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

Those programming truthfulness can learn a lot from mortality.

Andrew Steane

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

Awe and wonder

Books

Culture

Poetry

10 October 2024

6 min read

Charles Taylor on how poetry seeks cosmic connections

The philosopher yields an array of luminous insights.

Paul Weston

Paul Weston is a Fellow at Ridley Hall, Cambridge.

At dusk, three people sit on a field edge and look at the stars emerging.

Rad Pozniakov on Unsplash.

Charles Taylor, Cosmic Connections: Poetry in the Age of Disenchantment

At just shy of 600 pages, philosopher Charles Taylor’s latest book is not for the faint-hearted. At the heart of the book are fascinating questions. What kind of language should we use when we encounter ‘beauty’, or experience ‘wonder’ which seems to take us into a new kind of space, or make time stand still? It could be the sight of a breathtaking landscape. It could be a piece of music. It’s that sense of ‘connection’ with something bigger than ourselves. What might it mean? More particularly, how do we express what we intuitively feel to be real and deeply significant about such things when our usual language fails to capture it?

Taylor says that his latest book ‘is about (what I see as) the human need for cosmic connection . . . one shot through with joy, significance, inspiration’. And his hypothesis is that ‘the desire for this connection is a human constant, felt by (at least some) people in all ages and phases of human history’, even if ‘the forms this desire takes have been very different in the succeeding phases and stages of this history’. The book explores these questions by focusing on late-Romantic European poetry and traces its development through the work of later poets including Goethe, Rilke, Wordsworth, Hopkins, and T. S. Eliot.

The background to Taylor’s explorations is the ongoing impact of the Enlightenment on our language and understanding about ‘truth’ and ‘meaning’. Those familiar with his 2007 book A Secular Age will find echoes of it here, particularly the way in which post-Enlightenment language tended to develop the language of control. This, he argues, became increasingly dominant in Western modernity ‘because it is linked to a practical stance which is basically instrumentalist; we seek out the efficient causal relations in our world with the aim of discovering handles which will enable us to realise our purposes’. In reaction to this narrowing of possibilities, the Romantic poets focused on ‘the experience of connection, and the empowerment this brings: not a power over things, but one of self-realization’. And a key element of the Romantic movement was the recognition that a poem (alongside the wider arts) uncovers deeper meanings: it ‘reveals to us, brings us into contact with, a deep reality which would otherwise remain beyond our ken’.

‘Poetry goes beyond creating a mood, an atmosphere of feeling, and claims to give access to the inner force in a thing, not by describing it, but by making it palpable’

Charles Taylor

The book focuses on the variety of forms that this reaction took amongst Romantic poets. A unifying desire for ‘connection’ led to differing ideas about how poetic language makes this possible and what kinds of meaning are revealed. But the central belief remained constant: that poetic language was the key to addressing a prevailing cultural atmosphere of ‘disenchantment’, in which the desire for cosmic connection had been sidelined.

On the English side of the channel, Taylor finds in Keats’ poetry, a new form of expression summed up in his statement ‘Beauty is Truth and Truth Beauty’. In Taylor’s words, ‘Art raises the object to a new unity and intensity, which constitutes Beauty. But this is not something which just exists in the mind of the artist (or reader); it has reality, and hence Truth, even though this reality is partly brought to fruition by artistic (re)creation’.

Similarly, for Gerard Manley Hopkins, the form of poetic language itself can become a means of ‘connection’. As Taylor puts it, in Hopkins’ work, ‘Poetry goes beyond creating a mood, an atmosphere of feeling, and claims to give access to the inner force in a thing, not by describing it, but by making it palpable’. His poetry embodies this in that it ‘renders the rhythms of the being itself through the “sprung rhythm” of the verse’.

The second half of the book looks to poets of the last 200 years who have navigated parallel pathways towards this ideal of experienced fullness – in the face of increasing industrialisation and disenchantment. Baudelaire longs for the experience of fullness, but finds it barred by a state of melancholy that he described as ‘Spleen’. It is a melancholy brought about by the endless repetitions of mundane and urbanised life. Baudelaire’s attempt to find release from our imprisonment in trapped time is to face Spleen head on and to transform it through poetic contemplation.

T. S. Eliot’s poetry similarly aims to evoke the sources of a fuller life in a culture of decay by means of the ‘continual surrender’ of the poet to something more significant and valuable. In Eliot’s case this resolves in a more-or-less traditional sense of Christian order, but Taylor notes that his poetry doesn’t necessarily require this. Miłosz on the other hand, amidst the social and political upheavals in Poland, sought a higher form of poetry: a poetry that could rise above the discord of social turmoil in order to define a moment and clarify the path that needed to be taken.

‘The work of art opens for us a new field of meaning, by giving shape to it’.

Charles Taylor

At times, the weight and detail of Taylor’s exposition threatens to overwhelm the reader, but he offers an array of luminous insights along the way, and he is largely successful in keeping an eye on the broader questions and themes. Perhaps the most important here is his belief that the human desire for ‘cosmic connection’, with its yearning for joy, significance and inspiration is perennial. The book is in this sense an elaborate historical worked example of this desire, understood – perhaps imperceptibly – as a sense of ‘loss’ or ‘longing’. It is this ‘central aspiration of the Romantic period’, he concludes, that ‘remains powerful today’.

My own experiences of talking with people today seems to amply confirm Taylor’s view. And the search for language to describe the desire for this sense of connection (or the longing for it) continues to thrive even within a so-called secularised culture (most likely because of it). It too seeks language for expression and sometimes struggles for some of the same reasons that the Romantics identified. Taylor’s phrase ‘the immanent frame’ (from his A Secular Age) powerfully evokes a sense of being held on a restricted rein, unable to name or explore the realms of ‘beauty’ or the ‘transcendent’ beyond perceived cultural boundaries. Connected to this is the still commonly held belief that any supposed knowledge derived through the arts doesn’t put us in touch with anything other than ourselves.

Taylor convincingly transcends the supposed dichotomy between these so-called ‘subjective’ or on the other hand, ‘ontological’ possibilities. ‘The work of art’, he says, ‘opens for us a new field of meaning, by giving shape to it’. Moreover, it can ‘realize a powerful experience of fulfillment . . . of connection which empowers’. And Taylor backs his theory with autobiography. In Goethe’s ‘Wanderer Nachtlied’, he says, ‘There is a kind of rest/peace which I long for. I don’t fully understand it, but now I have some sense of it’. Or broadening the artistic field, he finds that listening to Chopin’s Fantasie-Impromptu in C sharp minor ‘opens me up to an unnameable longing’, or that when he listens to Mozart’s Jupiter Symphony, he feels ‘a striving upwards, an expression of praise and thanks, straining to reach some higher addressee (for me, this would be God), but I can imagine that someone else, feeling the ascending movement, would imagine another destination’).

Here then is a book that shines a light on the possibilities of ‘cosmic connection’. Taylor affirms our desires, reassuring us that we should not feel strange to share the same longings for ‘fullness’, for ‘transcendence’, for ‘joy’ and for ‘connection’ as did the Romantic poets. Our recognition of these desires may well have been evoked by the kind of poetry or music that Taylor talks about here. But for all of us, it is good to be reminded that in a western world still heavily influenced by the climate of secularisation many today are still searching for the sense of cosmic connection that Taylor describes.

Charles Taylor’s Cosmic Connections: Poetry in the Age of Disenchantment is published by Harvard University Press.