January 26, 2021 - 7:00am
Imagine your wife died suddenly, and you struggled to cope with the loss. Well, soon you’ll be able to converse with her synthetic ghost, via a chatbot! Microsoft has filed a patent to access “social data” such as images, social media posts, written letters and so on “to create or modify a special index in the theme of the specific person’s personality”. This could then be used “to train a chat bot to converse in the personality of the specific person”, while “a 2D or 3D model of a specific person may be generated” to go with it.
I volunteered for some years at a bereavement counselling service. The people who presented for help with bereavement weren’t those who were sad when loved ones died. Every normal human being grieves. The individuals who needed help were those who’d become stuck along the path to accepting loss, usually because the relationship with the lost loved one was complicated.
We’ve all long since become accustomed to the way algorithms customise our digital experience for us, based on user data we provide as the cost of free digital services. Google uses our previous search history to offer us whatever it thinks will be most relevant (or commercially advantageous to Google); Facebook and Twitter algorithms try and curate our newsfeed for maximum “engagement”.
Microsoft’s patent on “digital ghost” technology suggests the next step along from customised online media content is to offer a customised alternative reality. If the aim of bereavement counselling is to come to terms with an irreducible loss by working through difficult memories with sympathetic help from a counsellor, here the aim is rejecting any need to accept irreducible emotional realities full stop. There should be no limit to our ability to tailor digital realities to individual taste.
We should not welcome this development. The more we come to expect an online world tailored to individual desire, the less willing we’ll be to compromise on a shared world offline. Consider the recent invasion of Washington’s Capitol by individuals convinced they were saving America from a coup, even as their opponents read the same events (with the same absolute conviction) as evidence of a coup. Or consider the growing political pressure for fluid and customisable individual “gender identity” to take priority in law over the less malleable and more stubbornly binary traits of human biology.
The digital realm has phenomenal power to reflect, then mimic and commodify our inner worlds on an increasingly individualised basis. This is driving an increasingly widespread expectation that the same level of “mass customisation” is also possible offline, whether in consumer products or political realities. This expectation is in increasingly radical tension with the collective nature of any workable political settlement, and, more fundamentally, with our nature as social beings.
Join the discussion
Join like minded readers that support our journalism by becoming a paid subscriber
To join the discussion in the comments, become a paid subscriber.
Join like minded readers that support our journalism, read unlimited articles and enjoy other subscriber-only benefits.
SubscribeIt all sounds rather like the technological equivalent of a Victorian seance. But what I want to know is, where do I go to buy the advertising space? Just imagine the scene: “Yes, dear, we’re all so happy here on the Other Side — because we all use Brand X washing powder!”
Brilliant article. “The more we come to expect an online world tailored to individual desire, the less willing we’ll be to compromise on a shared world offline.” If only the whole world could see this and not only the profiteers hurtling us towards dehumanized, dystopian division and disaster.
You said everything I wanted to. Just here to appreciate the comment and congratulate the writer aswell
This is all very reminiscent of Charlie Brookner’s “Be Right Back” – where the bereaved woman ends up sticking the deceased partner’s robot in the attic because it couldn’t truly reflect the real person’s spirit.
Yes. My first thought reading the beginning of the article: BLACK MIRROR AGAIN
“…We should not welcome this development…”
That assumes humanity has agency over the direction the algorithmic and gene-edit technologies are taking us in short order. That assertion is debatable – because I don’t think we do. There is zero chance of tech advance panning out as planned for anyone. It is nothing to do with political systems, capitalism or socialism or crony communism, or choices made by individuals. It is to do with the nature of tech, the trajectory which is essentially beyond human agency.
I’d agree that the commercially driven agenda of “digital ghosts” is an unnerving prospect. But in a crude way, isn’t that what we do already, with photos, videos, diaries etc of those who have died? I recorded a couple of chats with a dying friend in 2019, just brief discussions about his life and work. His widow listens to them regularly, and for that matter so do I. This new development seems rather like a logical extension of that.
I think it’s the difference between an actual recording of what the dear departed actually said, and an AI simulation of a thing they might have said.
I think there are serious ethical issues here that need to be unpacked. Let’s assume for the moment that there is agreement that such a service is sound in principle (a point well worth exploring in itself). Given that such a service. would have a huge effect on someone’s emotional or spiritual condition, we presume that it would at least need to be accurate: that is, that the behaviour of the digital replication of the deceased should express the same sentiments that the deceased really would have in real life. What scientific validity could be legitimately be claimed for such an assertion? Where is the research that validates it? I don’t see anything in the patent to support that.
At the risk of sounding all sci-fi, imagine if an immensely respected political leader, someone like Nelson Mandela, gets one of these post-mortem AI makeovers. Do current political leaders go to him/it for advice? What do him/it’s followers do when current political leaders do something different from what ‘Nelson-in-heaven’ would have done.
It’s all very well talking about bereaved spouses getting comforted etc but the implications here are truly huge.
Now that’s scary! haha
Even more because it actually makes sense, as we get further and further away from reality and time in the digital world. Also reminds me of Harari’s Homo Deus and the human pursuit for immortality – now openly declared and stablished as a goal.
My overarching view is that pretty much all of humanity has *never* known what it actually is, which is what creates the dissonance between what humans think they are and what agency they think they possess, and what algorithms and gene-decoding are subsequently deciding they actually are. But the fact that the algorithms and the gene-crackers are holding up this mirror to humanity, creates feedback loops. If humans care to look at these different reflections of themselves, they will react and change themselves – full scale self-hacking when the bio-edit and mind-machine interface technologies eventually allow. Potentially in directions that are outright nihilistic.
What a dreary lament. If I wish to create my own online world nobody has the right to stop me.