Chapter 12 Homework Reading selections are grouped into clusters

Unlock access to all the studying documents.

View Full Document

Chapter 12

Artificial Intelligence

CHAPTER S

UMMARY

Chapter 12 concerns emerging artificial intelligence and its implications for humanity, present and future. The

chapter consists of three sections. The first is devoted to AI’s breaking free of evolution. Behrens and Rosen

ground the discussion with an opening piece on the legend of Prometheus, which raises important questions for the

chapter as a whole: Are there, should the be, limits to human knowledge? What are the consequences of

intellectual ambition?

This chapter invites students to contemplate the once unthinkable: the “mind” of a machine. Reading selections are

grouped into clusters. First come selections on the prospect and implications of humanity’s breaking free of

biological evolution through the design and application of a new, silicon-based intelligence. George Luger reviews

“The Legacy of Prometheus,” the myth of the Greek god who stole fire from Olympus and gave it to humans, making

possible all subsequent technology. In an online interview, historian Yuval Harari questions the premise that our

current bodies represent the pinnacle of human evolution. Tech writer Carolyn Mathas sets in opposition the views of

The chapter’s second cluster of readings opens with a more neutral view of what Dominic Basulto regards as a likely

“hybrid evolution” of silicon and biology. Three selections then consider the effects of AI on work. Scott Santens

investigates why “Robots Will Take Your Job”; an economist reviews a book that insists humans possess irreplaceable

skills and intuitions; and an article from The Economist magazine explores work-related anxieties concerning

automation. In the chapter’s final selections, we include a review of HBO’s hit series Westworld, the story of a Wild

West theme park populated by robotic actors. In a creepy, fascinating twist, the robots are more appealing and humane

The Legacy of Prometheus (p. 371)

George Luger

UMMARY

Luger is a computer scientist, both a scholar involved in AI research, and a textbook writer on AI. In this selection he

discusses the mythological associations of AI. He discusses both Prometheus, the god who stole fire from Olympus

TEACHING S

UGGESTIONS

Ask students to highlight Luger’s important statements about the origins of AI science and to turn those statements into

questions that might guide further reading. For instance, students might highlight this line: “The notion that human

efforts to gain knowledge constitute a transgression against the laws of God or nature is deeply ingrained in Western

thought.” An associated question might be: What evidence is there today of the conviction that humans should not seek

forbidden knowledge?

ANSWERS TO REVIEW Q

UESTIONS

1. Prometheus was a god who stole the secret of fire from the gods of Olympus and gave it to humans, enabling

2. The positive legacy of Prometheus’s gift of rational thought touches every aspect of human life, including

4. In the preface to Frankenstein, Shelley recounts listening to a conversation about new discoveries in science.

Will Technology Help Us Become Immortal? (p. 373)

Interview with Yuval Harari

UMMARY

This interview with the Israeli historian builds on breakthrough insights published first in

Sapiens: A Brief History

of Humankind. Harari is concerned with the relationship between history and biology. He takes a long view of human

TEACHING S

UGGESTIONS

Most students will be unfamiliar with the type of epochal history that Harari investigates. His perspective is sweeping,

not (as with many historians) narrow. It may be worth preparing students for substantial shift in perspective by

2030—Ray Kurzweil’s Predictions or Bill Joy’s Fears? (374)

Carolyn Mathas

UMMARY

Mathas summarizes Bill Joy’s and Ray Kurzweil’s the radically opposed futures concerning humans an artificial

intelligence. Are we on a collision course, leading to the doom of humankind (Bill Joy)? Are we on the threshold of a

summarizes the opposing positions and invites readers to weigh in.

TEACHING S

UGGESTIONS

This selection is structured as a straightforward comparison-contrast. Ask students to take notes on this structure, as

separate from content, and compare notes. How has Maathas organized the piece? Is she neutral in her presentation?

Does she favor one version of the future over the other?

