New Article on Artificial Wisdom, i.e. the Intersection of Ethics and Computer Science

New Article on Artificial Wisdom, i.e. the Intersection of Ethics and Computer Science August 10, 2018

An article that my colleague Ankur Gupta and I wrote together, exploring the intersection of science fiction, artificial intelligence, wisdom, ethics, and religion, has appeared in a special issue of the journal Religions. The special issue is titled “So Say We All: Religion and Society in Science Fiction” and our article has the title “Writing a Moral Code: Algorithms for Ethical Reasoning by Humans and Machines.” Here’s the abstract:

The moral and ethical challenges of living in community pertain not only to the intersection of human beings one with another, but also our interactions with our machine creations. This article explores the philosophical and theological framework for reasoning and decision-making through the lens of science fiction, religion, and artificial intelligence (both real and imagined). In comparing the programming of autonomous machines with human ethical deliberation, we discover that both depend on a concrete ordering of priorities derived from a clearly defined value system.

Visit the journal’s website to read the whole thing online or download the article in pdf format.

If the article seems too long to keep your interest, the XKCD cartoon below about Asimov’s Three Laws of Robotics sums up one crucial point that we explore. But I really do hope that the cartoon will whet your appetite to want to see what we do with it, rather than seeming acceptable as a substitute for reading it!

The journal, being open access online, makes articles available as soon as they are through the peer review and editing processes. Ours was the first through that process, and I’m eagerly looking forward to reading what the other contributions turn out to be!

Elsewhere online, you can explore some recent articles related to at least some aspects of these intersections, if not always quite as many of the threads as Ankur and I sought to incorporate:

The Theology of Battlestar Galactica (A New Podcast)

An NSF grant to explore the ethics of driverless cars

Only a Game on self-braking cars (that’s the blog of the author of The Virtuous Cyborg, where you’ll find more on this subject)

Saving ignorance from AI in Nautilus

New Humanist on AI and common sense

Vox and IO9 on woke droids and more from the Star Wars universe

David Brin on science fiction scenarios

Bias as the real danger of AI

Catholic thinkers on the ethics of artificial intelligence

Gillian Whitaker shared the authors contributing to a collection themed around AI and robots

Android Soup for the Soul: How Robots Model Humanity

AI and emergence: An essential meld?

Imago Hominis

In the European Union, the decision was made recently to recognize robots as persons – in the same sense that corporations can be persons before the law.

Summer Special: Could a robot ever have a real human identity?

Lots from (or via) 3 quarks daily:

How Big Data Is ‘Automating Inequality’

Automating Inequality: How High Tech Tools Profile, Police, and Punish the Poor

Artificial Intelligence Is Infiltrating Medicine — But Is It Ethical?

Will humans ever conquer mortality by merging with technology?

To Build Truly Intelligent Machines, Teach Them Cause and Effect

It’s Time for Technology to Serve all Humankind with Unconditional Basic Income

We’re talking about “sex robots” now

The meaning of life in a world without work

What if the Government Gave Everyone a Paycheck?

How a Pioneer of Machine Learning Became One of Its Sharpest Critics

Inside Trends And Forecast For The $3.9T AI Industry

The Disruption Ecosystem

The Threat of AI Weapons

The New Yorker on how frightened we should be of AI

5 Great TED Talks on the Potential of Artificial Intelligence

A couple from Steve Wiggins:

Eternity, Technically

Wired for Good

New York Times article on whether there is a “smarter path” to AI

AI and wealth redistribution

Religious Studies Opportunities Digest – 5 June 2018

https://relcfp.tumblr.com/post/174156653006/boston-university-graduate-student-conference

Finally, let me link to a classic article by Robert Geraci about AI.

"The discussion is not about changing weather but about the climate. I wonder how many ..."

Not Liberal, Just Literate
"You believe the sky is falling? Why? The weather has always changed."

Not Liberal, Just Literate
"I pulled out an old book from my dad's bookshelf: "Those Incredible Christians" and its ..."

Peer Reviews, Pseudoscience, and Denialism
"Thanks for the preview James. I'm very excited about the book and the anticipation is ..."

