1,000th Article
AI
Creed
Death & life
Digital
6 min read

AI deadbots are no way to cope with grief

The data we leave in the cloud will haunt and deceive those we leave behind.

Graham is the Director of the Centre for Cultural Witness and a former Bishop of Kensington.

A tarnished humaniod robot rests its head to the side, its LED eyes look to the camera.
Nicholas Fuentes on Unsplash.

What happens to all your data when you die? Over the years, like most people, I've produced a huge number of documents, letters, photos, social media posts, recordings of my voice, all of which exist somewhere out there in the cloud (the digital, not the heavenly one). When I die, what will happen to it all? I can't imagine anyone taking the time to climb into my Dropbox folder or Instagram account and delete it all? Does all this stuff remain out there cluttering up cyberspace like defunct satellites orbiting the earth?  

The other day I came across one way it might have a future - the idea of ‘deadbots’. Apparently, AI has now developed to such an extent that it can simulate the personality, speech patterns and thoughts of a deceased person. In centuries past, most people did not leave behind much record of their existence. Maybe a small number of possessions, memories in the minds of those who knew them, perhaps a few letters. Now we leave behind a whole swathe of data about us. AI is now capable of taking all this data and creating a kind of animated avatar, representing the deceased person, known as a ‘deadbot’ or even more weirdly, a ‘griefbot’. 

You can feel the attraction. An organisation called ‘Project December’ promises to ‘simulate the dead’, offering a ghostly video centred around the words ‘it’s been so long: I miss you.’ For someone stricken with grief, wondering whether there's any future in life now that their loved one has gone, feeling the aching space in the double bed, breakfast alone, the silence where conversation once filled the air, the temptation to be able to continue to interact and talk with a version of the deceased might be irresistible. 

There is already a developing ripple of concern about this ‘digital afterlife industry’. A recent article in Aeon explored the ethical dilemmas. Researchers in Cambridge University have already called for the need for safety protocols against the social and psychological damage that such technology might cause. They focus on the potential for unscrupulous marketers to spam surviving family or friends with the message that they really need XXX because ‘it's what Jim would have wanted’. You can imagine the bereaved ending up being effectively haunted by the ‘deadbot’, and unable to deal with grief healthily. It can be hard to resist for those whose grief is all-consuming and persistent. 

Yet it's not just the financial dangers, the possibility of abuse that troubles me. It's the deception involved which seems to me to operate in at a number of ways. And it's theology that helps identify the problems.  

The offer of a disembodied, AI-generated replication of the person is a thin paltry offering, as dissatisfying as a Zoom call in place of a person-to-person encounter. 

An AI-generated representation of a deceased partner might provide an opportunity for conversation, but it can never replicate the person. One of the great heresies of our age (one we got from René Descartes back in the seventeenth century) is the utter dualism between body and soul. It is the idea that we have some kind of inner self, a disembodied soul or mind which exists quite separately from the body. We sometimes talk about bodies as things that we have rather than things that we are. The anthropology taught within the pages of the Bible, however, suggests we are not disembodied souls but embodied persons, so much so that after death, we don't dissipate like ethereal ‘software’ liberated from the ‘hardware’ of the body, but we are to be clothed with new resurrection bodies continuous with, but different from the ones that we possess right now. 

We learned about the importance of our bodies during the COVID pandemic. When we were reduced to communicating via endless Zoom calls, we realised that while they were better than nothing, they could not replicate the reality of face-to-face bodily communication. A Zoom call couldn't pick up the subtle messages of body language. We missed the importance of touch and even the occasional embrace. Our bodies are part of who we are. We are not souls that happen to temporarily inhabit a body, inner selves that are the really important bit of us, with the body an ancillary, malleable thing that we don't ultimately need. The offer of a disembodied, AI-generated replication of the person is a thin paltry offering, as dissatisfying as a virtual meeting in place of a person-to-person encounter. 

Another problem I have with deadbots, is that they fix a person in time, like a fossilised version of the person who once lived. AI can only work with what that person has left behind - the recordings, the documents, the data which they produced while they were alive. And yet a crucial part of being human is the capacity to develop and change. As life continues, we grow, we shift, our priorities change. Hopefully we learn greater wisdom. That is part of the point of conversation, that we learn things, it changes us in interaction with others. There is the possibility of spiritual development of maturity, of redemption. A deadbot cannot do that. It cannot be redeemed, it cannot be transformed, because it is, to quote U2, stuck in a moment, and you can’t get out of it.  

This is all of a piece with a general trajectory in our culture which is to deny the reality of death. For Christians, death is an intruder. Death - or at least the form in which we know it, that of loss, dereliction, sadness - was not part of the original plan. It doesn't belong here, and we long for the day when one day it will be banished for good. You don’t have to be a Christian to feel the pain of grief, but paradoxically it's only when you have a firm sense of hope that death is a defeated enemy, that you can take it seriously as a real enemy. Without that hope, all you can do is minimise it, pretend it doesn't really matter, hold funerals that try to be relentlessly cheerful, denying the inevitable sense of tragedy and loss that they were always meant to express.  

Deadbots are a feeble attempt to try to ignore the deep gulf that lies between us and the dead. In one of his parables, Jesus once depicted a conversation between the living and the dead:  

“between you and us a great chasm has been fixed, so that those who might want to pass from here to you cannot do so, and no one can cross from there to us.”  