An Open Letter on Artificial Intelligence (p. 377)

Future of Life Institute

Research Priorities for Robust and Beneficial Artificial Intelligence (p. 378)

Stuart Russel, Daniel Dewey, and Max Tegmark

An Open Letter on AI: Why Now? (p. 380)

Seán Ó hÉigeartaigh

UMMARY

This cluster of readings concerns a remarkable open letter on artificial intelligence signed by an all-star list of

scientists, initially, after an AI conference, and subsequently signed by more than 8.000 scientists world wide. While

certainly welcoming of, and encouraging, developments in AI, the letter is founded in part on an underlying wariness

of intelligent machines. It would seem that those most familiar with the science are most wary of possibilities it may

TEACHING S

UGGESTIONS

This passage is typical of, and central to, the letter and its ancillary documents—the research priorities and

Ó hÉigeartaigh’s statement on timing: “It’s no longer enough to ask “can we build it?” Now that it looks like we can,

we have to ask: “How can we build it to provide most benefit? And how must we update our own systems – legal,

economic, ethical – so that the transition is smooth, and we make the most of the positives while minimizing the

negatives?” Engage students in these questions, breaking them into groups to discuss legal, economic, and ethical

considerations of AI. Students can report on each group’s discussions with the entire glass. As a class, you might also

discuss Basulto’s preference of emphasizing existential reward over existential risk. Establish as a basis for this

discussion what AI science will look like in a few hundred years. Assume AI has vast capabilities. What are the

implications for human evolution? Would students object to AI enhancements to their bodies—if, for instance, such

enhancements could prevent cancer or heart attacks?

Motion for a European Parliament Resolution to the Commission on Civil Law Rules on

Robotics (p. 383)

Committee on Legal Affairs, European Parliament

UMMARY

This document takes account of the present state of the art of AI and concludes that the science has advanced to the

point that its benefits and dangers to humanity are clear and that the entire European Union, as a block, should

establish regulations that ensure benefits and avoid perils. The document lays out its rationale in a long series of

TEACHING S

UGGESTIONS

Students can easily get lost in the Motion, due to its format, which is peculiar to legal writing. The document

is, grammatically, a single sentence—which is parsed for students in the headnote. Review this structure

with students to make sure they understand the Motion’s structure. At that point t hey can dive into the

Superintelligence: The Idea That Eats Smart People (p. 388)

Maciej Ceglowski

UMMARY

In 2016, Oxford philosopher and artificial intelligence researcher Nick Bostrom wrote Superintelligence, a book

arguing that if scientists and policy makers do not wisely (and soon) direct the development of AI, the technology will

one day overrun humanity. Many tech and scientific luminaries—including Stephen Hawking, Bill Gates, and Elon

Musk—share Bostrom’s concern. The potential for AI to cause an unprecedented harm, as well as good, has prompted

a worldwide debate about how best to proceed, including two documents you’ll find in this chapter: the Open Letter on

AI and the European Parliament’s draft resolution concerning rules on AI robotics.

In a keynote address delivered at a conference in Zagreb, Croatia in late 2016, tech entrepreneur Maciej (pronounced

MAH-tchay) Ceglowski takes a skeptical view of Bostrom’s arguments about runaway, superintelligent machines and

systematically breaks down what he terms Bostrom’s “crazy idea.” Ceglowski structures his critique as follows:

Introduction: Unknowns regarding AI are reminiscent of unknowns regarding first test of

atomic bomb.

Summary of Bostom’s book: Superintelligence.

Introducing the critique and its structure: outside perspective and inside perspective.

Critique, Part I

Inside Arguments against Bostrom’s superintelligence based on what we know

about how AI is programmed:

Critique, Part II

Outside Argument against Bostrom’s superintelligence based on what we know

Conclusion

“I want to put my cards on the table. What do I think AI is and the possibilities of it are?”

AI now is in same position like Alchemy was in 17th century. We have clues. But we

TEACHING S

UGGESTIONS

The lecture is an outstanding example of critique: well researched, logical, methodical, and clearly—even

entertainingly—presented. It’s long but worth the effort. Students who can follow and grasp the whole will gain

Don’t Fear AI (p. 405)

Dominic Basulto

UMMARY

Dominc Basulto concludes the cluster on the open letter and responds to the subtle for important wariness of AI. He

takes a more optimistic view of the advancing science and has no hesitancy when considering evolutionary changes in

the human form to part biological, part silicon.

