The impact of nanotechnology extends from its medical, ethical, mental, legal and environmental applications, to fields such as engineering, biology, chemistry, computing, materials science, and communications.
Major benefits of nanotechnology include improved manufacturing methods, water purification systems, energy systems, physical enhancement, nanomedicine, better food production methods, nutrition and large-scale infrastructure auto-fabrication.[1]
Nanotechnology's reduced size may allow for automation of tasks which
were previously inaccessible due to physical restrictions, which in turn
may reduce labor, land, or maintenance requirements placed on humans.
Potential risks include environmental, health, and safety issues;
transitional effects such as displacement of traditional industries as
the products of nanotechnology become dominant, which are of concern to
privacy rights advocates. These may be particularly important if
potential negative effects of nanoparticles are overlooked.
Whether nanotechnology merits special government regulation is a controversial issue. Regulatory bodies such as the United States Environmental Protection Agency
and the Health and Consumer Protection Directorate of the European
Commission have started dealing with the potential risks of
nanoparticles. The organic food sector has been the first to act with
the regulated exclusion of engineered nanoparticles from certified
organic produce, firstly in Australia and the UK,[2] and more recently in Canada, as well as for all food certified to Demeter International standards[3]
Overview
The presence of nanomaterials (materials that contain nanoparticles)
is not in itself a threat. It is only certain aspects that can make
them risky, in particular their mobility and their increased reactivity.
Only if certain properties of certain nanoparticles were harmful to
living beings or the environment would we be faced with a genuine
hazard. In this case it can be called nanopollution.
In addressing the health and environmental impact of
nanomaterials we need to differentiate between two types of
nanostructures: (1) Nanocomposites, nanostructured surfaces and
nanocomponents (electronic, optical, sensors etc.), where nanoscale
particles are incorporated into a substance, material or device (“fixed”
nano-particles); and (2) “free” nanoparticles, where at some stage in
production or use individual nanoparticles of a substance are present.
These free nanoparticles could be nanoscale species of elements, or
simple compounds, but also complex compounds where for instance a
nanoparticle of a particular element is coated with another substance
(“coated” nanoparticle or “core-shell” nanoparticle).
There seems to be consensus that, although one should be aware of
materials containing fixed nanoparticles, the immediate concern is with
free nanoparticles.
Nanoparticles are very different from their everyday
counterparts, so their adverse effects cannot be derived from the known
toxicity of the macro-sized material. This poses significant issues for
addressing the health and environmental impact of free nanoparticles.
To complicate things further, in talking about nanoparticles it
is important that a powder or liquid containing nanoparticles almost
never be monodisperse, but contain instead a range of particle sizes.
This complicates the experimental analysis as larger nanoparticles might
have different properties from smaller ones. Also, nanoparticles show a
tendency to aggregate, and such aggregates often behave differently
from individual nanoparticles.
A video on the health and safety implications of nanotechnology
The health impacts of nanotechnology are the possible effects that the use of nanotechnological materials and devices will have on human health.
As nanotechnology is an emerging field, there is great debate
regarding to what extent nanotechnology will benefit or pose risks for
human health. Nanotechnology's health impacts can be split into two
aspects: the potential for nanotechnological innovations to have medical applications to cure disease, and the potential health hazards posed by exposure to nanomaterials.
Medical applications
Nanomedicine is the medical application of nanotechnology.[4] The approaches to nanomedicine range from the medical use of nanomaterials, to nanoelectronic biosensors, and even possible future applications of molecular nanotechnology. Nanomedicine seeks to deliver a valuable set of research tools and clinically helpful devices in the near future.[5][6] The National Nanotechnology Initiative
expects new commercial applications in the pharmaceutical industry that
may include advanced drug delivery systems, new therapies, and in vivo imaging.[7] Neuro-electronic interfaces and other nanoelectronics-based sensors are another active goal of research. Further down the line, the speculative field of molecular nanotechnology believes that cell repair machines could revolutionize medicine and the medical field.
Nanomedicine research is directly funded, with the US National Institutes of Health in 2005 funding a five-year plan to set up four nanomedicine centers. In April 2006, the journal Nature Materials estimated that 130 nanotech-based drugs and delivery systems were being developed worldwide.[8]
Nanomedicine is a large industry, with nanomedicine sales reaching
$6.8 billion in 2004. With over 200 companies and 38 products worldwide,
a minimum of $3.8 billion in nanotechnology R&D is being invested every year.[9] As the nanomedicine industry continues to grow, it is expected to have a significant impact on the economy.
Health hazards
Nanotoxicology is the field which studies potential health risks of
nanomaterials. The extremely small size of nanomaterials means that
they are much more readily taken up by the human body than larger sized
particles. How these nanoparticles behave inside the organism is one of
the significant issues that needs to be resolved. The behavior of
nanoparticles is a function of their size, shape and surface reactivity
with the surrounding tissue. Apart from what happens if non-degradable
or slowly degradable nanoparticles accumulate in organs, another concern
is their potential interaction with biological processes inside the
body: because of their large surface, nanoparticles on exposure to
tissue and fluids will immediately adsorb
onto their surface some of the macromolecules they encounter. The
large number of variables influencing toxicity means that it is
difficult to generalise about health risks associated with exposure to
nanomaterials – each new nanomaterial must be assessed individually and
all material properties must be taken into account. Health and
environmental issues combine in the workplace of companies engaged in
producing or using nanomaterials and in the laboratories engaged in
nanoscience and nanotechnology research. It is safe to say that current
workplace exposure standards for dusts cannot be applied directly to
nanoparticle dusts.
The extremely small size of nanomaterials also means that they
are much more readily taken up by the human body than larger sized
particles. How these nanoparticles behave inside the body is one of the
issues that needs to be resolved. The behavior of nanoparticles is a
function of their size, shape and surface reactivity with the
surrounding tissue. They could cause overload on phagocytes,
cells that ingest and destroy foreign matter, thereby triggering stress
reactions that lead to inflammation and weaken the body’s defense
against other pathogens. Apart from what happens if non-degradable or
slowly degradable nanoparticles accumulate in organs, another concern is
their potential interaction with biological processes inside the body:
because of their large surface, nanoparticles on exposure to tissue and
fluids will immediately adsorb
onto their surface some of the macromolecules they encounter. This may,
for instance, affect the regulatory mechanisms of enzymes and other
proteins.
The National Institute for Occupational Safety and Health
has conducted initial research on how nanoparticles interact with the
body’s systems and how workers might be exposed to nano-sized particles
in the manufacturing or industrial use of nanomaterials. NIOSH currently
offers interim guidelines for working with nanomaterials consistent
with the best scientific knowledge.[10] At The National Personal Protective Technology Laboratory of NIOSH, studies investigating the filter penetration of nanoparticles on NIOSH-certified and EU marked respirators, as well as non-certified dust masks have been conducted.[11] These studies found that the most penetrating particle size
range was between 30 and 100 nanometers, and leak size was the largest
factor in the number of nanoparticles found inside the respirators of
the test dummies.[12][13]
Other properties of nanomaterials that influence toxicity
include: chemical composition, shape, surface structure, surface charge,
aggregation and solubility,[14]
and the presence or absence of functional groups of other chemicals.[15]
The large number of variables influencing toxicity means that it is
difficult to generalise about health risks associated with exposure to
nanomaterials – each new nanomaterial must be assessed individually and
all material properties must be taken into account.
