Article
AI - Artificial Intelligence
Culture
5 min read

What AI needs to learn about dying and why it will save it

Those programming truthfulness can learn a lot from mortality.

Andrew Steane has been Professor of Physics at the University of Oxford since 2002, He is the author of Faithful to Science: The Role of Science in Religion.

An angel of death lays a hand of a humanioid robot that has died amid a data centre
A digital momento mori.
Nick Jones/midjourney.ai

Google got itself into some unusual hot water in recently when its Gemini generative AI software started putting out images that were not just implausible but downright unethical. The CEO Sundar Pichai has taken the situation in hand and I am sure it will improve. But before this episode it was already clear that currently available chat-bots, while impressive, are capable of generating misleading or fantastical responses and in fact they do this a lot. How to manage this? 

Let’s use the initials ‘AI’ for artificial intelligence, leaving it open whether or not the term is entirely appropriate for the transformer and large language model (LLM) methods currently available. The problem is that the LLM approach causes chat-bots to generate both reasonable and well-supported statements and images, and also unsupported and fantastical (delusory and factually incorrect) statements and images, and this is done without signalling to the human user any guidance in telling which is which. The LLMs, as developed to date, have not been programmed in such a way as to pay attention to this issue. They are subject to the age-old problem of computer programming: garbage in, garbage out

If, as a society, we advocate for greater attention to truthfulness in the outputs of AI, then software companies and programmers will try to bring it about. It might involve, for example, greater investment in electronic authentication methods. An image or document will have to have, embedded in its digital code, extra information serving to authenticate it by some agreed and hard-to-forge method. In the 2002 science fiction film Minority Report an example of this was included: the name of a person accused of a ‘pre-crime’ (in the terminology of the film) is inscribed on a wooden ball, so as to use the unique cellular structure of a given piece of hardwood as a form of data substrate that is near impossible to duplicate.  

The questions we face with AI thus come close to some of those we face when dealing with one another as humans. 

It is clear that a major issue in the future use of AI by humans will be the issue of trust and reasonable belief. On what basis will we be able to trust what AI asserts? If we are unable to check the reasoning process in a result claimed to be rational, how will be able to tell that it was in fact well-reasoned? If we only have an AI-generated output as evidence of something having happened in the past, how will we know whether it is factually correct? 

Among the strategies that suggest themselves is the use of several independent AIs. If they are indeed independent and all propose the same answer to some matter of reasoning or of fact, then there is a prima facie case for increasing our degree of trust in the output. This will give rise to the meta-question: how can we tell that a given set of AIs are in fact independent? Perhaps they all were trained on a common faulty data set. Or perhaps they were able to communicate with each other and thus influence each other.  

The questions we face with AI thus come close to some of those we face when dealing with one another as humans. We know humans in general are capable of both ignorance and deliberate deception. We manage this by building up degrees of trust based on whether or not people show behaviours that suggest they are trustworthy. This also involves the ability to recognize unique individuals over time, so that a case for trustworthiness can be built up over a sequence of observations. We also need to get a sense of one another's character in more general ways, so that we can tell if someone is showing a change in behaviour that might signal a change in their degree of trustworthiness. 

In order to earn our trust, an AI too will have to be able to suffer and, perhaps, to die. 

Issues of trust and of reasonable belief are very much grist to the mill of theology. The existing theological literature may have much that can be drawn upon to help us in this area. An item which strikes me as particularly noteworthy is the connection between suffering and loss and earning of trust, and the relation to mortality. In brief, a person you can trust is one who has ventured something of themselves on their pronouncements, such that they have something to lose if they prove to be untrustworthy. In a similar vein, a message which is costly to the messenger may be more valuable than a message which costs the messenger nothing. They have already staked something on their message. This implies they are working all the harder to exert their influence on you, for good or ill. (You will need to know them in other ways in order to determine which of good or ill is their intention.)  

