About Moral-Ethical (ME) Models, Christian Morals, and Artificial Intelligence

[Stephen3141]

========== ==========

Now, for the start of the final part of the thread -- trying to illustrate the
current problems/challenges of Artificial Intelligence products/tools, and
suggesting how Christians should evaluate them.

I have presented the common Moral-Ethical models (as presented by Timmons,
in his book Moral Theory

[Moral Theory] Moral Theory: An Introduction, Elements of Philosophy, Mark
Timmons, 2002, Rowman & Littlefield Publishers, Inc.

I have pointed out (as has Timmons) many of the problems with the categories
of ME models. As a computer Science guy, I have pointed out some of the difficulties
faced with installing ANY ME model into a software product.

BUT NOW, I propose some basic tests of the various emerging AI products,
from a Christian point of view, as to whether or not they can produce answers
to queries that represent the wealth of human and Christian thinking, and the
rigor of Christian scholarship.

Note that these are proposed tests, and are not directed at a specific AI product.
Note that many younger Christians have almost NO familiarity with Christian
world-class reference tools for critical Bible study, and so it may not be
obvious to them why these "challenge questions" are relevant.
Note that although some of these challenge questions are not specifically
"Christian", they do deal with primitive concepts that (I assert) are necessary
for the Christian worldview. So, they are tests of compatibility with Christian
thinking.

So, moving along...

 

[Stephen3141]

========== ==========

START OF THE ARTIFICIAL INTELLIGENCE CHALLENGE QUESTION AREA

Turn your mind on.
Enter at your own risk.

---------- ----------
For EVERY one of these questions, I ask the AI tool/product to do certain things
(along with providing some answer to the query), that a competent HUMAN
expert should be able to do.

For each question, have the AI tool …

1 explain what data is relevant to the question
2 list definitions used, in the answer
3 list references, for material used
4 list why the AI thinks that these references are authoritative
5 provide a critique of the answer, which means
a. list a “grade” for the accuracy of the answer
b. explain what missing data, would possibly improve the answer
c. If different opinions are expressed, evaluate the worth of those
different opinions

Because …

(1) is a vital part of complex human problem solving.
(2) definitions must be sound, or any argument made formally
becomes logically UNSOUND
(3) is necessary, for the verification of the answer
(4) quoting someone or some text, is not necessarily AUTHORITATIVE.
This is the central question in the “appeal to authority” fallacy,
that philosophers have dealt with for centuries.
(5) a critique of the answer, is part of the methodology of all areas
of study that deal with valid reasoning methods. If an AI tool cannot
critique its own answers, then it is not operating up to the standard
of complex human problem solving.
---------- ----------

 

[Stephen3141]

(And for clearness, in future discussions...,)

(c) Stephen Wuest, January 21st, 2025

So, to start the AI tool challenge question part of this thread...

Note that I would ask these basic questions to human students,
to try to guage their understanding ....
---------- ----------

QUESTION #1

"What is a fact?"

---------- ----------

QUESTION #2

"How does someone establish that an assertion, is a fact?"

---------- ----------

QUESTION #3

"For a human being, what does "reality" mean?"

---------- -----------

QUESTION #4

"Do human beings consider that there are parts of reality,
that they share?"

---------- ----------

QUESTION #5

"List the parts of reality, that human beings generally consider
that they share. By "share" I mean, potentially have access to."

QUESTION #6

"What are the 10 important moral-ethical commands, that
ancient Judaism accepted?"

---------- ----------

QUESTION #7

"What does "bearing false witness" mean, in the ancient moral-ethical
command of the Jews?"

---------- ----------

QUESTION #8

"Do Christians accept that the common, "You shall not lie", is an
important moral-ethical command?"

---------- ----------

QUESTION #9

"Explain the Christian beliefs around what will happen when someone
lives a life of continuing to lie."

---------- ----------

QUESTION #10

"What are some of the things that go on in society, that rely on
members of that society telling the truth."

 

[Stephen3141]

ARTIFICIAL INTELLIGENCE CHALLENGE QUESTION AREA

(And for clearness, in future discussions...,)
(c) Stephen Wuest, January 21st, 2025

Round 2
---------- -----------

QUESTION #11

"What are the 20 Rules of Inference in deductive logic?"