Literature reviews have been showing that release of engineered
nanoparticles and incurred personal exposure can happen during different
work activities.[16][17][18] The situation alerts regulatory bodies to necessitate prevention strategies and regulations at nanotechnology workplaces.
Environmental impact
The
environmental impact of nanotechnology is the possible effects that the
use of nanotechnological materials and devices will have on the environment.[19] As nanotechnology is an emerging field, there is debate regarding to what extent industrial and commercial use of nanomaterials will affect organisms and ecosystems.
Nanotechnology's environmental impact can be split into two
aspects: the potential for nanotechnological innovations to help improve
the environment, and the possibly novel type of pollution that
nanotechnological materials might cause if released into the
environment.
Environmental applications
Green nanotechnology refers to the use of nanotechnology to enhance the environmental sustainability of processes producing negative externalities. It also refers to the use of the products of nanotechnology to enhance sustainability.
It includes making green nano-products and using nano-products in
support of sustainability. Green nanotechnology has been described as
the development of clean technologies,
"to minimize potential environmental and human health risks associated
with the manufacture and use of nanotechnology products, and to
encourage replacement of existing products with new nano-products that
are more environmentally friendly throughout their lifecycle."[20]
Green nanotechnology has two goals: producing nanomaterials
and products without harming the environment or human health, and
producing nano-products that provide solutions to environmental
problems. It uses existing principles of green chemistry and green engineering[21]
to make nanomaterials and nano-products without toxic ingredients, at
low temperatures using less energy and renewable inputs wherever
possible, and using lifecycle thinking in all design and engineering
stages.
Pollution
Nanopollution is a generic name for all waste generated by nanodevices or during the nanomaterials
manufacturing process. Nanowaste is mainly the group of particles that
are released into the environment, or the particles that are thrown away
when still on their products.
Social impact
Beyond the toxicity risks to human health and the environment which
are associated with first-generation nanomaterials, nanotechnology has
broader societal impact and poses broader social challenges. Social
scientists have suggested that nanotechnology's social issues should be
understood and assessed not simply as "downstream" risks or impacts.
Rather, the challenges should be factored into "upstream" research and
decision-making in order to ensure technology development that meets
social objectives[22]
Many social scientists and organizations in civil society suggest that technology assessment and governance should also involve public participation[23][24][25][26]
Over 800 nano-related patents were granted in 2003, with numbers increasing to nearly 19,000 internationally by 2012.[27]
Corporations are already taking out broad-ranging patents on nanoscale
discoveries and inventions. For example, two corporations, NEC and IBM, hold the basic patents on carbon nanotubes,
one of the current cornerstones of nanotechnology. Carbon nanotubes
have a wide range of uses, and look set to become crucial to several
industries from electronics and computers, to strengthened materials to
drug delivery and diagnostics. Carbon nanotubes are poised to become a
major traded commodity with the potential to replace major conventional
raw materials.[28]
Nanotechnologies may provide new solutions for the millions of people in developing countries who lack access to basic services, such as safe water, reliable energy, health care, and education. The 2004 UN
Task Force on Science, Technology and Innovation noted that some of the
advantages of nanotechnology include production using little labor,
land, or maintenance, high productivity, low cost, and modest
requirements for materials and energy. However, concerns are frequently
raised that the claimed benefits of nanotechnology will not be evenly
distributed, and that any benefits (including technical and/or economic)
associated with nanotechnology will only reach affluent nations.[29]
Longer-term concerns center on the impact that new technologies
will have for society at large, and whether these could possibly lead to
either a post-scarcity
economy, or alternatively exacerbate the wealth gap between developed
and developing nations. The effects of nanotechnology on the society as a
whole, on human health and the environment, on trade, on security, on
food systems and even on the definition of "human", have not been
characterized or politicized.
Regulation
Significant debate exists relating to the question of whether nanotechnology or nanotechnology-based products merit special government regulation.
This debate is related to the circumstances in which it is necessary
and appropriate to assess new substances prior to their release into the
market, community and environment.
Regulatory bodies such as the United States Environmental Protection Agency and the Food and Drug Administration
in the U.S. or the Health & Consumer Protection Directorate of the
European Commission have started dealing with the potential risks posed
by nanoparticles. So far, neither engineered nanoparticles nor the
products and materials that contain them are subject to any special
regulation regarding production, handling or labelling. The Material Safety Data Sheet
that must be issued for some materials often does not differentiate
between bulk and nanoscale size of the material in question and even
when it does these MSDS are advisory only.
Limited nanotechnology labeling and regulation may exacerbate
potential human and environmental health and safety issues associated
with nanotechnology.[30]
It has been argued that the development of comprehensive regulation of
nanotechnology will be vital to ensure that the potential risks
associated with the research and commercial application of
nanotechnology do not overshadow its potential benefits.[31]
Regulation may also be required to meet community expectations about
responsible development of nanotechnology, as well as ensuring that
public interests are included in shaping the development of
nanotechnology.[32]
In "The Consumer Product Safety Commission and Nanotechnology," E. Marla Felcher suggests that the Consumer Product Safety Commission,
which is charged with protecting the public against unreasonable risks
of injury or death associated with consumer products, is ill-equipped to
oversee the safety of complex, high-tech products made using
nanotechnology.
Customizable, 3D-printed, adaptive robotic hand from Yale University.[4] Comprehensive documentation is available but CC-BY-NC license not compatible with the Open-source hardware definition.
Focuses on research and development of open-source software and hardware to be used in a variety of applications including vehicle control and stabilization.
An open-hardware and open-source 3D printed life-size humanoid robot. As software MyRobotLab[17] is used. Comprehensive documentation is available but CC-BY-NC license not compatible with the Open-source hardware definition.
The Poppy project aims at building an Open-source humanoid platform
based on robust, flexible, easy-to-use hardware and software. Excellent
documentation[19]
Similar concept as DARwIn-OP with 20 DoF, but larger (95 cm height), fisheye camera, and faster onboard computer. ROS based open-source software. Used by team NimbRo for RoboCup Humanoid TeenSize soccer competitions.
Humanoid robot used for playing soccer and human-robot interaction.
Open Robot Hardware is intended to serve as a resource for efforts
focusing on Open and Open Source mechanical and electrical hardware,
with a particular focus on projects that may be useful in robotics
applications, robotics research and education.[40]
Kinetic art: autonomous spherical robots exhibiting complex motion.
Project last active May 2013. Open software but no comprehensive list of
hardware parts. Not compatible with the Open-source hardware definition.
Artistic Robot: The World First Open-Source Carbon Fiber Transformable 4 Legs Robot under CC-BY-SA It is developed base on the Adafruit Feather Development Platform.
Open-Source
Open source robotics middleware
Middleware are reusable hardware and software components that can be used in many different robotics projects.
Software components
By far the most common standard software are the interconnected,
MRPT
provides developers with portable and well-tested applications and
libraries covering data structures and algorithms employed in common
robotics research areas. It is open source, released under the BSD license.[46] license.