Mortality brings this issue of cost to a point of considerable sharpness. A person willing to die on behalf of what they claim certainly invests a lot in their contribution. They earn attention. It is not a guarantee of rationality or factual correctness, but it is a demonstration of commitment to a message. It signals a sense of importance attached to whatever has demanded this ultimate cost. Death becomes a form of bearing witness.  

A thought-provoking implication of the above is that in order to earn our trust, an AI too will have to be able to suffer and, perhaps, to die. 

In the case of human life, even if making a specific claim does not itself lead directly to one's own death, the very fact that we die lends added weight to all the choices we make and all the actions we take. For, together, they are our message and our contribution to the world, and they cannot be endlessly taken back and replaced. Death will curtail our opportunity to add anything else or qualify what we said before. The things we said and did show what we cared about whether we intended them to or not. This effect of death on the weightiness of our messages to one another might be called the weight of mortality. 

In order for this kind of weight to become attached to the claims an AI may make, the coming death has to be clearly seen and understood beforehand by the AI, and the timescale must not be so long that the AI’s death is merely some nebulous idea in the far future. Also, although there may be some hope of new life beyond death it must not be a sure thing, or it must be such that it would be compromised if the AI were to knowingly lie, or fail to make an effort to be truthful. Only thus can the pronouncements of an AI earn the weight of mortality. 

For as long as AI is not imbued with mortality and the ability to understand the implications of its own death, it will remain a useful tool as opposed to a valued partner. The AI you can trust is the AI reconciled to its own mortality. 

Review
Culture
Film & TV
Language
Music
6 min read

The Phoenician Scheme - opening the mind to wider horizons

Wes Anderson's new film widens our vision to a bigger world

Oliver is a Junior Research Fellow at Pembroke College, Oxford, writing and speaking about theology and AI.

Characters from a Wes Anderson film sit in a stylish plane interior.
Benicio del Toro and Mia Threapleton star.

Wes Anderson’s latest film – The Phoenician Scheme – has caused as much confusion amongst critics and viewers as it has the usual delight. It tells the story of Anatole – Zsa-Zsa – Korda, his mad-cap business scheme across an imagined near-Eastern world, and his growing relationship with his daughter (apparently), Liesl, a novitiate nun. There are the usual Anderson-ian tropes and characters, with superb cameos by Tom Hanks, Richard Ayoade, and Benedict Cumberbatch (worth watching in itself), and a real star turn for the young Liesl, Mia Threapleton.  

I first watched it on a transatlantic flight (viewer advisory: there are several scenes in rickety planes). I was hooked from the first moment. Why? Not just the usual Anderson style and panache and dead-pan weird story and acting. It was the music. Anderson himself first trained as a musician. It shouldn’t be a surprise that amidst the rest of Anderson’s meticulously designed and curated world the music should carry so much meaning.  

The opening scene (no spoiler, it’s in the trailer), involves the burning wreckage of a plane (viewer advisory). There are birds – crows, hovering. And from the wreckage, bloodied but unbowed, emerges Korda. We hear from a voiceover that this is by no means the first assassination attempt he has survived. It won’t be his last. But the music at this precise point? It is a dark and brooding short melodic fragment. Does this portray a dark and brooding – evil, even – presence in the main character? Indeed, this dark melodic fragment follows Korda around the whole film, a leitmotif.  

But far from it. And this is what delighted me and hooked me. Because this isn’t just any old dark and brooding melodic fragment. It is the opening notes of Stravinsky’s magnificent ballet score, his first hit for the Russian impresario in Paris, Diaghilev and his ‘Ballets Russes’, The Firebird. Now here’s the fun thing. If you know the ballet, you know that it is the magic of the firebird’s feather which brings new life out of death in the ballet’s wonderful conclusion. And that is because the Firebird story itself is based on another mythical bird-creature – the phoenix (remember the title of the movie). The mythical phoenix is a bird which cyclically dies in flames, only to be reborn from the ashes to new life. So immediately, even though all we can see is the burnt-out wreckage of a plane, what we might think to ourselves if we know our Stravinsky, is that perhaps what this melodic fragment signifies, far from a brooding menacing presence, is someone who is constantly going to reemerge from the ashes to new life. In fact, I immediately felt I would be surprised if that wouldn’t happen. Korda himself says at a certain point ‘I won’t die, I never do’. Just from a musical fragment, the whole story can be seen in one glimpse.  