Preview the Mandaean Book of John

Browse Our Archives

Follow Us!


TRENDING AT PATHEOS Progressive Christian
What Are Your Thoughts?leave a comment
  • John MacDonald

    Interesting article. I was reminded of the Bridge Officer’s Test in Star Trek TNG where only after ordering close friend Geordi La Forge to make a repair in an area where he would be exposed to fatal doses of radiation did Troi pass the test.

  • Neil Brown

    Loved the article, both the parts I agreed with, and the parts I didn’t (both the parts I understood and the parts I didn’t….)
    I confess that I find the “self driving car problem” seems to be massively over-thinking the problem.
    A crucial rule for all road users is to be predictable – you aren’t the only actor and you cannot predict other actors unless they are predictable, and so you must also be predictable.
    So rather than trying to encode moral values, I’d much rather encode predictability and put research effort into sensor technology so that the predictions have high quality data to work with. If I can trust other actors to be predictable, I don’t need to be nearly so clever myself (and please, keep the human off the roads, unless they are in tanks like the rest of us).

    • Ankur Gupta

      I think the point you bring up about predictability is precisely the reason *why* we have to talk about ethics of driverless cars. In other words, let’s suppose that it’s easy to design a machine that operates in an ideal world with rules that no one is allowed to break. However, currently that is not the situation we face. We have to develop algorithms to help us along the journey towards 100% utilization of autonomous vehicles. Assuming that the problem is trivial in that ideal domain is somewhat of a cop-out.

      Further to the point is that even if we were able to assume a “closed” and idealized system (where either a car is automated, or it’s not allowed), you still need to be able to deal with other unexpected circumstances, such as debilitating weather or unexpected breakdowns of roads or wildlife interactions like deer hits or Godzilla. And, in the scenario that those unexpected events cause a decision to be made, we have to think about this issue carefully as above.

      • Neil Brown

        Thanks for your thought. I exactly disagree (though I’m not an expert, so it doesn’t mean much).
        Predictability is precisely the reason why we must NOT have “ethics” in driverless cars. If they attempt to behave “ethically”, then they are less likely to behave predictably (because in any non-trivial situation, ethical choice is based on perception, and different actors perceive different things).

  • John MacDonald

    Hi,
    I don’t regularly post on my Palpatine’s Way blog any more, but the fancy hit me yesterday and I did a post last night that readers of this blog may be interested in on Continental Philosophy’s critique of traditional Artificial Intelligence’s Cartesian assumptions of the essence of human awareness reducing to the subject as a thinking substance (res cogitans). The post is here: http://palpatinesway.blogspot.com/2019/09/continental-philosophy-and-critique-of.html There were a few problems with the post I did last night that I was alerted to today, but I think I fixed them!

    • John MacDonald

      One big takeaway I have from the study of Contemporary Continental Philosophy (postmodernism) and Cognitive Science is that the core assumption of modernism, the ground of human awareness being the res cogitans (thinking substance), is completely wrong. We are fundamentally in the world, disclosing the being of things with our involvement with things, and disclosing ourselves through the way we mirror the world.

      So, for instance, the being of a hammer as equipment doesn’t purely belong to the object, since a rock could serve the same function as a hammer, but we don’t regard the rock as equipment. Rather, being as equipmentality is in between subject (carpenter) and object (hammer), uniting them as a kind of copula.

      We not isolated thinking substances, as is shown with the cross translating of the various sensory areas of the brain which allows mirroring and self description metaphor (I’m boiling mad), as well as the projection of inner experience such as time onto the world allowing skillfully coping with things, such as the substancetality of substances (I encounter the table as a substance: one and the same thing “over time” and despite change).

      • John MacDonald

        Similarly, Heidegger points out a person is not a thing with properties, but a work in process. The woman is not, for instance a “carpenter thing,” but what it means for her to be a carpenter is constantly being dis-closed/revealed to her and those who know her in the activity of carpentry!

