Research Brief 17, dated December 2020, examines artificial intelligence’s capabilities, limitations, and potential effects on employment. The authors argue that AI is unlikely to cause lasting mass unemployment and emphasize human-computer collaboration through “superminds” rather than simple worker replacement. The brief reviews machine learning, natural language processing, robotics, and computer vision; discusses technical challenges and emerging research; and considers how AI may eliminate, create, and transform jobs. It recommends education, reskilling, worker support, and policy changes involving schools, businesses, worker organizations, community organizations, and government to help distribute AI’s benefits more broadly.
Free preview text
The first 5 pages, shown as text so search engines and readers can find this document.
Page 1
RESEARCH BRIEF 17 • DECEMBER 2020
ARTIFICIAL INTELLIGENCE AND
THE FUTURE OF WORK
THOMAS W. MALONE, DIRECTOR, MIT CENTER FOR COLLECTIVE INTELLIGENCE
PATRICK J. MCGOVERN PROFESSOR OF MANAGEMENT, MIT SLOAN SCHOOL OF
MANAGEMENT
DANIELA RUS, DIRECTOR, COMPUTER SCIENCE AND ARTIFICIAL INTELLIGENCE
LABORATORY (CSAIL)
ANDREW & ERNA VITERBI PROFESSOR OF ELECTRICAL ENGINEERING AND COMPUTER
SCIENCE
MIT TASK FORCE ON THE WORK OF THE FUTURE MEMBER
ROBERT LAUBACHER, ASSOCIATE DIRECTOR, MIT CENTER FOR COLLECTIVE
INTELLIGENCE
Page 2
Artificial Intelligence and the Future
of Work
Thomas W. Malone, Daniela Rus, and Robert Laubacher
Executive Summary
Artificial intelligence (AI) is one of the most important technologies in the world today. The United States
and China compete for dominance in its development (Lee, 2018). CEOs believe it will significantly change
the way they do business (Rao et al., 2019). And it has helped companies such as Facebook, Google, and
Apple to become among the largest in the world.
But how will this technology affect work in the future? Will it lead to a permanent underclass of people
who are no longer employable because their jobs are being done by computers? Will superintelligent
computers someday take over the world, finding little use for the humans who created them? Or will
robotic servants usher in a golden age of human leisure and prosperity?
In this report, we conclude that recent fears about AI leading to mass unemployment are unlikely to be
realized. Instead, we believe that—like all previous labor-saving technologies—AI will enable new
industries to emerge, creating more new jobs than are lost to the technology. But we see a significant need
for governments and other parts of society to help smooth this transition, especially for the individuals
whose old jobs are disrupted and who cannot easily find new ones.
In addition, even though AI is advancing rapidly, we believe that we are at least many decades away
from the day when computers have complete, human-level artificial intelligence. For the foreseeable
future, therefore, the most promising uses of AI will not involve computers replacing people, but rather,
people and computers working together—as “superminds”—to do both cognitive and physical tasks that
could not be done before.
Page 3
3
RECOMMENDATIONS
Specifically, we recommend the following actions for the following key stakeholders, which range from
schools to businesses and government. In some instances, we call for stakeholders to provide direct
assistance to people whose work life is disrupted by AI. In other cases, we call for stakeholders to pay for
services delivered by other entities.
– K-12 schools, colleges, and universities
• Ensure that every high school has a computer science teacher with appropriate subject training.
• Enhance secondary-school programs by expanding current computer literacy curricula to include
computational thinking.
• Build community college programs that include reskilling tracks and online micro-degree offerings
matched to the needs of local employers. Offerings should include apprenticeships that provide
on-the-job training.
• Expand post-secondary-school enrollment—through traditional, online, and hybrid programs—to
better educate the population as a whole, enhancing not only general cognitive skills but also
social skills.
– Businesses
• Focus on applying AI to work with people in achieving new or better outcomes—rather than to
replace people.
• Offer training for employees whose positions will be eliminated or transformed by AI to prepare
them for other jobs.
• Provide appropriate continuing education to decision-makers, developers, and users of AI; this
should include training in testing and evaluation practices.
• Create nontraditional training initiatives focused on enhancing the skills of partners and customers.