Deadbots, like ‘direct cremations’, where the body is disposed without any funeral, denying the bereaved the chance to grieve, like the language around assisted dying that death is ‘nothing at all’ and therefore can be deliberately hastened, are an attempt to bridge that great chasm, which, this side of the resurrection, we cannot do. 

Deadbots in one sense are a testimony to our remarkable powers of invention. Yet they cannot ultimately get around our embodied nature, offer the possibility of redemption, or deal with the grim reality of death. They offer a pale imitation of the source of true hope - the resurrection of the body, the prospect of meeting our loved ones again, yet transformed and fulfilled in the presence of God, even if it means painful yet hopeful patience and waiting until that day. 

Celebrate with us - we're 2!

Since March 2023, our readers have enjoyed over 1,000 articles. All for free. This is made possible through the generosity of our amazing community of supporters.

If you’re enjoying Seen & Unseen, would you consider making a gift towards our work?

Do so by joining Behind The Seen. Alongside other benefits, you’ll receive an extra fortnightly email from me sharing my reading and reflections on the ideas that are shaping our times.

Graham Tomlin

Editor-in-Chief

Snippet
AI
Culture
Digital
Sustainability
3 min read

AI Barbie: does anyone think about destruction?

We choose waste and consumption over stewardship.

Jean is a consultant working with financial services and Christian organisations. She also writes and broadcasts.

An AI generated image of a Barbie-like Toy
AIn a Barbie world.

If you spend any time on any social media platform you would have probably seen the ChatGPT Barbie trend. Resembling packaged toys, the AI depicts you like a doll or action figure. At first, I thought I was only seeing it because of the LinkedIn algorithm. But then I started to see articles in my feed from mainstream media outlets teaching people how to do it.  

Generally, speaking, I am not a trend follower. I am one of those annoying people who doesn’t get involved with what everyone is doing just because everyone is doing it. Thankfully, I don’t suffer from FOMO (the Fear Of Missing Out) and I don’t think I am swayed much by peer pressure. But I like to stay informed about what is going on. So I can have something to talk about when I meet people in new settings and to remain relevant. So, when this started popping up in my feeds, I investigated it, and I was pleasantly surprised. 

I am not anti-AI. I have embraced and seen the benefits of AI in my own life (this sounds a bit weird, but I think you get my point). I understand and accept that it will, can and has improved productivity and creativity. I use ChatGPT all the time for social media content and captions, brainstorming, titles for articles, coding problems, research and language translations.  

But like many, I have long been sceptical about the growth of AI use and the viability of its long-term sustainability. I wouldn’t describe myself as a climate warrior, but I do believe that we have a responsibility to ourselves and the generations after us to use the finite resources of the planet frugally. The AI-powered Barbie trend throws that out of the window.  

The current Trump administration has facilitated a shift away from ESG (environmental, social and governance) targets in the world of business. For the most part, the criticism of this in the media (social and mainstream) has been focused on DEI targets. But perhaps, in the face of slow economic growth and because this began before the Trump administration took office, the move away from environmental targets or what I would call environmental stewardship, or frugality has received limited coverage.   

I have never understood why proponents of the climate emergency, have made themselves bedfellows and in some cases, wholehearted supporters of the AI revolution. A typical data centre uses between 11-19 million litres per day water just to cool its servers, that’s the equivalent of a small town of 30,000-50,000 people. The International Energy Agency (IEA) predicts by 2030 that there will be a doubling of electricity demand from data centres globally equating to slightly more than the entire electricity consumption of Japan. This growth will be driven by the use of AI in the US, China, and Europe. That’s why vocal support of the climate emergency and advocating escalated transition to AI, as is the position of the UK government, currently seems paradoxical to me.  

This isn’t hyperbole, Sam Altman, CEO of Open AI recently tweeted asking folks to reduce their use of the ChatGPT’s image generator because Open AI’s servers were overheating.  

That is why I have been pleasantly surprised, by some of coverage on the Barbie trend. Arguments are now being made more loudly about the true cost of unlimited AI expansion.  

I am not against progress or AI expansion entirely, and I have some support for the argument that governments have pursued net zero policies at a rate that is impractical, expensive and unviable for the average consumer in Western democracies. However, the Barbie trend reveals our tendency to choose waste and consumption for fleeting pleasure. For many of us, we have probably just thought, ‘It’s just a bit of harmless fun’. But the truth is it isn’t, it’s just that we can’t see the damage we are doing to the environment. That’s without going into the financial and privacy costs associated with the AI revolution. It really is a case of that age old adage, ‘Out of sight, out of mind’.  

The challenge is now that we know, what do we do? Do we continue to be part of wasteful AI trends? Or do we use AI to add value, increase productivity and solve problems?  

Celebrate our 2nd birthday!

Since Spring 2023, our readers have enjoyed over 1,000 articles. All for free. 
This is made possible through the generosity of our amazing community of supporters.

If you enjoy Seen & Unseen, would you consider making a gift towards our work?

Do so by joining Behind The Seen. Alongside other benefits, you’ll receive an extra fortnightly email from me sharing my reading and reflections on the ideas that are shaping our times.

Graham Tomlin
Editor-in-Chief