  • John MacDonald

    Harvard psychologist Steven Pinker shared this article today about why computer scientists are having such a hard time teaching computers to read: https://www.wired.com/story/adaptation-if-computers-are-so-smart-how-come-they-cant-read/?utm_source=twitter&utm_medium=social&utm_campaign=onsite-share&utm_brand=wired&utm_social-type=earned

    Here is an excerpt which, as a former elementary school teacher, really seems on point:

    All of these questions are easy for people. But no AI yet devised comes close—because each of these questions requires a reader to follow a chain of inferences that are only implicit in the story, and current techniques do not carry out inference in this sense. What is implicit is largely outside their scope. Such chains of reasoning often demand that the reader put together background knowledge about people and objects and, more generally, about how the world works. No current system has a broad enough fund of general knowledge to do this well.

    So, even though people often think all Philosophy reduces to Analytic Philosophy and Cognitive Science, the contemporary continental and existential traditions being mere literary curiosities, there is a very real sense that Heidegger’s major question from Being and Time is still relevant today, namely, how are we to understand out being-in-the-world?

    • John MacDonald

      For example, when children (or anyone) are reading, they unconsciously employ the 4 R’s strategies: Retell Relate Reflect (sometimes Review). These strategies are taught explicitly as metacognition, with the goal being for students to internalize them and be aware of these strategies, so that when meaning making breaks down, students know which strategies to try to employ.

      For instance, effective programming by computer scientists to get computers to read effectively is going to have to deal with the following metacognitive structures involved in good reading:

      Retell

      Some things to write about:

      Retell what you have read.
      Summarize what you have read.
      Describe the major ideas.
      Describe the most significant and relevant parts.
      Describe parts that you find confusing, attempt to
      work out their meaning through your writing, and
      formulate questions about them for discussion.

      – Some possible Retell prompts:

      This was about …
      I noticed that …
      The most significant ideas/event was …
      An event that I especially liked was …
      I particularly liked/valued/enjoyed …
      An idea that captured my interest was …
      Words that I found particularly effective were …
      An idea that inspired me was …
      The author’s style is …
      I didn’t understand this part because …

      Relate

      Some things to write about:

      Make connections between the text and your own life
      (e.g., personal experiences, feelings), other texts (e.g.,
      books, music, films, television programs, websites), or
      the world at large (e.g., current events).
      Relate your thoughts about the text to your own life,
      giving examples.

      – Some possible Relate prompts:

      Something I identify with is …
      What I found especially meaningful or significant was

      This text relates to my life in the following ways: …
      This book, compared with others by the same author, is …
      The text is similar to my experience when …
      What is described in this text is very much like what is
      happening today in …

      Reflect

      Some things to write about:

      Why this text is important to you
      The author’s conclusions· Your judgement about the
      text/author’s conclusions· Your opinions about the
      ideas/events in the text
      Your opinions about the style of this text compared
      with the style of other texts on the same topic
      Your judgement about the credibility of the text
      Whether you feel the text is current and/or accurate
      Whether reading the text has changed the way you
      feel or think about something, and if so, how
      Any relevant questions you have
      Anything about the text that still confuses or
      frustrates you
      How reading this text has or has not helped you meet
      the learning goals you set for yourself
      Any new goals you are setting for yourself

      -Some possible Reflect prompts:

      The passages that I find most meaningful are …
      because …
      I think that …
      I wonder why or what if …
      I realize now that …· How …?
      This author’s style differed from … on the same topic
      in the following ways …
      This information is very different from what I read in …
      What is the author trying to make me think/do?
      I wonder why I feel so …
      Reading this text changed the way I think and behave
      because …
      A question raised in my mind is …
      My predictions about this text were …
      Something that pushed my thinking was …
      This text was a challenge for me because …
      I learned that …
      I changed the way I think/feel about … because …
      I think that I will investigate other points of view
      because …
      An idea that inspired me was … because …
      The author wrote the story with an underlying
      message that is … because …
      I would recommend this book to a friend or to a
      reader who is especially interested in …

      3 R’s Reference:
      Ministry of Education. A Guide to Effective Literacy Instruction. Vol.2. Toronto: Queen’s Printer, 2006
      .

      There are other metacognitive approaches that go into teaching reading, but this is an example of the core elements computer scientists are going to need to program in order to get computers to be competent readers