– Worker organizations
• To meet the needs of an increasingly dynamic world of work, current worker organizations (such as
labor unions and professional associations) or new ones (perhaps called “guilds”) should expand
their roles to provide benefits previously tied to formal employment (such as insurance and
pensions, career development, social connections, a sense of identity, and income security).
– Community organizations
• Use strong local ties and deep understanding of the nature of the specific challenges faced by
community members to help workers deal with disruptions caused by AI.
• Provide aggressive communication programs that inform displaced workers about their reskilling
and placement opportunities, endeavoring to inspire them to imagine new career pathways.
Page 4
4
– Government
• Increase federal and state government investment in post-secondary education and
reskilling/training programs to make the American workforce once again the best educated in the
world.
• Reshape the legal and regulatory framework that governs work to encourage job creation and to
adapt to other disruptions created by AI.
What Is Artificial Intelligence?
There are many ways to define AI, but one simple definition is “intelligence demonstrated by machines”
(Wikipedia, 2020). A leading textbook in the field defines it more precisely as machines thinking and/or
acting humanly and/or rationally (Russell and Norvig, 2020).
In general, the field of artificial intelligence seeks to advance the science and engineering of intelligence,
with the goal of creating machines with human-like characteristics. This includes developing machines with a
wide range of human-inspired capabilities, including communication, perception, planning, reasoning,
knowledge representation, the ability to move and manipulate objects, and learning. AI approaches
problems using tools and techniques from a wide variety of other fields, including probability and statistics,
symbolic computation, search and optimization, game theory, information engineering, mathematics,
psychology, linguistics, and philosophy.
BRIEF HISTORY OF AI
Artificial intelligence traces its beginnings to Alan Turing, who in a 1950 paper imagined a machine that
could communicate—via an exchange of typed messages—so capably that people conversing with it
could not tell whether they were interacting with a machine or another person (Turing 1950). The term
artificial intelligence was coined in mid-1955, in a proposal by a group of computer scientists, including
Marvin Minsky of MIT, to hold a workshop at Dartmouth College the following summer. The goal of the
envisioned workshop was “to find how to make machines use language, form abstractions and concepts,
solve kinds of problems now reserved for humans, and improve themselves” (McCarthy et al., 1955). The
workshop was held between June and August 1956; many consider the event to mark the birth of the field
of AI.
In the years following that seminal workshop, labs were established at MIT, Stanford, and Carnegie Mellon
to pursue research in AI. Other labs soon followed. Some took highly focused, narrow approaches. MIT’s
Minsky, in contrast, advocated combining a broad, multifaceted range of methods—an approach outlined
in his influential 1961 paper “Steps Toward Artificial Intelligence” (Minsky, 1961). MIT’s Patrick Winston
Page 5
5
nicely characterized the contributions of two of the pioneers of the field of AI: “Turing told us that we could
do this … Minsky told us what to do” (MIT Sloan and CSAIL, 2017, Module 1).
The next several decades saw a broad range of advances in AI. The first wave of breakthroughs relied on
such approaches as dividing complex problems into sets of small, constituent tasks that machines could do
(Slagle, 1961). A later wave involved encoding specialized human expertise into rules for the machine to
follow (Buchanan and Shortliffe, 1984).
By the mid-1980s, however, advances slowed, and the field experienced what came to be called the “AI
winter.” Some new discoveries occurred over the next few decades, but the field’s previously confident
projections that rapid progress would continue were not realized.
Progress in the field has since accelerated, notably over the past decade or so. This has been enabled by
the convergence of advances in three areas: the explosion of data, ever-increasing computing power, and
new algorithms (especially algorithms called “neural networks” and “deep learning”). Together, these
advances have created a technological tidal wave. The first sign appeared in 2011, when IBM’s Watson
program beat the best human players of the TV game show Jeopardy. Other advances followed in quick
succession. In 2015, for example, Google’s AlphaGo beat a grand master of the game of Go, and in
2017, AlphaGo beat the number one–ranked human player of the game, a feat that had previously been
considered impossible, since Go is far more complex than chess.
SPECIALIZED VS. GENERAL INTELLIGENCE
Many people don’t realize that even the most advanced examples of AI today involve machines that can
only perform very specific tasks. This specialized AI, sometimes called narrow or weak AI, is only able to
tackle a limited number of problems within a narrow scope. Examples include Google’s AlphaGo, IBM’s
Watson, and numerous speech transcription programs.