TEACHING S

UGGESTIONS

Basulto helps to articulate the stakes involved in getting AI right. Before students read this selection, introduce the

terms “existential risk” and “existential reward,” focusing on existential and the scope and significance of that word.

Robots Will Take Your Job; Will They Guarantee Your Income? (p. 407)

Scott Santens

UMMARY

Santens does a good job of explaining to an audience of non-experts how AI algorithms learn, essentially teaching

themselves by working through vast databases in order to continually improve performance. The approach is called

deep learning. Santens makes a clear distinction between work that is jeopardized by AI science (routine work) and

TEACHING S

UGGESTIONS

One of the reasons students come to college is to gain skills that will gain them entry in the workforce. The Santens

article should be ringing loud alarm bells for your students, who would be well advised to take developments in AI and

robotics seriously as they plan their working lives. Emphasize for students the differences between routine and non-

routine work to be sure they appreciate the distinctions. And second point for discussion: Everyone likes their holidays.

But to what extent are vacations defined in opposition to work? Can there be such a thing as leisure when one is not

working? What would be the challenges of not working? the benefits? For instance, what would it mean to go “on

vacation”? Wouldn’t life be one continual vacation? Or would vacations not be possible?”

A Review of Humans are Underrated by Geoff Colvin (p. 411)

Tyler Cowen

UMMARY

This book review of Geoff Colvin’s Humans are Underrated summarizes the book, picks a few bones, and broadly

agrees that Colvin has done a good job arguing that human beings possess a capacity for empathy and interpersonal

TEACHING S

UGGESTIONS

Cowens is an economist who reviews a book written by business writer Geoffrey Colvin. The piece is a good example

of critique. Cowens broadly endorses Colvin’s book, calling it “a worthy addition to the growing collection of books

Automation and Anxiety (p. 413)

The Economist

UMMARY

This overview of automation of the workplace steers a middle course between optimism and pessimism that wide

swaths of the labor force will be displaced. True, many jobs will be automated out of existence. The authors refer to

the much-cited Frey and Osborne report that estimates an astonishing “47% of workers in American has jobs at high

risk of automation.” As a counter to the pessimism such a statistic will engender, the authors are quick to observe (and

TEACHING S

UGGESTION

One activity relating to this selection: ask students to go online and locate the Frey and Osborne article on

Sympathy for the Robot: Visions of AI in Westworld (p. 420)

Christopher Orr

UMMARY

Christopher Orr locates HBO’s Westworld in the tradition of dystopian literature and film that turn on the question:

What if our creations turn against us? He admires the series, writing that its dystopian themes have “rarely been

developed with [such] sophistication and ingenuity.” He focuses on the ways Westworld “reframes” or “inverts” our

TEACHING S

UGGESTIONS

Imaginative literature and film have played an essential role in the development of artificial intelligence, and

Christopher Orr has done a good job in setting one of the most provocative examples—the HBO series Westworld—in

a larger context of fiction and film. He begins with Frankenstein and he references Prometheus. Ask students to

investigate any of the films, stories, or novels that Orr mentions in his article and learn enough about it to identify AI-

Testing the Turing Test (p. 425)

The Turing Test—George Luger (p. 425)

Intelligent Machines that Compose Sonnets—National Public Radio (p. 427)

Intelligent Machines that Compose Music (p. 428)

Intelligent Machines that Draw and Paint (p. 428)

Intelligent Machines that Chat with You (p. 428)

UMMARY

This final cluster of readings gives students chance to investigate (play with) AI systems online and to gauge their

human-like qualities. How intelligent are these so-called intelligent agents? The first selection in the cluster is George

TEACHING S

UGGESTIONS

Of the online AI machines students view and interact with in this chapter, two are especially impressive: The Bach-like

musical compositions and the Rembrandt portrait—both powered by deep learning algorithms. The Chatbots, by