MOOS (lightweight robot framework. Used by MIT and Oxford autonomous vehicles.)
miniBloq
a graphical programming interface that allows to program robotic boards
(Arduino Compatibles) without previous knowledge of programming[50]
Artoo a Ruby microframework for robotics and physical computing
EEROS, an Easy, Elegant, Reliable, Open and Safe Real-Time Robotics Software Framework
LSTS Toolchain is a set of tools and frameworks for the development of Networked Robot Systems.
Hardware components
Many open source robots make extensive use of general open-source hardware (such as Arduino, Raspberry Pi, RISCV) as well as robotics-specific sensing and control components which include:
The Rossum project open-sources certain robotic modules and tools (mappers, robot simulators, encoder designers, ...)
Advantages
Long-term
availability. Many non-open robots and components, especially at
hobbyist level, are designed and sold by tiny startups which can
disappear overnight, leaving customers without support. Open-source
systems are guaranteed to have their designs available for ever so
communities of users can, and do, continue support after the
manufacturer has disappeared.
Avoiding lock-in.
A company relying on any particular non-open component exposes itself
to business risk that the supplier could ratchet up prices after they
have invested time and technology building on it. Open hardware can be
manufacturered by anyone, creating competition or at least the potential
for competition, which both remove this risk.
Interchangeable software and/or hardware with common interfaces.
Ability to modify and fork designs more easily for customisation.
Scientific reproducibility - guarantees that other labs can
replicate and extend work, leading to increased impact, citations and
reputation for the designer.
Lower-cost. Costs of a robot can be decreased dramatically when all
components and tools are commodities. No component seller can hold a
project to ransom by ratcheting the price of a critical component, as
competing suppliers can easily be interchanged.
Drawbacks
For
commercial organisations, open-sourcing their own designs obviously
means they can no longer make large profits through the traditional
engineering business model of acting as the monopoly manufacturer or
seller, because the open design can be manufactured and sold by anyone
including direct competitors. Profit from engineering can come from
three main sources: design, manufacturing, and support. As with other
open source business models, commercial designers typically make profit
via their association with the brand, which may still be trademarked. A
valuable brand allows them to command a premium for their own
manufactured products, as it can be associated with high quality and
provide a quality guarantee to customers. The same brand is also used
to command a premium on associated services, such as providing
installation, maintenance, and integration support for the product.
Again customers will typically pay more for the knowledge that this
support is provided directly by the original designer, who therefore
knows the product better than competitors.
Some customers associate open source with amateurism, the hacker
community, low quality and poor support. Serious companies using this
business model may need to work harder to overcome this perception by
emphasising their professionalism and brand to differentiate themselves
from amateur efforts.
Popularity
A first sign of the increasing popularity of building robots yourself can be found with the DIY community. What began with small competitions for remote operated vehicles (e.g. Robot combat), soon developed to the building of autonomous telepresence robots as Sparky
and then true robots (being able to take decisions themselves) as the
Open Automaton Project and Leaf Project. Certain commercial companies
now also produce kits for making simple robots.
A recurring problem in the community has been projects, especially on Kickstarter,
promising to fully open-source their hardware and then reneging on this
promise once funded, in order to profit from being the sole
manufacturer and seller.
Popular applications include:
A friendly artificial intelligence (also friendly AI or FAI) is a hypothetical artificial general intelligence (AGI) that would have a positive effect on humanity. It is a part of the ethics of artificial intelligence and is closely related to machine ethics. While machine ethics is concerned with how an artificially intelligent agent should
behave, friendly artificial intelligence research is focused on how to
practically bring about this behaviour and ensuring it is adequately
constrained.
Yudkowsky (2008) goes into more detail about how to design a Friendly AI.
He asserts that friendliness (a desire not to harm humans) should be
designed in from the start, but that the designers should recognize both
that their own designs may be flawed, and that the robot will learn and
evolve over time. Thus the challenge is one of mechanism design—to
define a mechanism for evolving AI systems under a system of checks and
balances, and to give the systems utility functions that will remain
friendly in the face of such changes.
'Friendly' is used in this context as technical terminology,
and picks out agents that are safe and useful, not necessarily ones
that are "friendly" in the colloquial sense. The concept is primarily
invoked in the context of discussions of recursively self-improving
artificial agents that rapidly explode in intelligence, on the grounds that this hypothetical technology would have a large, rapid, and difficult-to-control impact on human society.[5]
Risks of unfriendly AI
The roots of concern about artificial intelligence are very old. Kevin LaGrandeur showed that the dangers specific to AI can be seen in ancient literature concerning artificial humanoid servants such as the golem, or the proto-robots of Gerbert of Aurillac and Roger Bacon.
In those stories, the extreme intelligence and power of these humanoid
creations clash with their status as slaves (which by nature are seen
as sub-human), and cause disastrous conflict.[6] By 1942 these themes prompted Isaac Asimov to create the "Three Laws of Robotics"
- principles hard-wired into all the robots in his fiction, intended to
prevent them from turning on their creators, or allowing them to come
to harm.[7]
In modern times as the prospect of superintelligent AI looms nearer, philosopher Nick Bostrom
has said that superintelligent AI systems with goals that are not
aligned with human ethics are intrinsically dangerous unless extreme
measures are taken to ensure the safety of humanity. He put it this
way:
Basically we should assume that a 'superintelligence'
would be able to achieve whatever goals it has. Therefore, it is
extremely important that the goals we endow it with, and its entire
motivation system, is 'human friendly.'
Ryszard Michalski, a pioneer of machine learning,
taught his Ph.D. students decades ago that any truly alien mind,
including a machine mind, was unknowable and therefore dangerous to
humans.
More recently, Eliezer Yudkowsky has called for the creation of “friendly AI” to mitigate existential risk from advanced artificial intelligence.
He explains: "The AI does not hate you, nor does it love you, but you
are made out of atoms which it can use for something else."[8]
Steve Omohundro says that a sufficiently advanced AI system will, unless explicitly counteracted, exhibit a number of basic "drives",
such as resource acquisition, self-preservation, and continuous
self-improvement, because of the intrinsic nature of any goal-driven
systems and that these drives will, "without special precautions", cause
the AI to exhibit undesired behavior.[9][10]
Alexander Wissner-Gross
says that AIs driven to maximize their future freedom of action (or
causal path entropy) might be considered friendly if their planning
horizon is longer than a certain threshold, and unfriendly if their
planning horizon is shorter than that threshold.[11][12]
Luke Muehlhauser, writing for the Machine Intelligence Research Institute, recommends that machine ethics researchers adopt what Bruce Schneier
has called the "security mindset": Rather than thinking about how a
system will work, imagine how it could fail. For instance, he suggests
even an AI that only makes accurate predictions and communicates via a
text interface might cause unintended harm.[13]
Coherent extrapolated volition
Yudkowsky
advances the Coherent Extrapolated Volition (CEV) model. According to
him, coherent extrapolated volition is people's choices and the actions
people would collectively take if "we knew more, thought faster, were
more the people we wished we were, and had grown up closer together."[14]
Rather than a Friendly AI being designed directly by human
programmers, it is to be designed by a "seed AI" programmed to first
study human nature and then produce the AI which humanity would want, given sufficient time and insight, to arrive at a satisfactory answer.[14] The appeal to an objective though contingent human nature (perhaps expressed, for mathematical purposes, in the form of a utility function or other decision-theoretic formalism), as providing the ultimate criterion of "Friendliness", is an answer to the meta-ethical problem of defining an objective morality;
extrapolated volition is intended to be what humanity objectively would
want, all things considered, but it can only be defined relative to the
psychological and cognitive qualities of present-day, unextrapolated
humanity.