Narrow AI systems solve problems by working in predefined ways, and they often rely on complex pattern
recognition. Typically, they look at vast amounts of data, extract patterns, make predictions, and act
based on those predictions. To play Go, for example, the computer can study every game ever recorded
and model the likely outcome of every possible move. Narrow AI solutions exist for a wide range of
specific problems and can do a lot to improve efficiency and productivity in the work world. For example,
in addition to the version of IBM’s Watson system that plays Jeopardy!, a substantially different version of
the system can help doctors make better decisions.
Specialized intelligence can be contrasted with general intelligence, which humans possess: the ability to
undertake a wide variety of tasks in a broad range of settings (Malone, 2018, 24; Russell & Norvig,
2020). A long-term objective of the AI field is to develop systems that can exhibit artificial general
intelligence (AGI, also called strong AI). A recent step toward AGI was taken by OpenAI in 2020, with the
Artificial Intelligence and the Future of Work: summary, contents and background
This guide helps you understand the scope of Artificial Intelligence and the Future of Work before or alongside reading the original document. It is based on the catalogue record, credited creator, source, classification and topics shown on this page. It does not replace the original work and does not invent claims about passages that Doxxly cannot verify.
Summary and overview
Artificial Intelligence and the Future of Work is a 39-page documentcredited to Thomas W. Malone, Daniela Rus, and Robert Laubacher. Research Brief 17, dated December 2020, examines artificial intelligence’s capabilities, limitations, and potential effects on employment. The authors argue that AI is unlikely to cause lasting mass unemployment and emphasize human-computer collaboration through “superminds” rather than simple worker replacement. The brief reviews machine learning, natural language processing, robotics, and computer vision; discusses technical challenges and emerging research; and considers how AI may eliminate, create, and transform jobs. It recommends education, reskilling, worker support, and policy changes involving schools, businesses, worker organizations, community organizations, and government to help distribute AI’s benefits more broadly. It appears in the Technology & Security collection, which groups it with other material that readers may use for comparison and further study.
The item can be approached in several ways. A reader looking for a quick orientation can begin with the free five-page preview, inspect the page structure and use the metadata to judge relevance. Someone undertaking deeper research can work through the complete file, search its indexed text, record useful page references and compare its treatment of Artificial intelligence, Future of work, Human-computer collaboration, Superminds, Automation, Workforce reskilling, Employment policy, Machine learning, Robotics with the related resources linked below.
Rather than reducing the work to a single claim, this page treats it as a source that should be read in context. The title, creator, publication source, licence and format all help explain what kind of evidence it may offer. Readers should distinguish the original author's statements from Doxxly's catalogue description, and should cite the original edition when using the material in academic, professional or published work.
Key topics and reader takeaways
The catalogue connects this document with Artificial intelligence, Future of work, Human-computer collaboration, Superminds, Automation, Workforce reskilling, Employment policy, Machine learning, Robotics. These terms offer practical entry points rather than a substitute for close reading. Start by identifying the definitions, examples, arguments, tables or visual explanations attached to each topic. Then note where the creator draws connections, makes distinctions or moves from background material to a conclusion, recommendation, calculation or narrative development.
A useful reading method is to separate description from interpretation. First record what the item actually presents: names, headings, figures, formulas, quotations or stated propositions. Next consider why those details matter within the whole work. Finally compare them with another source from the same subject area. This process helps turn a long digital file into evidence that can be evaluated rather than merely skimmed.
Table of contents and reading guide
The following is an editorial content guide based on the verified catalogue topics. It is not represented as the original author's formal table of contents. Use it to organise notes while moving through the 39 pages in the player.
Part 1
Artificial intelligence
Use this section of the reading guide to identify how Artificial intelligence appears in Artificial Intelligence and the Future of Work, note important terms, and connect the subject to the wider purpose of the work.
Part 2
Future of work
Use this section of the reading guide to identify how Future of work appears in Artificial Intelligence and the Future of Work, note important terms, and connect the subject to the wider purpose of the work.
