We are once again talking about “digital resurrection”, a misleading and inappropriate term that, in reality, constitutes another problematic aspect of informational identity and control over data concerning us — or its economic exploitation by Andrea Monti Initially published in Italian by Italian Tech – La Repubblica
Periodically, there is renewed talk of “digital resurrection”, the creation of clones of deceased people made possible by the synergistic use of generative AI, holographic technologies and analysis of content relating to the “deceased loved one”.
Obviously, the clone is only an informational appearance. It is neither “alive” nor “conscious” but must only appear to be so; it has no memory or identity, but must give the impression of possessing them. That is enough to create a new market.
From virtual duets to ABBA avatars: the evolution of simulation
Such phenomena are certainly not new, if we think of “Unforgettable”, the famous duet between Natalie Cole and her father Nat. Back then, in 1992, the technology was clearly rudimentary and the interaction between father and daughter was only apparent, but from the viewer’s perspective, the result is not too different from what can be achieved with more sophisticated systems. Twenty years later, in 2022, it is Abba’s turn to replicate their 1979 appearance in Voyage, a concert featuring only their clones. And how can we forget Carrie Fisher’s posthumous cameo in an episode of the Star Wars franchise?
The commodification of intimacy: from entertainment to mourning
Today, the exploitation of informational identity — the set of data that defines who we are — is beginning to move beyond the confines of entertainment and invade highly personal spaces, as in the case of a murder victim whose relatives make the killer ask for forgiveness from a digital replica of the deceased.
This is symptomatic of the toxic relationship that is developing with chatbots and generative technologies, fuelled by the unscrupulous and deceptive use of words such as “digital resurrection” for commercial purposes. But it is also a powerful cognitive virus that conveys ideologies such as transhumanism, which preach the overcoming of the human condition and death through technological enhancement, to give life to a “superior humanity”.
The legal knot: who controls post-mortem informational identity?
The uninvited guest in this debate between sociology, psychology, information technology and pragmatic commercial interests is called “data” — or rather, the definition of its legal nature.
The creation of a personalised chatbot — because that is what we are talking about when we talk about digital resurrection — requires access to a significant amount of data relating to the subject whose outward personality one wishes to (attempt to) imitate. This means that there can only be two options for obtaining the result: either the person concerned voluntarily hands it over to a company that develops chatbots, or the data is handed over by someone else.
The rights of the legitimate heirs to the data of the deceased (especially if famous)
If this someone else is the legitimate heir — i.e. the person recognised by law as such — the problem does not arise. Photographs, videos, recordings and other content belonging to the deceased are part of the so-called “estate” and therefore become directly available to the heir, together with the related rights. Therefore, not only does the heir have the right to access, without limitation or conditions, for example, the deceased’s email accounts, social media profiles or storage, but they can also do whatever they want with them. This implies (having) the animated fetish of the deceased created and, in the case of a famous person, exploiting the possibility of having them appear in events, films or broadcasts for financial gain.
This aspect is particularly important in relation to the economic exploitation of the “memory” of famous people. The heirs of an artist, in fact, not only acquire the rights to their ancestor’s creative output, but also the right to earn money from their identity, thanks to copyright law, which allows rights to be attributed to a person’s image. Of course, when it was conceived, no one imagined that the law protecting portraits could also be applied to the possibility of animating them. But this does not change the terms of the issue: the right to one’s portrait extends to any technologically possible use, present and future. It is no coincidence that in the world of cinema, the main concern of actors—and their heirs—is that they will be easily replaced by digitised clones capable of acting, or at least “being on set” with a high degree of autonomy.
Contracts, consent and copyright in the post-mortem information age
A little more complex is the case of those who have acquired the right to use the data of the deceased through a contract entered into when they were alive, for example, by accepting a hypothetical “transfer of ownership” clause in exchange for a “free” service or because a “legitimate interest” of a company was deemed to prevail over the right of a person.
When a person dies, the contracts they have entered into are terminated (one of the first sad tasks following a bereavement is precisely the closure of utilities or bank accounts). Therefore, with the “upstream” contract no longer in force, those who had acquired the right to dispose of a person’s data as they wished would no longer be able to do so, unless the heirs agreed to enter into a new agreement.
The impact of the exploitation of individual memory on AI companies
If confirmed, such a scenario could have a huge impact on the operations of AI companies, and it is plausible to believe that, perhaps in the US, legal action will be taken to seek compensation for the heirs of someone whose informational identity has ended up in a dataset used to train a model.
The first round, fought in the European Union, saw a German court recognise that the data of (living) users can be used by a Big Tech company on the basis of a “legitimate interest without the need to contract or pay anything”.
Now we have to see if — and where — there will be a second round, to decide whether this “free for all” also applies to the data of former customers, not only and not so much to organise a fake digital resurrection, but to continue to feed the hunger of algorithms and fill the coffers of Big Tech.