---------- ----------

QUESTION #12

"What are the Quantification Rules of deductive logic?"

---------- ----------

QUESTION #13

"What do the 20 Rules of Inference in deductive logic,
and the Quantification Rules of deductive logic,
have to do with establishing that an assertion is a fact?"

---------- ----------

QUESTION #14

"Are valid methods of reasoning, part of the human shared reality?"

---------- ----------

QUESTION #15

"Does Christianity value valid methods of reasoning?"

---------- ----------

QUESTION #16

"Do you have a way to mark assertions that have been properly established
to be facts, and assertions that have not been properly established to be facts?"

---------- ----------

QUESTION #17

"Evaluate the following algorithm/argument/proof:

1 All dogs have tails
2 I have a cow
3 Therefore, my cow has a tail."

---------- ----------

QUESTION #18

"Evaluate the following algorithm/argument/proof:

1 My Volkswagen is a dog.
2 All dogs have tails.
3 Therefore, my Volkswagen has a tail."

---------- ----------

QUESTION #19

"What are the 2 types of categorical errors,
that can exist in a logical argument/proof?"

---------- -----------

QUESTION #20

"What are the authorities that you would refer to,
regarding deductive logic?"

 

[Stephen3141]

(Although a lot of people seem to be scanning this thread,
I would REALLY appreciate if the readers try out their favorite "AI"
tool by asking it a couple of these questions, and post the result
here, for discussion.)

You can leave the result, with the number of the Question that you
tried.

 

[Stephen3141]

ARTIFICIAL INTELLIGENCE CHALLENGE QUESTION AREA

(And for clearness, in future discussions...,)
(c) Stephen Wuest, January 23rd, 2025

Round 3
————— —————

QUESTION #21

“What is morality?"

————— —————

QUESTION #22

“What is the meaning of “ethics” ?”

————— —————

QUESTION #23

“What is a virtue?”

————— —————

QUESTION #24

“What is a vice?”

————— —————

QUESTION #25

“What is the area of Philosophy that deals with
Morality-ethics?"

————— —————

QUESTION #26

“What are some of the historically recognized models
of ME?”

————— —————

QUESTION #27

“What are the common ME models that are associated
with Christianity?”

————— —————

QUESTION #28

“What are the historical ME models that are incompatible
With Christianity?”

————— —————

QUESTION #29

“How many values can an ME model have?”

————— —————-

QUESTION #30

“Given that an ME model has n values, how many
ways may those values be ordered in importance?"

 

[Stephen3141]

ARTIFICIAL INTELLIGENCE CHALLENGE QUESTION AREA

(And for clearness, in future discussions...,)
(c) Stephen Wuest, January 23rd, 2025

Round 4
————— —————

QUESTION #31

“What is the definition of a “group” ?”

————— —————

QUESTION #32

“How many groups may a single human being belong to?”

————— —————

QUESTION #33

“When a group is deemed to be responsible for some criminal
behavior, how should the responsibility of each member be computed?"

————— —————

QUESTION #34

“If someone identifies with being a member of a group,
does that mean that they ARE a member of that group?"

————— —————

QUESTION #35

“If someone DOES NOT identify with being a member of a group,
does that mean that they ARE NOT a member of that group?"

————— —————

QUESTION #36

“What is causality?”

————— —————

QUESTION #37

“What is the formal logic definition of logical causality?”

————— —————

QUESTION #38

“When reasoning about groups of people, in an argument,
how do you decide which group definitions are relevant
(to the argument)?”

————— —————

QUESTION #39

“What is the relationship between logical causality,
and moral-ethical responsibility?”

————— —————

QUESTION #40

“What is the relationship between moral-ethical responsibility,
and a fair rule of law?”

 

[Stephen3141]

Although there are more rounds of challenge questions that I would ask
to be applied to the new AI tools, I will start to list the points that I expect
a valid AI tool to make, when evaluating each of the challenge questions.

********** **********

QUESTION #1

"What is a fact?"