Part 3
Human-computer collaboration
Use this section of the reading guide to identify how Human-computer collaboration appears in Artificial Intelligence and the Future of Work, note important terms, and connect the subject to the wider purpose of the work.
Part 4
Superminds
Use this section of the reading guide to identify how Superminds appears in Artificial Intelligence and the Future of Work, note important terms, and connect the subject to the wider purpose of the work.
Part 5
Automation
Use this section of the reading guide to identify how Automation appears in Artificial Intelligence and the Future of Work, note important terms, and connect the subject to the wider purpose of the work.
Part 6
Workforce reskilling
Use this section of the reading guide to identify how Workforce reskilling appears in Artificial Intelligence and the Future of Work, note important terms, and connect the subject to the wider purpose of the work.
Part 7
Employment policy
Use this section of the reading guide to identify how Employment policy appears in Artificial Intelligence and the Future of Work, note important terms, and connect the subject to the wider purpose of the work.
Part 8
Machine learning
Use this section of the reading guide to identify how Machine learning appears in Artificial Intelligence and the Future of Work, note important terms, and connect the subject to the wider purpose of the work.
When a heading in the original differs from this guide, follow the original heading in your citation. Page numbers can also vary between editions. Doxxly's direct page field is useful for navigation, but formal references should include the title, creator, source and edition details available in the file itself.
About Thomas W. Malone, Daniela Rus, and Robert Laubacher
Doxxly credits Thomas W. Malone, Daniela Rus, and Robert Laubacher as the author or creator of Artificial Intelligence and the Future of Work. This page does not add an unverified biography, job title or institutional affiliation. The attribution is preserved so readers can distinguish the creator from the hosting library and follow the original record when an external author link is available.
Author context matters because a work reflects a particular purpose, audience and point of view. Before relying on the material, consider what the creator set out to explain, when the source was produced and whether later research or revised guidance should also be consulted. For literature and philosophy, compare translations and editions; for technical material, check whether a newer version has superseded the one shown here.
About the source organisation
The recorded source for this item is the credited original source. Doxxly provides access and discovery tools but does not claim to be the original publisher unless explicitly stated. Source links, creator attribution and the recorded Public domain licence are displayed to help readers assess provenance and permitted use.
Provenance is especially important for reports, standards, presentations and working spreadsheets. Check the original source for amendments, supporting files and current policy. A trustworthy research workflow keeps the source URL with any notes or downloaded copy, verifies important figures against the original, and avoids treating an archived document as automatically current.
How to read and use this document
Begin with the title and opening pages, then use the guide above to decide whether to read sequentially or move to a particular topic. The Doxxly player provides previous and next navigation, a direct page field, zoom steps, fit-to-width and fit-to-page modes, rotation and fullscreen. These controls are useful when a page contains small type, a wide table, an image, a scanned layout or a dense sequence of notes.
The first five pages are open to every visitor. A plan unlocks the complete item and full-document search. Search is best used for distinctive names, technical terms or short phrases; broad words can produce too many possible contexts. When you find a useful passage, read the surrounding pages so that a sentence, chart or formula is not separated from its explanation.
If downloading is enabled, members can obtain the original PDF file. Keep the original format when formulas, slide sequencing, pagination or layout carries meaning. Respect the Public domain terms displayed on this page, and use the rights-report link if you believe the attribution or availability needs review.
Research questions to consider
As you read Artificial Intelligence and the Future of Work, ask which audience the creator addresses, what problem or idea organises the work, what evidence is presented, and which assumptions remain unstated. Note whether important terms are defined consistently and whether examples represent a broader pattern or a narrow case. For quantitative material, record units, time periods, formulas and data sources. For interpretive material, separate the creator's language from your own summary.
A second set of questions concerns relevance today. Which parts remain durable, which depend on their historical or institutional setting, and what would need confirmation from a newer source? Following the tags and the Technology & Security collection can reveal useful comparisons. Agreement between sources can strengthen confidence, while disagreement can point to changing definitions, methods, evidence or priorities.
Finally, decide what you need from the item: background knowledge, a quotation, a worked example, a model, a visual explanation or a starting bibliography. That purpose should guide the depth of reading and the notes you keep. Doxxly helps locate and navigate the source; careful interpretation remains the reader's responsibility.