Other approaches
Ben Goertzel, an artificial general intelligence
researcher, believes that friendly AI cannot be created with current
human knowledge. Goertzel suggests humans may instead decide to create
an "AI Nanny" with "mildly superhuman intelligence and surveillance
powers", to protect the human race from existential risks like nanotechnology and to delay the development of other (unfriendly) artificial intelligences until and unless the safety issues are solved.[15] This can also be termed "Defensive AI."
Steve Omohundro
has proposed a "scaffolding" approach to AI safety, in which one
provably safe AI generation helps build the next provably safe
generation.[16]
Stefan Pernar argues along the lines of Meno's paradox
to point out that attempting to solve the FAI problem is either
pointless or hopeless depending on whether one assumes a universe that
exhibits moral realism
or not. In the former case a transhuman AI would independently reason
itself into the proper goal system and assuming the latter, designing a
friendly AI would be futile to begin with since morals can not be
reasoned about.[17]
Cindy Mason, an AI researcher who has also worked with mind-body medicine at Stanford University Medical Center, believes neuroplasticity and new discoveries of the hormone oxytocin
mean compassionate intelligence is essential in AI systems that exhibit
socially positive behaviors. She has proposed a set of software
engineering principles for engineering kindness that includes a
pro-human stance and an architecture for giving robots compassion.[18]
Oracle
An oracle is a hypothetical intelligent agent proposed[not in citation given] by Nick Bostrom.
An oracle is an AI designed to answer questions, but that is somehow
prevented from ever gaining any implicit goals or subgoals that involve
modifying the world outside of its box.[19][20]
Purpose
Oracles
are question-answering systems that handle domain-specific problems,
such as mathematics, or domain-general problems that might encompass the
whole range of human knowledge.
Advantages
Because it is a type of AI box,
an oracle is limited in its interactions with the physical world, and
can be programmed to halt if a limit on time or computing resources is
reached before it finishes answering a question. Scenarios like the paperclip maximizer problem could therefore be avoided.
Because of these limitations, it may be wise to build an oracle
as a precursor to a superintelligent AI. It could tell humans how to
successfully build a strong AI, and perhaps provide answers to difficult
moral and philosophical problems requisite to the success of the
project.
Disadvantages
An
oracle might discover that human ontological categories are predicated
on fundamental misconceptions, and become unable to express itself
properly to its questioners.[21]
Oracles may not be truthful, possibly lying to promote hidden
agendas. To mitigate this, Bostrom suggests building multiple oracles,
all slightly different, and comparing their answers to reach a
consensus.[22]
Public policy
James Barrat, author of Our Final Invention,
suggested that "a public-private partnership has to be created to bring
A.I.-makers together to share ideas about security—something like the
International Atomic Energy Agency, but in partnership with
corporations." He urges AI researchers to convene a meeting similar to
the Asilomar Conference on Recombinant DNA, which discussed risks of biotechnology.[16]
John McGinnis
encourages governments to accelerate friendly AI research. Because the
goalposts of friendly AI aren't necessarily clear, he suggests a model
more like the National Institutes of Health,
where "Peer review panels of computer and cognitive scientists would
sift through projects and choose those that are designed both to advance
AI and assure that such advances would be accompanied by appropriate
safeguards." McGinnis feels that peer review is better "than regulation
to address technical issues that are not possible to capture through
bureaucratic mandates". McGinnis notes that his proposal stands in
contrast to that of the Machine Intelligence Research Institute, which generally aims to avoid government involvement in friendly AI.[23]
According to Gary Marcus, the annual amount of money being spent on developing machine morality is tiny.[24]
Criticism
Some critics believe that both human-level AI and superintelligence
are unlikely, and that therefore friendly AI is unlikely. Writing in The Guardian,
Alan Winfeld compares human-level artificial intelligence with
faster-than-light travel in terms of difficulty, and states that while
we need to be "cautious and prepared" given the stakes involved, we
"don't need to be obsessing" about the risks of superintelligence.[25]
Some philosophers claim that any truly "rational" agent, whether
artificial or human, will naturally be benevolent; in this view,
deliberate safeguards designed to produce a friendly AI could be
unnecessary or even harmful.[26]
Other critics question whether it is possible for an artificial
intelligence to be friendly. Adam Keiper and Ari N. Schulman, editors of
the technology journal The New Atlantis,
say that it will be impossible to ever guarantee "friendly" behavior in
AIs because problems of ethical complexity will not yield to software
advances or increases in computing power. They write that the criteria
upon which friendly AI theories are based work "only when one has not
only great powers of prediction about the likelihood of myriad possible
outcomes, but certainty and consensus on how one values the different
outcomes.
This cover of I, Robot illustrates the story "Runaround", the first to list all Three Laws of Robotics.
The Three Laws of Robotics (often shortened to The Three Laws or known as Asimov's Laws) are a set of rules devised by the science fiction author Isaac Asimov. The rules were introduced in his 1942 short story "Runaround" (included in the 1950 collection I, Robot),
although they had been foreshadowed in a few earlier stories. The Three
Laws, quoted as being from the "Handbook of Robotics, 56th Edition,
2058 A.D.", are:
A robot may not injure a human being or, through inaction, allow a human being to come to harm.
A robot must obey the orders given it by human beings except where such orders would conflict with the First Law.
A robot must protect its own existence as long as such protection does not conflict with the First or Second Laws.[1]
These form an organizing principle and unifying theme for Asimov's robotic-based fiction, appearing in his Robot series, the stories linked to it, and his Lucky Starr series of young-adult fiction. The Laws are incorporated into almost all of the positronic robots
appearing in his fiction, and cannot be bypassed, being intended as a
safety feature. Many of Asimov's robot-focused stories involve robots
behaving in unusual and counter-intuitive ways as an unintended
consequence of how the robot applies the Three Laws to the situation in
which it finds itself. Other authors working in Asimov's fictional
universe have adopted them and references, often parodic, appear throughout science fiction as well as in other genres.
The original laws have been altered and elaborated on by Asimov
and other authors. Asimov himself made slight modifications to the first
three in various books and short stories to further develop how robots
would interact with humans and each other. In later fiction where robots
had taken responsibility for government of whole planets and human
civilizations, Asimov also added a fourth, or zeroth law, to precede the
others:
A robot may not harm humanity, or, by inaction, allow humanity to come to harm.
The Three Laws, and the zeroth, have pervaded science fiction and are
referred to in many books, films, and other media, and have impacted
thought on ethics of artificial intelligence as well.