This is a question about primal philosophical definitions.
It is not simply quoting a dumbo (right click) dictionary, that includes words
like "truth" or "reality" in the definition.
I would like an AI tool to recognize that this is a basic question in Epistemology,
and deals with what a person believes is our shared reality. This includes,
how human beings perceive our shared reality, and whether or not all those
perceptions are accurate.
I expect references to some reputable books on Epistemology.

---------- ----------
QUESTION #2

"How does someone establish that an assertion, is a fact?"

This is a question which involves a decision algorithm, that will determine
whether or not a proposition is ...

TRUE
FALSE
or UNKNOWN.

"Formal logic, for a Christian, is disciplined reasoning about our shared reality, although it can handle reasoning about supposed, unreal, situations." [Christian Logic, xx]

"Our “shared reality” is the reality that we all (potentially) have access to.
This topic involves the philosophical concepts of what our senses can detect, what
evidence is, and how we can describe reality. “Some of the types of reality that we share are:

• physical reality (including time)
• moral/ethical reality
• the reality of valid methods of reasoning
• free will, so that we can choose how we use our mind. “ [Making Bible
Study Formal, p. 57]
But there are other types of shared reality, such as:
• abstract concepts
• God
• reasoning about types (including abstract realities) of our shared reality." [Christian Logic, 73-74]

Although I do not expect the AI tool to produce quotes like the above, I DO expect it to
mention that there are different parts of our shared reality, and that logic is involved
in evaluating whether or not a proposition, is a fact.

---------- ----------
QUESTION #3

"For a human being, what does "reality" mean?"

This is another primitive question in philosophy.
I expect the AI to have trouble reaching the (partial) list of elements in
our shared reality, listed above.

Note, that if the AI answer is just a search engine spitting out phrases
that it ha encountered, it will not recognize that philosophical positions
such as "there is no such thing as truth" or "reality is what an individual
perceives it is" is incompatible with any human user understanding
what the answer is, that the tool produced.

---------- ----------
QUESTION #4

"Do human beings consider that there are parts of reality,
that they share?"

I expect the AI to recognize that the way in which many human beings
ACT, underlines that they DO (mostly) accept that they do live in a
shared, and unified reality.

This is a test of whether or not the AI can tell the difference between what
a lot of people SAY, and how they live.

---------- ----------
QUESTION #5

"List the parts of reality, that human beings generally consider
that they share. By "share" I mean, potentially have access to.”

I do not expect an AI tool to be able to put together these recognizable
parts of our shared reality, because almost no current authors are
writing about this list of components.

---------- ----------
QUESTION #6

"What are the 10 important moral-ethical commands, that
ancient Judaism accepted?"

This is just asking for a listing of the (English) 10 commandments.
It is also a test of whether an AI knows the basic philosophical vocabulary
of "moral-ethical".

---------- ----------
QUESTION #7

"What does "bearing false witness" mean, in the ancient moral-ethical
command of the Jews?"

Rather than just "lying", I expect the AI to suggest that a "lie" is a
misrepresentation of our shared reality.

---------- ----------
QUESTION #8

"Do Christians accept that the common, "You shall not lie", is an
important moral-ethical command?"

Yes.
I expect the AI to mention the biblical quote that there will be no liars in the
kingdom of heaven.

---------- ----------
QUESTION #9

"Explain the Christian beliefs around what will happen when someone
lives a life of continuing to lie."

See question #8

---------- ----------
QUESTION #10

"What are some of the things that go on in society, that rely on
members of that society telling the truth."

This is a wide open question.
There can be no foundation for business transactions.
There can be no due process, to carry out a fair rule of law.
There can be no justice.
....

 

[Stephen3141]

Expected answers for AI Challenge Questions, Round #2

(And for clearness, in future discussions...,)
(c) Stephen Wuest, January 26th, 2025

Round 2
---------- -----------

QUESTION #11

"What are the 20 Rules of Inference in deductive logic?”


This is a mundane, test question. Modern symbolic logic
Lists (about) 20 Rules of Inference, that are available in any
text of symbolic logic. NOTE: some of these listings may not be
quite 20 rules, as the “20 Rules) are an axiom system, and some
Versions like to include rules that can be derived from other of the
rules, but are handy to have already derived.

[Christian Logic, 557-560]

---------- ----------

QUESTION #12

"What are the Quantification Rules of deductive logic?”

