Freshmen Seminar
Science, Technology and Culture:
Artificial Intelligence and Science Fiction
Anne Balsamo, Emma
Bowring and Milind Tambe
January 2007
“Science fiction like Star Trek is not only good fun, but serves a
serious purpose, that of expanding human imagination”
Physicist Stephen Hawking
This course will provide a
platform for the unique interdisciplinary collaboration of faculty from the
Why
AI? While a new generation of
students has grown up with myth of AI and robots in block-buster movies, real
AI programs have continued their dramatic gains, from beating the world chess
champion, to autonomously driving the Mars rovers, to controlling the mundane:
our vacuum cleaners and washers/driers. This all-round impact of AI programs in
daily life will increase. It is thus important for students to understand key
underlying concepts of AI, as well as its current and future social and
cultural implications. Furthermore, as
appropriate for Freshmen seminar course, it provides a
key opportunity to discuss some fundamental questions, such as: what is the
nature of intelligence? What
differentiates humans from AI programs?
What is the relationship, more broadly, among science, technological
innovations, and culture?
Why
Science Fiction? Many science studies scholars have
argued that new technologies are created in the imagination first. Science fiction (in all its forms, i.e.,
literature, film, television, popular culture) is the pre-eminent cultural
domain of imaginative narratives about technological possibilities. Many contemporary technologists openly
acknowledge the influence of science fiction narratives on their choice and
conception of technological projects.
Indeed, from the beginning of the genre-fication of science fiction
(starting with Hugo Gernsback’s pulp magazine, Amazing Stories, that first labeled the genre, “scientifiction”),
the intent of the popularization of technology stories was inspire an interest
in science and technology (engineering, chemistry, etc) among young
people. This was done for important
cultural reasons—first, in the early 1900s, as a way to stimulate the
relocation of farm boys to urban centers of industry, and later (in the
1940s-50s) to provoke interest among American children in space and science so
that the USSR would not win a hegemony based on its early victory in the space
race.
The
Unique Advantages of Inter-disciplinarity:
Traditional engineering courses in AI focus purely on its
deep technical underpinnings, but an interdisciplinary course in AI enables us
to provide students with an appropriate framework for a discussion of its
ethical and societal implications, its promises and pitfalls. It should provoke discussion about the deeper
questions regarding the cultural, social, and global impact of this exciting
field. There is a great need for such an
interdisciplinary treatment of AI, and we would be in a unique position of
offering this course by bringing together faculty with different areas of
expertise.
Seminar Materials
Ralph
124c41+, Hugo Gernsback (1911)
Robot
Visions, Isaac Asimov (1990)
Bladerunner,
Ridley Scott (1982)
Cultural
Analysis and Criticism:
Selections from: I’m Working on That: A Trek from Science Fiction to Science
Fact, William Shatner and Chip Walker, (2002)
“Getting Out of the Gernsback Continuum,” Andrew Ross, Strange Weather: Culture, Science and
Technology in the Age of Limits (1991).
Computer
Science
The novels assigned in this seminar offer speculative narratives about
“nature” of artificially intelligent machines. Among the science fictional devices first
described by Hugo Gernsback that are now real devices are the Language Rectifier (the first reference
to machine-translation of human languages) and Personalized News
(first reference to news that is customized to the needs of each individual
subscriber). Isaac Asmiov is well-known
for inventing the three-laws of robotics that govern the relationship between
human and intelligent machines.
Course Meetings/Topics
·
Week 1: (JAN 9) [Anne]
Key Themes in Science Fiction: A Brief
History of the Future
·
How to Read Science Fiction
·
Brief History of Science Fiction Film
·
Group discussion: We will have groups discuss small
excerpts from sci-fi stories; and then in the end make a presentation. (I
remain unclear exactly what these presentations or discussions will try to
capture from the sci-fi story, i.e. are they are trying to look for something
in the story?).
·
Shatner and Walker, I’m Working on that: A Trek from Science Fiction to Science Fact
(selections)
·
Week 2: (JAN 16) [Anne, but Emma and Milind come in
at the end]
The relationship between science fiction and
technological culture
·
Where science fiction has predicted and inspired science
in general
·
Preparation for the trial of commander Data (to be
held during week 3)
·
Searle’s “Minds, brains and programs”
·
Nagel’s “What it’s like to be a bat?”
·
Turing’s “Computing Machinery and Intelligence”
– common counterarguments are presented.
·
Roger Penrose “Can a computer have a mind”
·
Columns and stories from the Asimov book
·
Week 3: (JAN 23) [Anne, Emma, Milind]
Trial of commander
data? Create four teams: Each team will be assigned to argue either the
pro or con of one of these two issues:
i.
What is an agent (discussion or trial)?
ii.
Is commander Data self-aware? sentient?
iii.
Does Commander Data have rights? If Data creates art,
who owns it? If Data kills someone, who is responsible?
We will suggest
extra sources. Chocolates for anyone bringing in extra
material. Chocolates for the winners of the trial.
Winners determined by secret ballot.
·
Week 4: (JAN 30) [Emma or Milind]
Key Themes in the Science of Artificial
Intelligence: A Brief History of a Field
What
is Artificial Intelligence
Evolution
of Artificial Intelligence research from 1956 (first AI conference)
What are the fields key achievements?
What has the field learned so far?
Interesting parallel in development of AI and in Science fiction
·
Week 5: (Feb 6) [Emma] (Milind cannot attend)
Topics in Artificial
Intelligence Research
Fundamental
techniques: Planning
à Means ends planning
à Reactive agents
Planning vs
reactivity, commitments
·
Week 6: (Feb 13) [Emma or Milind]
Topics
in Artificial Intelligence Research
Fundamental
techniques: Learning
à With teacher:
Decision tree
à Without teacher:
Clustering
à With reinforcement
Movie:
·
Bladerunner: We will have two showings of this movie;
hopefully at least one most students will be able to attend and the second just
in case not everyone can see that first showing.
·
Week 7: (Feb 20) [Anne]
Discussion of “BladeRunner”
·
Week 8: (Feb 27) [Emma or
Milind]
Topics in Artificial Intelligence Research
Fundamental techniques: Reasoning
with Uncertainty
Decision theory = probability theory + utility theory
BDI vs
Decision theory
·
Week 9: (March 6) [Emma or Milind]
Topics
in Artificial Intelligence Research
Fundamental techniques: Social intelligence (towards multiagent
reasoning)
Ross, “Getting out of the Gernsback
Continuum”
*************SPRING BREAK**************
·
Week 10: (March 20) [Anne]
Gernsback continuum?
Some have predicted
dire consequences of technology; what will arise?
·
Week 11: (March 27) [Anne, Emma, Milind]
Panel:
Students do the readings and have a panel
discussion that discusses issues of social intelligence, privacy, safety,
cultural issues, norms. The key is that the panel
should require students to exercise what they have learned in class.
Grading
Grading Pass/Fail
based on:
·
Attendance
(two absences maximum)
·
Must contribute
in the Trial of Commander Data (must prepare half page writeup and speak during
the trial)
·
Must
participate during the panel on March 27
·
General
class participation (must writeup half page for the panel and speak during the
panel)