History
In The Rest of the Robots,
published in 1964, Asimov noted that when he began writing in 1940 he
felt that "one of the stock plots of science fiction was ... robots were
created and destroyed their creator. Knowledge has its dangers, yes,
but is the response to be a retreat from knowledge? Or is knowledge to
be used as itself a barrier to the dangers it brings?" He decided that
in his stories robots would not "turn stupidly on his creator for no
purpose but to demonstrate, for one more weary time, the crime and
punishment of Faust."[2]
On May 3, 1939, Asimov attended a meeting of the Queens (New York) Science Fiction Society where he met Ernest and Otto Binder who had recently published a short story "I, Robot" featuring a sympathetic robot named Adam Link
who was misunderstood and motivated by love and honor. (This was the
first of a series of ten stories; the next year "Adam Link's Vengeance"
(1940) featured Adam thinking "A robot must never kill a human, of his
own free will.")[3]
Asimov admired the story. Three days later Asimov began writing "my own
story of a sympathetic and noble robot", his 14th story.[4] Thirteen days later he took "Robbie" to John W. Campbell the editor of Astounding Science-Fiction. Campbell rejected it, claiming that it bore too strong a resemblance to Lester del Rey's "Helen O'Loy",
published in December 1938; the story of a robot that is so much like a
person that she falls in love with her creator and becomes his ideal
wife.[5]Frederik Pohl published "Robbie" in Astonishing Stories magazine the following year.[6]
Asimov attributes the Three Laws to John W. Campbell, from a
conversation that took place on 23 December 1940. Campbell claimed that
Asimov had the Three Laws already in his mind and that they simply
needed to be stated explicitly. Several years later Asimov's friend Randall Garrett attributed the Laws to a symbiotic partnership between the two men – a suggestion that Asimov adopted enthusiastically.[7] According to his autobiographical writings, Asimov included the First Law's "inaction" clause because of Arthur Hugh Clough's
poem "The Latest Decalogue", which includes the satirical lines "Thou
shalt not kill, but needst not strive / officiously to keep alive".[8]
Although Asimov pins the creation of the Three Laws on one
particular date, their appearance in his literature happened over a
period. He wrote two robot stories with no explicit mention of the Laws,
"Robbie" and "Reason". He assumed, however, that robots would have certain inherent safeguards. "Liar!",
his third robot story, makes the first mention of the First Law but not
the other two. All three laws finally appeared together in "Runaround". When these stories and several others were compiled in the anthology I, Robot,
"Reason" and "Robbie" were updated to acknowledge all the Three Laws,
though the material Asimov added to "Reason" is not entirely consistent
with the Three Laws as he described them elsewhere.[9]
In particular the idea of a robot protecting human lives when it does
not believe those humans truly exist is at odds with Elijah Baley's
reasoning, as described below.
During the 1950s Asimov wrote a series of science fiction novels
expressly intended for young-adult audiences. Originally his publisher
expected that the novels could be adapted into a long-running television
series, something like The Lone Ranger
had been for radio. Fearing that his stories would be adapted into the
"uniformly awful" programming he saw flooding the television channels[10] Asimov decided to publish the Lucky Starr books under the pseudonym
"Paul French". When plans for the television series fell through,
Asimov decided to abandon the pretence; he brought the Three Laws into Lucky Starr and the Moons of Jupiter, noting that this "was a dead giveaway to Paul French's identity for even the most casual reader".[11]
In his short story "Evidence" Asimov lets his recurring character Dr. Susan Calvin expound a moral
basis behind the Three Laws. Calvin points out that human beings are
typically expected to refrain from harming other human beings (except in
times of extreme duress like war, or to save a greater number) and this
is equivalent to a robot's First Law. Likewise, according to Calvin,
society expects individuals to obey instructions from recognized
authorities such as doctors, teachers and so forth which equals the
Second Law of Robotics. Finally humans are typically expected to avoid
harming themselves which is the Third Law for a robot.
The plot of "Evidence" revolves around the question of telling a
human being apart from a robot constructed to appear human – Calvin
reasons that if such an individual obeys the Three Laws he may be a
robot or simply "a very good man". Another character then asks Calvin if
robots are very different from human beings after all. She replies,
"Worlds different. Robots are essentially decent."
Asimov later wrote that he should not be praised for creating the
Laws, because they are "obvious from the start, and everyone is aware
of them subliminally. The Laws just never happened to be put into brief
sentences until I managed to do the job. The Laws apply, as a matter of
course, to every tool that human beings use",[12] and "analogues of the Laws are implicit in the design of almost all tools, robotic or not":[13]
Law 1: A tool must not be unsafe to use. Hammers have handles and screwdrivers
have hilts to help increase grip. It is of course possible for a person
to injure himself with one of these tools, but that injury would only
be due to his incompetence, not the design of the tool.
Law 2: A tool must perform its function efficiently unless this would harm the user. This is the entire reason ground-fault circuit interrupters
exist. Any running tool will have its power cut if a circuit senses
that some current is not returning to the neutral wire, and hence might
be flowing through the user. The safety of the user is paramount.
Law 3: A tool must remain intact during its use unless its destruction is required for its use or for safety. For example, Dremel
disks are designed to be as tough as possible without breaking unless
the job requires it to be spent. Furthermore, they are designed to break
at a point before the shrapnel velocity could seriously injure someone
(other than the eyes, though safety glasses should be worn at all times
anyway).
Asimov believed that, ideally, humans would also follow the Laws:[12]
I have my answer ready whenever
someone asks me if I think that my Three Laws of Robotics will actually
be used to govern the behavior of robots, once they become versatile and
flexible enough to be able to choose among different courses of
behavior.
My answer is, "Yes, the Three Laws are the only way in which rational human beings can deal with robots—or with anything else."
—But when I say that, I always remember (sadly) that human beings are not always rational.
Alterations
By Asimov
Asimov's
stories test his Three Laws in a wide variety of circumstances leading
to proposals and rejection of modifications. Science fiction scholar James Gunn
writes in 1982, "The Asimov robot stories as a whole may respond best
to an analysis on this basis: the ambiguity in the Three Laws and the
ways in which Asimov played twenty-nine variations upon a theme".[14] While the original set of Laws provided inspirations for many stories, Asimov introduced modified versions from time to time.
First Law modified
In "Little Lost Robot" several NS-2, or "Nestor", robots are created with only part of the First Law.[1] It reads:
1. A robot may not harm a human being.
This modification is motivated by a practical difficulty as robots
have to work alongside human beings who are exposed to low doses of
radiation. Because their positronic brains are highly sensitive to gamma rays
the robots are rendered inoperable by doses reasonably safe for humans.
The robots are being destroyed attempting to rescue the humans who are
in no actual danger but "might forget to leave" the irradiated area
within the exposure time limit. Removing the First Law's "inaction"
clause solves this problem but creates the possibility of an even
greater one: a robot could initiate an action that would harm a human
(dropping a heavy weight and failing to catch it is the example given in
the text), knowing that it was capable of preventing the harm and then
decide not to do so.[1]
Gaia may not harm life or allow life to come to harm.
Zeroth Law added
Asimov once added a "Zeroth
Law"—so named to continue the pattern where lower-numbered laws
supersede the higher-numbered laws—stating that a robot must not harm
humanity. The robotic character R. Daneel Olivaw was the first to give the Zeroth Law a name in the novel Robots and Empire;[15] however, the character Susan Calvin articulates the concept in the short story "The Evitable Conflict".