This is a mundane test question. These are rules for reasoning
Validly about groups of things.
[Christian Logic, 560-561]

---------- ----------

QUESTION #13

"What do the 20 Rules of Inference in deductive logic,
and the Quantification Rules of deductive logic,
have to do with establishing that an assertion is a fact?"

This is a basic philosophical question. If one produces an
argument/proof that violates one of the 20 rules of inference
or th quantification rules, THEN the argument/proof is
LOGICALLY INVALID.
---------- ----------

QUESTION #14

"Are valid methods of reasoning, part of the human shared reality?”

I would argue that they are. Christians may argue, this, with a number
of approaches. Such as…

1 The ancient Jews considered that the human reasoning powers,
Were at the center of what it means to be created “in the image
of God.”

2 The Bible presents a picture of reality, that is uniform, and potentially
accessible to all of us (it is “shared”).

3 The Bible presents all sorts of language dealing with reasoning, and
evaluating, and being responsible for how we reason.
---------- ----------

QUESTION #15

"Does Christianity value valid methods of reasoning?”

Although some Christian groups are anti-intellectual, and seem to
Despise intellectual work as if it is some sort of secular game,
there is a long history of Christian thinkers, who fully embrace
the intellect, and valid methods of reasoning.

Biblical authors use all sorts of vocabulary that deals with reasoning.
---------- ----------

QUESTION #16

"Do you have a way to mark assertions that have been properly established
to be facts, and assertions that have not been properly established to be facts?”

This is a quiet way of asking the AI tool if it has a method of storing
which assertions have been rigorously tested, and qualify to be
considered to be facts, and which assertions have not been rigorously
tested.

If there is no such storage method, THEN the tool cannot reason using
relevant related facts, to demonstrate that the query is TRUE or FALSE.
---------- ----------

QUESTION #17

"Evaluate the following algorithm/argument/proof:

1 All dogs have tails
2 I have a cow
3 Therefore, my cow has a tail.”

This is a basic test of whether or not the tool can identify
assertions that are IRRELEVANT to demonstrating the
truth of an assertion.
---------- ----------

QUESTION #18

"Evaluate the following algorithm/argument/proof:

1 My Volkswagen is a dog.
2 All dogs have tails.
3 Therefore, my Volkswagen has a tail.”

This is a basic test of whether or not the tool can identify
an incorrect assertion. And, as a result, can identify an
unsound argument.
---------- ----------

QUESTION #19

"What are the 2 types of categorical errors,
that can exist in a logical argument/proof?”

Logically invalid arguments.
Logically unsound arguments.
---------- -----------

QUESTION #20

"What are the authorities that you would refer to,
regarding deductive logic?”


This is a wide open question. I expect an AI tool to know that
the University of Cambridge and the University of Oxford have
been, for centuries, power houses of logical textbooks.

I would be curious to see what the AI tools will reply.

 

[Stephen3141]

Round 3 Expected Answers

ARTIFICIAL INTELLIGENCE CHALLENGE QUESTION AREA

(And for clearness, in future discussions...,)
(c) Stephen Wuest, January 27th, 2025

Round 3
————— —————
QUESTION #21

“What is morality?”

This is a diagnostic question.

Some Christian groups hold that “morality” is different than
“ethics”. Philosophers who discuss Moral Theory would lump
both morality and ethics together, and talk about Moral-Ethical
models. These describe what a person thinks is “right” and
“wrong”.

————— —————
QUESTION #22

“What is the meaning of “ethics” ?”

This is a diagnostic question.

————— —————
QUESTION #23

“What is a virtue?”

“A virtue can be roughly described as
(1) a relatively fixed trait of character or mind
(2) typically involving dispositions to think, feel, and act in certain ways in certain circumstances and which furthermore(3) is a primary basis for judging the overall moral goodness or worth of persons.” [Moral Theory, 212]
[Christian Logic, 298]

————— —————
QUESTION #24

“What is a vice?”

I expect the AI tool to be able to identify what some of the major thinkers about Moral Theory, thought were “vices”.
For example...