There are two other Stravinsky ballets which Anderson skilfully deploys (although less intrusively than the Firebird theme): the joyous whirligig of the opening of Petrushka, and the searing epilogue of the ballet Apollo. Now the Petrushka music does seem to be associated with another character, just like Firebird is associated with Korda. In the movie, Petrushka appears in two moments of significance for Liesl, (apparently) Korda’s daughter, the novitiate nun (and therefore herself already intimately associated with music – The Sound of Music). But the telling thing here is that, unlike Firebird, Petrushka (the ballet) doesn’t end well for its eponymous puppet-hero. Petrushka is killed by another puppet, with only a fleeting appearance at the end as a ghost. So the music of the ballet of Petrushka, despite the excerpt we hear being full of joyousness and innocent youthful energy, and its association with Liesl, suggests that her journey in the film is going to go in a very different direction to the convent of her initial intentions. Once again, knowing the music and the whole pattern of it can foretell an entire history that will unfold, even just from a mere fragment.  

Now the next thing that is so fascinating here is the combination of Stravinsky and Wes Anderson. Stravinsky wrote several ballet scores for the ‘Ballet Russes’ and Diaghilev in the glamour of Paris of the 1920s and 1930s (amongst other famous ones are The Rite of Spring (which caused a riot), Orpheus, and Pulcinella). They are highly stylised pieces, often returning to Classical ideas and tropes (musically, as well as in theme), presenting stylised and formal dances, tableaux. And whilst all these descriptions could be applied to Anderson’s films, The Phoenician Scheme itself presents a series of quirkily introduced tableaux, with their own distinctive characters and settings. And, in the concluding scene, set in a theatre, all the characters are present all at once. A miniature mechanical device representing all of Korda’s business interests appears on a stage. And the music at that point? The opening movement of Pictures at an Exhibition (by Mussorgsky, a Russian composer from the generation before Stravinsky), music which presents its own series of musical tableaux. Artistic tableau, musical tableau, ballet, and now film presented as a series of tableaux all coming together in Anderson’s fertile imagination.  

But there is one last thing that is fascinating for us in this presentation of music and art and film and plot. There is a much earlier precursor for the technique I referred to above, of one musical fragment potentially carrying with it the implication and meaning of the whole work. That earlier precursor for this technique is found in the New Testament. The authors of the New Testament, especially Paul, were saturated in the texts which we now call the Old Testament, or what they thought of as their Scriptures (just as, we might say, Anderson is clearly saturated in Stravinsky). Scholars think the New Testament writers assumed a familiarity with those Scriptures in the hearers and readers of their new writings, or, alternatively, they were helping their hearers and readers newly think and imagine along the lines set out in the Scriptures. Time and again, as Richard Hays masterfully showed (in Echoes of Scripture in the Letters of Paul, and Echoes of Scripture in the Gospels), the authors resort to a technique called metalepsis. That is, in quoting or near quoting a few words or a phrase from their Scriptures, not only are the hearers/readers meant to understand that it is a quotation, but to import the sense of the entire passage or even book from which that miniature quotation emerges. It was Richard Hays’s groundbreaking work on this literary hermeneutical aspect which caused a sensation in New Testament studies in the 1980s and 1990s when it first emerged, because it opened up whole new lines of interpretation, without any question remaining about their veracity. What it means is that, as we read the New Testament, we have constantly to be aware of what Scriptures the writer had in mind, either consciously or semi-consciously, in order to allow that thought-world to permeate our reading. It is a reminder, whatever we are reading or watching or listening to, never to be too reductive about our own cultural horizons when we approach such a text, but to be listening and open and willing to be enlarged by the life-world of the text before us, as the great philosopher Paul Ricoeur used to say.  

Support Seen & Unseen

Since Spring 2023, our readers have enjoyed over 1,500 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