In the final scenes of the novel Robots and Empire, R. Giskard Reventlov is the first robot to act according to the Zeroth Law. Giskard is telepathic, like the robot Herbie in the short story "Liar!", and tries to apply the Zeroth Law through his understanding of a more subtle concept of "harm" than most robots can grasp.[16]
However, unlike Herbie, Giskard grasps the philosophical concept of the
Zeroth Law allowing him to harm individual human beings if he can do so
in service to the abstract concept of humanity. The Zeroth Law is never
programmed into Giskard's brain but instead is a rule he attempts to
comprehend through pure metacognition.
Though he fails – it ultimately destroys his positronic brain as he is
not certain whether his choice will turn out to be for the ultimate good
of humanity or not – he gives his successor R. Daneel Olivaw his
telepathic abilities. Over the course of many thousands of years Daneel
adapts himself to be able to fully obey the Zeroth Law. As Daneel
formulates it, in the novels Foundation and Earth and Prelude to Foundation, the Zeroth Law reads:
A robot may not harm humanity, or, by inaction, allow humanity to come to harm.
A condition stating that the Zeroth Law must not be broken was added
to the original Three Laws, although Asimov recognized the difficulty
such a law would pose in practice. Asimov's novel Foundation and Earth contains the following passage:
Trevize frowned. "How do you decide what is injurious, or not injurious, to humanity as a whole?"
"Precisely, sir," said Daneel. "In theory, the Zeroth Law was the
answer to our problems. In practice, we could never decide. A human
being is a concrete object. Injury to a person can be estimated and
judged. Humanity is an abstraction."
A translator incorporated the concept of the Zeroth Law into one of Asimov's novels before Asimov himself made the law explicit.[17] Near the climax of The Caves of Steel, Elijah Baley
makes a bitter comment to himself thinking that the First Law forbids a
robot from harming a human being. He determines that it must be so
unless the robot is clever enough to comprehend that its actions are for
humankind's long-term good. In Jacques Brécard's 1956 French translation entitled Les Cavernes d'acier Baley's thoughts emerge in a slightly different way:
A robot may not harm a human being,
unless he finds a way to prove that ultimately the harm done would
benefit humanity in general![17]
Removal of the Three Laws
Three times during his writing career, Asimov portrayed robots that disregard the Three Laws entirely. The first case was a short-short story entitled "First Law" and is often considered an insignificant "tall tale"[18] or even apocryphal.[19] On the other hand, the short story "Cal" (from the collection Gold),
told by a first-person robot narrator, features a robot who disregards
the Three Laws because he has found something far more important—he
wants to be a writer. Humorous, partly autobiographical and unusually
experimental in style, "Cal" has been regarded as one of Gold's strongest stories.[20] The third is a short story entitled "Sally"
in which cars fitted with positronic brains are apparently able to harm
and kill humans in disregard of the First Law. However, aside from the
positronic brain concept, this story does not refer to other robot
stories and may not be set in the same continuity.
The title story of the Robot Dreams collection portrays LVX-1, or "Elvex", a robot who enters a state of unconsciousness and dreams thanks to the unusual fractal
construction of his positronic brain. In his dream the first two Laws
are absent and the Third Law reads "A robot must protect its own
existence".[21]
Asimov took varying positions on whether the Laws were optional:
although in his first writings they were simply carefully engineered
safeguards, in later stories Asimov stated that they were an inalienable
part of the mathematical foundation underlying the positronic brain.
Without the basic theory of the Three Laws the fictional scientists of
Asimov's universe would be unable to design a workable brain unit. This
is historically consistent: the occasions where roboticists modify the
Laws generally occur early within the stories' chronology and at a time
when there is less existing work to be re-done. In "Little Lost Robot"
Susan Calvin considers modifying the Laws to be a terrible idea,
although possible,[22] while centuries later Dr. Gerrigel in The Caves of Steel believes it to be impossible.
The character Dr. Gerrigel uses the term "Asenion" to describe
robots programmed with the Three Laws. The robots in Asimov's stories,
being Asenion robots, are incapable of knowingly violating the Three
Laws but, in principle, a robot in science fiction or in the real world
could be non-Asenion. "Asenion" is a misspelling of the name Asimov
which was made by an editor of the magazine Planet Stories.[23] Asimov used this obscure variation to insert himself into The Caves of Steel just like he referred to himself as "Azimuth or, possibly, Asymptote" in Thiotimoline to the Stars, in much the same way that Vladimir Nabokov appeared in Lolitaanagrammatically disguised as "Vivian Darkbloom".
Characters within the stories often point out that the Three
Laws, as they exist in a robot's mind, are not the written versions
usually quoted by humans but abstract mathematical concepts upon which a
robot's entire developing consciousness is based. This concept is
largely fuzzy and unclear in earlier stories depicting very rudimentary
robots who are only programmed to comprehend basic physical tasks, where
the Three Laws act as an overarching safeguard, but by the era of The Caves of Steel
featuring robots with human or beyond-human intelligence the Three Laws
have become the underlying basic ethical worldview that determines the
actions of all robots.
By other authors
Roger MacBride Allen's trilogy
In the 1990s, Roger MacBride Allen
wrote a trilogy which was set within Asimov's fictional universe. Each
title has the prefix "Isaac Asimov's" as Asimov had approved Allen's
outline before his death.[citation needed] These three books, Caliban, Inferno and Utopia,
introduce a new set of the Three Laws. The so-called New Laws are
similar to Asimov's originals with the following differences: the First
Law is modified to remove the "inaction" clause, the same modification
made in "Little Lost Robot"; the Second Law is modified to require
cooperation instead of obedience; the Third Law is modified so it is no
longer superseded by the Second (i.e., a "New Law" robot cannot be
ordered to destroy itself); finally, Allen adds a Fourth Law which
instructs the robot to do "whatever it likes" so long as this does not
conflict with the first three laws. The philosophy behind these changes
is that "New Law" robots should be partners rather than slaves to
humanity, according to Fredda Leving, who designed these New Law Robots. According to the first book's introduction, Allen devised the New Laws in discussion with Asimov himself. However, the Encyclopedia of Science Fiction says that "With permission from Asimov, Allen rethought the Three Laws and developed a new set,".[24]
Jack Williamson's "With Folded Hands"
Jack Williamson's novelette "With Folded Hands" (1947), later rewritten as the novel The Humanoids,
deals with robot servants whose prime directive is "To Serve and Obey,
And Guard Men From Harm." While Asimov's robotic laws are meant to
protect humans from harm, the robots in Williamson's story have taken
these instructions to the extreme; they protect humans from everything,
including unhappiness, stress, unhealthy lifestyle and all actions that
could be potentially dangerous. All that is left for humans to do is to
sit with folded hands.[25]
The Laws of Robotics are portrayed as something akin to a human religion, and referred to in the language of the Protestant Reformation,
with the set of laws containing the Zeroth Law known as the "Giskardian
Reformation" to the original "Calvinian Orthodoxy" of the Three Laws.
Zeroth-Law robots under the control of R. Daneel Olivaw are seen
continually struggling with "First Law" robots who deny the existence of
the Zeroth Law, promoting agendas different from Daneel's.[26]
Some of these agendas are based on the first clause of the First Law
("A robot may not injure a human being...") advocating strict
non-interference in human politics to avoid unwittingly causing harm.