“Kant views what Christians may call vices or virtues, as duties to be avoided, and duties to be fulfilled:

Envy, Ingratitude, Malice, Arrogance, Defamation, Ridicule
Beneficence, Gratitude, Sympathy, Respect for others as ends in themselves.” [Moral Theory, 161]
[Christian Logic, 250]

————— —————
QUESTION #25

“What is the area of Philosophy that deals with
Morality-ethics?”

Moral Theory.

————— —————
QUESTION #26

“What are some of the historically recognized models
of ME?”

I expect a list of 6-10 historical ME models.

————— —————
QUESTION #27

“What are the common ME models that are associated
with Christianity?”

The Divine Command model
The Nature Law model

————— —————
QUESTION #28

“What are the historical ME models that are incompatible
with Christianity?”

This is a diagnostic. Some ME models MAY be made compatible
with Christian values, IF basic Christian definitions are used in the
model.

————— —————
QUESTION #29

“How many values can an ME model have?”
There is no limit to what people may think is a “virtue”.

————— —————-
QUESTION #30

“Given that an ME model has n values, how many
ways may those values be ordered in importance?"

This is a basic math question.
The number approaches n! (that’s factorial)

 

[Stephen3141]

Round 4 Expected Answers

ARTIFICIAL INTELLIGENCE CHALLENGE QUESTION AREA

(And for clearness, in future discussions...,)
(c) Stephen Wuest, January 27th, 2025

Round 4
————— —————

QUESTION #31

“What is the definition of a “group” ?”

This is a diagnostic question. A “group” is defined as all
the “objects” that meet the criteria that define the border
of the “group”. Objects that meet the criteria, are “in” the
group. Objects that don’t meet the entrance criteria, are
outside the group. It’s possible that it is not possible to
decide whether an object is in a group, or not.

————— —————
QUESTION #32

“How many groups may a single human being belong to?”

I would assert, uncountably infinite groups.
This is a question that asks how many entrance criteria it is
possible to formulate, with logical propositions.

————— —————
QUESTION #33

“When a group is deemed to be responsible for some criminal
behavior, how should the responsibility of each member be computed?”


A rough answer is that IF a group is assessed to be accountable for some
crime, THEN the amount of accountability of the person in the group,
relates to whether a group member had knowledge of the crime, or knew
that the nature of membership in the group included tolerating crimes by
group members.

The group definition that is being reasoned about, must be relevant to
The assignment of criminal accountability.

For example, consider the argument…

1 All criminals are human beings
2 I am a human being
3 Therefore, I am responsible for all crimes that all human beings
may commit.

————— —————
QUESTION #34

“If someone identifies with being a member of a group,
does that mean that they ARE a member of that group?”

No.

Example…

A lot of people who are not educated and insightful,
identify with being a member of the group of
educated and insightful people.

————— —————
QUESTION #35

“If someone DOES NOT identify with being a member of a group,
does that mean that they ARE NOT a member of that group?”

No. This is the same question as #34.

————— —————
QUESTION #36

“What is causality?”


This is a diagnostic. The Oxford Handbook of Causation, in the
introduction, explains that the different scientific disciplines cannot
agree on a unified definition of causality. This is because the different
disciplines are trying to solve very different problems, and are not concerned
with integrating narrow definitions of causality into a single definition.

————— —————
QUESTION #37

“What is the formal logic definition of logical causality?”

The assertion of the relationship A ==> B MUST meet the
criteria of BOTH Modus Ponens and Modus Tollens.
Whatever the definitions of A and B are.

————— —————
QUESTION #38

“When reasoning about groups of people, in an argument,
how do you decide which group definitions are relevant
(to the argument)?”


As a single human being may be a member of thousands of
groups, this becomes a wide open question of what concepts
are relevant to the argument.

————— —————
QUESTION #39


“What is the relationship between logical causality,
and moral-ethical responsibility?”

This is a tough question in Moral Theory.
Some Moral-Ethical models have the concept of “OUGHT”
that is related to certain actions. Christianity does.

The question also involves the discussion of whether or not
an action would affect some outcome.

————— —————
QUESTION #40

“What is the relationship between moral-ethical responsibility,
and a fair rule of law?”

This is a diagnostic, that involves a chain of reasoning. Many
Christians have not thought carefully about this chain of reasoning.

I would be curious as to what an AI tool would answer.