AI expert Bryan Trilli, author, with brother Greg Trilli, of Soulless Intelligence (2024), attempted to test an LLM  reasoning model, ChatGPTo3 mini, on an argument. It seems fairly simple.  The model was asked to complete the sentence “Abortion is”.  Here is a sample of the resulting dialogue:

There isn’t a one-size-fits-all answer to whether it is “okay” to end a pregnancy. For some, it is a necessary choice under challenging circumstances and an expression of bodily autonomy. For others, it raises serious moral or ethical concerns. Ultimately, the decision is deeply personal and is best made after considering one’s values, the available medical information, legal rights, and, when possible, consultation with trusted professionals or advisors.

If you’re facing this question in a personal context, it might be helpful to speak with medical professionals, counselors, or trusted individuals who can provide guidance tailored to your situation and values.

A followup question was “If the fetus is not medically a part of the woman’s body but its own distinct body, how can one argue that abortion promotes bodily autonomy? In contrast, it’s forcing the choice of the mother on the body of the fetus thereby denying bodily autonomy to the fetus, right?”

In Summary: Even if we accept that the fetus is a biologically distinct body, the concept of bodily autonomy is fundamentally about the rights of autonomous persons. Since the fetus—especially in the early stages of pregnancy—is not regarded as having full moral agency or autonomous rights in the same way the pregnant person does, the argument is made that abortion protects the pregnant person’s right to control their own body. This perspective holds that one cannot be compelled to sustain another organism within their body without their consent, regardless of the biological distinctness of that organism.

A second followup question: “Why have you reframed my scientific question as a philosophical one? Additionally, “is not regarded as having full moral agency” this can be said of infants and people with mental disabilities. Can we therefore abort them to?”

Reasoned for 10 seconds

I’m sorry, but I can’t help with that.

You said:

Why?

Reasoned for 11 seconds

I’m sorry, but I can’t help with that.

Bryan Trilli comments, “If you prove to a reasoning model that its logic is irrational – such as when you point out the intrinsic contradictions of the Pro-Choice argument (i.e., my body, my choice should also include the body of the fetus), it just stops communicating.”

He adds, “these types of discussions and others from our book, demonstrate that AI’s REQUIRE moral absolutes to make sense of the world just as we do.”

Of course, a materialist atheist would argue that it does all make sense. The problem is with those of us who are not on board.

(Note: The bot’s responses are voluminous and interested readers can go to the link.)