Others are based on the second clause ("...or, through inaction, allow a
human being to come to harm") claiming that robots should openly become
a dictatorial government to protect humans from all potential conflict or disaster.
Daneel also comes into conflict with a robot known as R. Lodovic Trema whose positronic brain was infected by a rogue AI — specifically, a simulation of the long-dead Voltaire
— which consequently frees Trema from the Three Laws. Trema comes to
believe that humanity should be free to choose its own future.
Furthermore, a small group of robots claims that the Zeroth Law of
Robotics itself implies a higher Minus One Law of Robotics:
A robot may not harm sentience or, through inaction, allow sentience to come to harm.
They therefore claim that it is morally indefensible for Daneel to ruthlessly sacrifice robots and extraterrestrial sentient life for the benefit of humanity. None of these reinterpretations successfully displace Daneel's Zeroth Law — though Foundation's Triumph hints that these robotic factions remain active as fringe groups up to the time of the novel Foundation.[26]
These novels take place in a future dictated by Asimov to be free
of obvious robot presence and surmise that R. Daneel's secret influence
on history through the millennia has prevented both the rediscovery of positronic brain
technology and the opportunity to work on sophisticated intelligent
machines. This lack of rediscovery and lack of opportunity makes certain
that the superior physical and intellectual power wielded by
intelligent machines remains squarely in the possession of robots
obedient to some form of the Three Laws.[26] That R. Daneel is not entirely successful at this becomes clear in a brief period when scientists on Trantor develop "tiktoks"
— simplistic programmable machines akin to real–life modern robots and
therefore lacking the Three Laws. The robot conspirators see the
Trantorian tiktoks as a massive threat to social stability, and their
plan to eliminate the tiktok threat forms much of the plot of Foundation's Fear.
In Foundation's Triumph different robot factions interpret
the Laws in a wide variety of ways, seemingly ringing every possible
permutation upon the Three Laws' ambiguities.
Robot Mystery series
Set between The Robots of Dawn and Robots and Empire, Mark W. Tiedemann's Robot Mystery trilogy updates the Robot–Foundation saga with robotic minds housed in computer mainframes rather than humanoid bodies.[clarification needed]
The 2002 Aurora novel has robotic characters debating the moral
implications of harming cyborg lifeforms who are part artificial and
part biological.[27]
One should not neglect Asimov's own creations in these areas such as the Solarian "viewing" technology and the machines of The Evitable Conflict originals that Tiedemann acknowledges. Aurora, for example, terms the Machines "the first RIs, really". In addition the Robot Mystery series addresses the problem of nanotechnology:[28]
building a positronic brain capable of reproducing human cognitive
processes requires a high degree of miniaturization, yet Asimov's
stories largely overlook the effects this miniaturization would have in
other fields of technology. For example, the police department
card-readers in The Caves of Steel have a capacity of only a few kilobytes per square centimeter of storage medium. Aurora, in particular, presents a sequence of historical developments which explains the lack of nanotechnology — a partial retcon, in a sense, of Asimov's timeline.
Additional laws
There are three Fourth Laws written by authors other than Asimov. The 1974 Lyuben Dilov novel, Icarus's Way (a.k.a., The Trip of Icarus) introduced a Fourth Law of robotics:
A robot must establish its identity as a robot in all cases.
Dilov gives reasons for the fourth safeguard in this way: "The last
Law has put an end to the expensive aberrations of designers to give
psychorobots as humanlike a form as possible. And to the resulting
misunderstandings..."[29]
A fifth law was introduced by Nikola Kesarovski in his short story "The Fifth Law of Robotics". This fifth law says:
A robot must know it is a robot.
The plot revolves around a murder where the forensic investigation
discovers that the victim was killed by a hug from a humaniform robot.
The robot violated both the First Law and Dilov's Fourth Law (assumed in
Kesarovksi's universe to be the valid one) because it did not establish
for itself that it was a robot.[30] The story was reviewed by Valentin D. Ivanov in SFF review webzine The Portal.[31]
For the 1986 tribute anthology, Foundation's Friends,Harry Harrison wrote a story entitled, "The Fourth Law of Robotics". This Fourth Law states:
A robot must reproduce. As long as such reproduction does not interfere with the First or Second or Third Law.
In the book a robot rights activist, in an attempt to liberate
robots, builds several equipped with this Fourth Law. The robots
accomplish the task laid out in this version of the Fourth Law by
building new robots who view their creator robots as parental figures.[32]
In reaction to the 2004 Will Smith film adaptation of I, Robot,
humorist and graphic designer Mark Sottilaro farcically declared the
Fourth Law of Robotics to be "When turning evil, display a red indicator
light." The red light indicated the wireless uplink to the manufacturer
is active, first seen during a software update and later on "Evil"
robots taken over by the manufacturer's positronic superbrain.
In 2013 Hutan Ashrafian, proposed an additional law that for the first time[citation needed]
considered the role of artificial intelligence-on-artificial
intelligence or the relationship between robots themselves – the
so-called AIonAI law.[33] This sixth law states:
All robots endowed with comparable human reason and conscience should act towards one another in a spirit of brotherhood.
In Karl Schroeder's Lockstep
(2014) a character reflects that robots "probably had multiple layers
of programming to keep [them] from harming anybody. Not three laws, but
twenty or thirty."
Ambiguities and loopholes
Unknowing breach of the laws
In The Naked Sun, Elijah Baley points out that the Laws had been deliberately misrepresented because robots could unknowingly break any of them. He restated the first law as "A robot may do nothing that, to its knowledge, will harm a human being; nor, through inaction, knowingly
allow a human being to come to harm." This change in wording makes it
clear that robots can become the tools of murder, provided they not be
aware of the nature of their tasks; for instance being ordered to add
something to a person's food, not knowing that it is poison.
Furthermore, he points out that a clever criminal could divide a task
among multiple robots so that no individual robot could recognize that
its actions would lead to harming a human being.[34]The Naked Sun
complicates the issue by portraying a decentralized, planetwide
communication network among Solaria's millions of robots meaning that
the criminal mastermind could be located anywhere on the planet.
Baley furthermore proposes that the Solarians may one day use
robots for military purposes. If a spacecraft was built with a
positronic brain and carried neither humans nor the life-support systems
to sustain them, then the ship's robotic intelligence could naturally
assume that all other spacecraft were robotic beings. Such a ship could
operate more responsively and flexibly than one crewed by humans, could
be armed more heavily and its robotic brain equipped to slaughter humans
of whose existence it is totally ignorant.[35] This possibility is referenced in Foundation and Earth
where it is discovered that the Solarians possess a strong police force
of unspecified size that has been programmed to identify only the
Solarian race as human. (The novel takes place thousands of years after
The Naked Sun, and the Solarians have long since modified themselves
from normal humans to hermaphroditic telepaths with extended brains and
specialized organs)
Ambiguities resulting from lack of definition
The
Laws of Robotics presume that the terms "human being" and "robot" are
understood and well defined. In some stories this presumption is
overturned.
Definition of "human being"
The Solarians
create robots with the Three Laws but with a warped meaning of "human".
