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.

Article

Assisted dying

Comment

Culture

Politics

3 December 2024

5 min read

The assisted dying debate revealed the real role of Parliament

MPs from areas where people are vulnerable and at risk were more sensitive to the dangers.

M. Ciftci

Mehmet Ciftci has a PhD in political theology from the University of Oxford. His research focuses on bioethics, faith and politics.

An MP stands and speaks in a parliamentary debate.

MP Diane Abbott speaks in the debate.

Parliament TV.

What would be the effect of allowing assisted suicide for those ‘people who lack agency, the people who know what it is to be excluded from power and to have decisions made for them’, asked Danny Kruger MP, as he wrapped up his speech? ‘What are the safeguards for them? Let me tell the House: we are the safeguard—this place; this Parliament; you and me. We are the people who protect the most vulnerable in society from harm, yet we stand on the brink of abandoning that role.’

His words capture an important aspect of Friday’s debate: what is the point of Parliament? Do MPs meet to turn public opinion polls into policies? If the majority are in favour of something, do MPs have nothing left to do but to follow the public and sort out the fine details? We might instinctively say ‘Yes!’ It seems right and democratic to treat those whom we elect as people we select and send to do our bidding. And the polls do seem to show the majority of people supporting assisted suicide, at least in principle – although there are good reasons to be sceptical about those figures and about the conclusions drawn from them.

But there are numerous times when the majority are known to be in favour of something but politicians refuse to endorse it. Polls repeatedly show that a majority are in favour of reintroducing the death penalty. Why might it be right for MPs sometimes to ignore what the purported majority thinks and to use their own judgement?

Because Parliament is not just a debating chamber.

An older way of referring to it was to call it the ‘High Court of Parliament’ because ‘parliament, classically, was where individuals could seek the redress of grievances through their representatives,’ as law lecturer Dr Robert Craig writes. It performed its function admirably in response to the Horizon scandal: a legitimate grievance was brought to its attention, and it responded to redress the wrongs done to the sub-postmasters by passing a law to ‘overturn a series of judgments that could only have been obtained, and were only obtained, by a toxic, captured and wilfully blind corporate culture’.

Friday’s debate featured many MPs who understood what they were there to do. They acknowledged the ‘terrible plight of the people who are begging us for this new law’ as Danny Kruger said. But they also spoke up for those who were in danger of being harmed and wronged by the bill: the disabled and the dying, and all the vulnerable who were not there to speak on their own behalf.

Many echoed the concerns expressed by Diane Abbott about coercion: ‘Robust safeguards for the sick and dying are vital to protect them from predatory relatives, to protect them from the state and, above all, to protect them from themselves. There will be those who say to themselves that they do not want to be a burden. … Others will worry about assets they had hoped to leave for their grandchildren being eroded by the cost of care. There will even be a handful who will think they should not be taking up a hospital bed.’ And evidence of coercion is hard to find and trace: ‘Coercion in the family context can be about not what you say but what you do not say—the long, meaningful pause.’

An analysis shared on X by law lecturer Philip Murray found an association between the level of deprivation in a constituency and how likely a Labour MP was to vote against the bill. He also shared figures showing that 2/3 of MPs from ethnic minorities voted against it. In other words, MPs from areas where people are vulnerable and at risk were more sensitive to the dangers of helping people to kill themselves.

The second reading of the bill on Friday was a crucial moment for them to decide whether the bill would fix an injustice or whether it would itself cause harm.

But it seems that many MPs did not appreciate what the debate was about or what they had gathered to do. Layla Moran MP said: ‘The media are asking all of us, “Are you for or against the Bill?”, but I urge hon. Members to think about the question differently. The question I will be answering today is, “Do I want to keep talking about the issues in the Bill?”’ But James Cleverly MP intervened: “she is misrepresenting what we are doing at this point. We are speaking about the specifics of this Bill: this is not a general debate or a theoretical discussion, but about the specifics of the Bill.” He was right to be impatient. Unlike the Oxford Union, the vote has consequences. Parliamentarians are not there merely to debate. As the term ‘High Court of Parliament’ suggests, when MPs (either on their own initiative or as a government) propose bills, what they are often doing is conveying a plea to redress some grievance, and their debates are to decide whether to respond by making laws to grant justice to the wronged.

The second reading of the bill on Friday was a crucial moment for them to decide whether the bill would fix an injustice or whether it would itself cause harm, because the scrutiny that the bill will undergo in the following stages is not likely to be as rigorous as with government bills. As a Private Member’s Bill, the assisted dying proposal is free to be scrutinised by a committee selected by the MP who has proposed the bill, i.e. Kim Leadbeater. When the bill reaches the stage for a final vote in the Commons at the third reading, no further amendments can be made and the time for debate is likely to be short.

It is rare but bills are sometimes defeated at the third reading. With eighteen abstentions on Friday and at least thirty-six MPs claiming they might change their minds later, there is still hope.

Each sitting of the Commons begins every day with a prayer by the Speaker’s Chaplain, who prays that MPs ‘may they never lead the nation wrongly through love of power, desire to please, or unworthy ideals but laying aside all private interests and prejudices, keep in mind their responsibility to seek to improve the condition of all mankind.’

We can only hope and pray that at their next opportunity, MP will consider this bill in light of their responsibilities as the country’s High Court, charged with protecting the most vulnerable in society from harm.