI models can translate the impulses generated in the brain into concrete images, it is reasonable to ask what kind of data will we be able to extract from the mind in a few decades? And even more importantly, who and what will do with this new kind of data source?

Of course, machine-learning models that read our minds will not appear overnight. It is not certain that they will ever become widespread either. Just think of self-driving cars, which, despite all the promises they make, are still not on the roads by the hundreds.

An interesting question is whether our thoughts are as much personal data to be protected as, say, what we write in our private emails. Are they perhaps considered sensitive health data, which are already subject to strict rules in most countries? Solutions based on processing brain impulses, and later more complex thoughts, are unlikely to be feasible offline. Given the hardware requirements of more advanced AI-based solutions, it is more likely that the data processing will be done by some kind of cloud-based tool. Who will be responsible for data management and what will their responsibilities be?

You are the product again (?)

In the example of social media, we are already familiar with the phenomenon of how an initiative that originally promised only social benefits turns its users into commodities in order to maximize profits.

Although this seems like wild speculation today, the moon landing was once wild speculation, so it’s probably worth thinking about. The question is, for whom and how much are our thoughts worth? The answer seems trivial enough: for many market players, and a lot. The promise of Neuralink so far feels like a luxury product. If we are to believe it, secure data storage and privacy are likely to be among the services we can expect.

If brain signals can only be processed in such an invasive way (after a device has been surgically implanted), there is little reason to ‘fear’ mass use. However, if this changes (and it may be in the company’s best interest to do so, given the widening user base), it may not be for long. With technologies available to the masses, companies tend to generate revenue from whatever they can. Why shouldn’t these be our thoughts in the medium term?

Should we be worried?

As in most cases, the future will presumably not take an extremely utopian or dystopian direction. It is also possible that such relatively meaningless statements as that Neuralink has achieved “promising neuron spike detection” are, for the time being, coming more from the side of marketing than actual scientific results. We are also a long way from thinking about the possibility of our digital resurrection. It is also true that we cannot prepare for the unpredictable. What we can do, however, is to lay down the principles on which we want to act, for example in the event of the emergence of a technology that will learn more about us, who we are and what we think than ever before. In this respect, the very act of starting the discourse is a step forward.

István ÜVEGES is a researcher in Computer Linguistics at MONTANA Knowledge Management Ltd. and a researcher at the HUN-REN Centre for Social Sciences, Political and Legal Text Mining and Artificial Intelligence Laboratory (poltextLAB). His main interests include practical applications of Automation, Artificial Intelligence (Machine Learning), Legal Language (legalese) studies and the Plain Language Movement.