Revision db474adb9b1414d1a3190d758159d3a9cb7665af (click the page title to view the current version)
Changes from db474adb9b1414d1a3190d758159d3a9cb7665af to 659de93641c9a50cf24a325ca89d58671e4b5021
---
title: Questions for Discussion
---
**Under construction**
Please refer to the [Reading List]() for links and references.
# What if the machines take control?
The second compulsory assignment is a verbal presentation or
panel debate on a given question or claim.
In a **debate**, two students prepare a position and two students
prepare a counter-position.
In a **presentation** two students illuminate the question or
position from different angles.
In both cases, the students should use the literature to prepare
their position.
We look for a more scholarly approach than a casual pub conversation.
R&N:51 call this a design flaw.
A number of topics are presented below. You may propose your own.
Groups of one or three are permissible if the numbers do not work
out, but please avoid it if you can.
Can it be avoided under *bounded rationality*?
**Deadline** Form groups and select topic by Sunday 27 February.
Note that design is an evolution, and final goals cannot be
predicted.
# Debates
Note that most debate questions can also be made as a presentation.
## AI is a threat to human life and society.
**Reading** (in order of priority)
1. R&N (Sec. 1.5).
1. Schaathun (2022)
1. [Schaathun (2022)](https://www.sciencedirect.com/science/article/pii/S0142694X22000102)
1. Dick: *Do androids dream of electric sheep?*
1. Simon (1996:Ch. 6)
# Can machines think?
Debate: two students to argue *yes* and two students to argue *no*
## Can machines think?
**Reading** (in order of priority)
1. Epstein (2009:Chapters 1 and 3)
2. Weizenbaum (1976:Chapter 10)
3. Heinlein: *The Moon is a harsh Mistress*
# Eastern Philosophy
## Is the human mind any different from a computer?
**Reading** (in order of priority)
1. Epstein (2009:Chapters 1, 3, 4)
2. [Lucas 1961](https://www.cambridge.org/core/journals/philosophy/article/minds-machines-and-godel1/727219EDEB5DD3679E56CF3D335C90C1)
2. Weizenbaum (1976:Chapter 10)
# Other Topics
## What are the main challenges in AI Ethics?
**Reading**
1. [Jobin et al 2019](https://arxiv.org/pdf/1906.11668.pdf)
1. R&N Chapter 28
## How can we avoid the machines taking control?
R&N:51 call it a design flaw, if machines take control contrary
to human interest.
Is this really possible to avoid?
According to Simon note that design is an evolution.
Final goals are incompatible with our limited ability to divine about
the future.
**Reading** (in order of priority)
1. R&N (Sec. 1.5).
1. [Schaathun (2022)](https://www.sciencedirect.com/science/article/pii/S0142694X22000102)
1. Dick: *Do androids dream of electric sheep?*
1. Simon (1996:Ch. 6)
## Eastern Philosophy
This module is biased towards Western philosophy,
drawing heavily on ideas developed in a long tradition of thought,
rooted in Ancient Greek philosophy and the Judeo-Christian
worldview. This is as it has to be. We are who we are, and
there is no other tradition where the module convener would
be qualified to convene the module.
However, a meeting with other traditions of thought would be very
valuable, if someone would be up for the challenge.
This is not an easy task, because we should be looking for
scientific rigour, and not a superficial approach.
However, if someone is familiar with philosophical traditions
of other cultures, you are welcome to take any question suggested
above or otherwise discussed in the module, and recast it within
your own cultural frame?
How is the relationship between Man and Machine viewed in Indian
philosophy?
How is intelligence viewed in Chines thought?
How is intelligence viewed in Chinese thought?
What constitutes reason and rationality in Persian tradition?
There are plenty of questions and plenty of viewpoints,
and note that I have no idea about the answers what-so-every.
Curiosity is all I can contribute.
Note that the task is scholarly. As a minimum, one has to
point to scholarly references when citing the great thinkers.
I do not expect references in a language I can read, but they
need to be disclosed honestly and accurately.
