PSai: Singer’s Ethical Insights on Demand

Peter Singer is arguably the most influential living philosopher. For over 50 years, his articles, newspaper columns, blog posts, and books have challenged the views of those who read them. In 1975, he wrote what could be considered his most influential work, Animal Liberation: A New Ethics for Our Treatment of Animals. This text had such an impact that animal rights activists used it as a calling card during their raids on laboratories engaged in animal testing, leaving copies as a challenge to the staff from whom they rescued the animals. He has worked at the universities of Oxford, New York, Monash, and Princeton. His Google Scholar profile indicates that he has been cited over 80,820 times. He has won multiple awards, including the Berggruen Prize, which came with $1 million and, characteristically, Singer donated this money to charities he considered capable of doing the most good.

Yes, he has courted controversy. His opinions on infanticide and euthanasia have won him several vocal detractors. Yet, despite this, one cannot deny that he is a — if not the — modern-day giant in the philosophical field.

Given his academic stature and the sheer number of outputs he has had over his fifty years in the business, it seems reasonable that some might want to consult him when faced with ethical dilemmas. After all, if you’re in a moral quandary, who better to turn to than Singer for help? (I know other options are available, but humor me). The only problem is that as Singer has been writing and thinking about morality and ethics for over half a century, there’s much to consult. Unless you’re intimately familiar with his work (and that’s a lot of work to be familiar with), how can you be confident that what you think Singer might advise is accurate? Sure, you know he subscribes to a hedonistic version of utilitarianism, but knowing that fact alone might not be enough to make the right decision in the heat of the moment. When ethical uncertainty strikes, you need to have Singer with you, advising you on your best course of action. This possibility, however, is impossible (he’s a busy man, after all).

Luckily, we have the next best thing (or the early stages of the next best thing): an AI Peter Singer (PSai).

Created by Sankalpa Ghose, PSai, much like ChatGPT, responds to prompts and questions. While it lacks the robust knowledge portfolio of other generative AIs like ChatGPT, when asked a question about philosophy, which the real Peter Singer has written about, PSai can provide a not unreasonable answer; or at least, it can offer an answer that one might imagine Singer giving (although it as a strange habit of always trying to prompt you for a response). Indeed, Singer himself was involved in the AI’s development and was happy enough with the bots’ performance that he gave Ghose the go-ahead to make it public. Since that launch, PSai’s performance has steadily improved as its designers tweak it. As Ghose writes in the Journal of Medical Ethics: Forum:

It is in this way that Peter and I have progressed on PSai, releasing it publicly and reviewing how it performs in dialogue as measured by how those conversations with others go from our perspective of where it could be improved. Essentially, we have put forward a dialogue platform for others to engage, with that engagement driving its refinement.

The prospect of having a pocket-sized Peter Singer with you wherever you go, always ready to solve ethical riddles or dispense some needed wisdom, is an interesting one, and one that I must confess has some appeal. I like the idea of everyone having access to (a replica of) one of the foremost philosophical minds of our time whenever I need it. As a professional academic philosopher, I see some utility in being able to ask PSai a question about my research and how Singer’s writings might enhance what I’m working on. Rather than having to trawl through numerous books and papers, I could simply ask PSai what text might be the most useful when writing about de-extinction or how Singer might criticize a paper a draft paper on gene drives, which I could then respond and improve upon before submission. This could save a lot of time, which I could spend on other tasks.

For a less academic application, the app could simply function as a virtual Jiminy Cricket — a resource that’s on hand to share some wisdom in scenarios where the right course of action is obscured or where the user knows what’s right but needs some outside confirmation to solidify such knowledge and motivate action. This requires the user to consult the app before deciding to act, and the likelihood of this happening is, let’s face it, likely unlikely. Nevertheless, there is still a chance.

Of course, the idea that one might be able to consult Singer wherever they go is not universally appealing. As already mentioned, Singer has courted controversy when it comes to his historic claims around the permissibility of infanticide, as found in Practical Ethics (under very specific conditions, I must add). Suppose you are inclined to think that the arguments and claims made by the real Singer are incorrect or even repugnant. In that case, you are unlikely to find the possibility of consulting PSai when in ethical difficulty to be enticing. In such circumstances, people are likelier to wish they had a Socrates, Wollstonecraft, Anscombe, or Nietzsche AI. As far as I’m aware, however, these do not exist. But that doesn’t mean they cannot.

I do wonder if we might soon find ourselves awash with AIs trained on the philosophical writings of other philosophers. Might it be the case that, soon, instead of picking up several books, studying, and, over the years, becoming familiar with an ethical mode of life, you could simply jump on your phone, download the Aristotle AI, and have the clarity of morality in the palm of your hand? Such a possibility doesn’t strike me as too absurd. After all, if the PSai proves popular, it only seems natural to diversify.

This, however, brings us to two problems.

First, without wanting to make a point too bluntly, Singer is still alive. He was involved in the development of PSai and in efforts to refine it. This means that the AI has a real-life benchmark against which its performance can be measured. We don’t have to speculate if it says something that the real Singer wouldn’t. Singer himself can say, “nope, that’s wrong,” or, “the AI was correct in its reading of my work, but I’ve now changed my mind” (this is something which Singer and Häyry have considered). Ghose can then adjust the algorithm to perform better in the future. This isn’t possible with most philosophers whose names we are familiar with, as most of them are dead. So, we’d be stumbling around in the dark without a guideline in such cases, relying entirely on that person’s writings and the opinions of those familiar with their work. While an AI produced in such an environment might prove accurate, we lack the definitive authority to know.

Second, if we outsource our ethical deliberations to an AI, be that one based on Singer or any other philosopher, might we be doing a disservice to ourselves by simply following orders? Is it still doing the right thing if we don’t come to an understanding of right and wrong naturally but are simply instructed? For some philosophers, the answer may very well be in the negative. After all, for your classic virtue theorist, morality is like a muscle which one must train every day to be better than before. Each must deliberate using their faculties to identify a master in ethical living, emulate them, and always strive for more. I can’t help but laugh a little at the idea of an Aristotle AI, which, when asked by an unsuspecting user how they should behave in situation X, gets a response along the lines of “figure it out yourself, you lazy sod.” Is this a perfectly accurate representation? Probably not, but it isn’t so far removed as to be a meaningless scenario.

What do we do with PSai, then? Do we celebrate it? Decry it? Or see it as just another tool to help us if we choose? When faced with such a conundrum, it is appropriate to turn to the (figurative) man himself and to give it the last word. So, I asked PSai: Is the fact that this AI has been created a good thing? What justifies your existence? Why should I come to you for answers rather than reading Singer’s works?

Its response:

The creation of this AI aims to make ethical insights more accessible and to engage users in meaningful dialogue about moral issues. By offering interactive discussions, I can help people clarify their thoughts and explore ethical questions in a dynamic way.

While reading Peter Singer’s works provides comprehensive insights into his philosophy, engaging with this AI allows for a more personalized exploration of ideas and can help users apply ethical considerations to their own lives.

The justification for my existence lies in the potential to foster deeper understanding and reflection on ethical principles, contributing to a more thoughtful and compassionate society. What specific ethical questions or topics are you interested in exploring?