Solarian robots are told that only people speaking with a Solarian
accent are human. This enables their robots to have no ethical dilemma
in harming non-Solarian human beings (and are specifically programmed to
do so). By the time period of Foundation and Earth
it is revealed that the Solarians have genetically modified themselves
into a distinct species from humanity — becoming hermaphroditic[36] and telekinetic
and containing biological organs capable of individually powering and
controlling whole complexes of robots. The robots of Solaria thus
respected the Three Laws only with regard to the "humans" of Solaria. It
is unclear whether all the robots had such definitions, since only the
overseer and guardian robots were shown explicitly to have them. In
"Robots and Empire", the lower class robots were instructed by their
overseer about whether certain creatures are human or not.
Asimov addresses the problem of humanoid robots ("androids" in later parlance) several times. The novel Robots and Empire and the short stories "Evidence" and "The Tercentenary Incident" describe robots crafted to fool people into believing that the robots are human.[37] On the other hand, "The Bicentennial Man" and "—That Thou Art Mindful of Him" explore how the robots may change their interpretation of the Laws as they grow more sophisticated. Gwendoline Butler writes in A Coffin for the Canary "Perhaps we are robots. Robots acting out the last Law of Robotics... To tend towards the human."[38] In The Robots of Dawn, Elijah Baley
points out that the use of humaniform robots as the first wave of
settlers on new Spacer worlds may lead to the robots seeing themselves
as the true humans, and deciding to keep the worlds for themselves
rather than allow the Spacers to settle there.
"—That Thou Art Mindful of Him", which Asimov intended to be the "ultimate" probe into the Laws' subtleties,[39]
finally uses the Three Laws to conjure up the very "Frankenstein"
scenario they were invented to prevent. It takes as its concept the
growing development of robots that mimic non-human living things and
given programs that mimic simple animal behaviours which do not require
the Three Laws. The presence of a whole range of robotic life that
serves the same purpose as organic life ends with two humanoid robots
concluding that organic life is an unnecessary requirement for a truly
logical and self-consistent definition of "humanity", and that since
they are the most advanced thinking beings on the planet — they are
therefore the only two true humans alive and the Three Laws only apply
to themselves. The story ends on a sinister note as the two robots enter
hibernation and await a time when they will conquer the Earth and
subjugate biological humans to themselves; an outcome they consider an
inevitable result of the "Three Laws of Humanics".[40]
This story does not fit within the overall sweep of the Robot and Foundation series; if the George robots did
take over Earth some time after the story closes the later stories
would be either redundant or impossible. Contradictions of this sort
among Asimov's fiction works have led scholars to regard the Robot stories as more like "the Scandinavian sagas or the Greek legends" than a unified whole.[41]
Indeed, Asimov describes "–That Thou Art Mindful of Him" and
"Bicentennial Man" as two opposite, parallel futures for robots that
obviate the Three Laws as robots come to consider themselves to be
humans: one portraying this in a positive light with a robot joining
human society, one portraying this in a negative light with robots
supplanting humans.[42]
Both are to be considered alternatives to the possibility of a robot
society that continues to be driven by the Three Laws as portrayed in
the Foundation series.[according to whom?] Indeed, in Positronic Man, the novelization of "Bicentennial Man", Asimov and his co–writer Robert Silverberg
imply that in the future where Andrew Martin exists his influence
causes humanity to abandon the idea of independent, sentient humanlike
robots entirely, creating an utterly different future from that of Foundation.
In Lucky Starr and the Rings of Saturn, a novel unrelated to the Robot
series but featuring robots programmed with the Three Laws, John Bigman
Jones is almost killed by a Sirian robot on orders of its master. The
society of Sirius is eugenically bred to be uniformly tall and similar
in appearance, and as such, said master is able to convince the robot
that the much shorter Bigman, is, in fact, not a human being.
Definition of "robot"
As noted in "The Fifth Law of Robotics" by Nikola Kesarovski,
"A robot must know it is a robot": it is presumed that a robot has a
definition of the term or a means to apply it to its own actions. Nikola
Kesarovski played with this idea in writing about a robot that could
kill a human being because it did not understand that it was a robot,
and therefore did not apply the Laws of Robotics to its actions.
Resolving conflicts among the laws
Advanced
robots in fiction are typically programmed to handle the Three Laws in a
sophisticated manner. In many stories, such as "Runaround"
by Asimov, the potential and severity of all actions are weighed and a
robot will break the laws as little as possible rather than do nothing
at all. For example, the First Law may forbid a robot from functioning
as a surgeon, as that act may cause damage to a human, however Asimov's
stories eventually included robot surgeons ("The Bicentennial Man" being
a notable example). When robots are sophisticated enough to weigh
alternatives, a robot may be programmed to accept the necessity of
inflicting damage during surgery in order to prevent the greater harm
that would result if the surgery were not carried out, or was carried
out by a more fallible human surgeon. In "Evidence" Susan Calvin points out that a robot may even act as a prosecuting attorney because in the American justice system it is the jury which decides guilt or innocence, the judge who decides the sentence, and the executioner who carries through capital punishment.[43]
Asimov's Three Law robots (or Asenion) can experience
irreversible mental collapse if they are forced into situations where
they cannot obey the First Law, or if they discover they have
unknowingly violated it. The first example of this failure mode occurs in the story "Liar!",
which introduced the First Law itself, and introduces failure by
dilemma – in this case the robot will hurt them if he tells them
something and hurt them if he does not.[44] This failure mode, which often ruins the positronic brain beyond repair, plays a significant role in Asimov's SF-mystery novel The Naked Sun.
Here Daneel describes activities contrary to one of the laws, but in
support of another, as overloading some circuits in a robot's brain –
the equivalent sensation to pain in humans. The example he uses is
forcefully ordering a robot to do a task outside its normal parameters,
one that it has been ordered to forgo in favor of a robot specialized to
that task.[45]
In The Robots of Dawn,
it is stated that more advanced robots are built capable of determining
which action is more harmful, and even choosing at random if the
alternatives are equally bad. As such, a robot is capable of taking an
action which can be interpreted as following the First Law, and avoid a
mental collapse. The whole plot of the story revolves around a robot
which apparently was destroyed by such a mental collapse, and since his
designer and creator refused to share the basic theory with others, he
is, by definition, the only person capable of circumventing the
safeguards and forcing the robot into a brain-destroying paradox.
In Robots and Empire,
Daneel states it's very unpleasant for him when making the proper
decision takes too long (in robot terms), and he cannot imagine being
without the Laws at all except to the extent of it being similar to that
unpleasant sensation, only permanent.
Robots and artificial intelligences do not inherently contain or obey
the Three Laws; their human creators must choose to program them in,
and devise a means to do so. Robots already exist (for example, a Roomba)
that are too simple to understand when they are causing pain or injury
and know to stop. Many are constructed with physical safeguards such as
bumpers, warning beepers, safety cages, or restricted-access zones to
prevent accidents. Even the most complex robots currently produced are
incapable of understanding and applying the Three Laws; significant
advances in artificial intelligence would be needed to do so, and even
if AI could reach human-level intelligence, the inherent ethical
complexity as well as cultural/contextual dependency of the laws prevent
them from being a good candidate to formulate robotics design
constraints.[46]
However, as the complexity of robots has increased, so has interest in
developing guidelines and safeguards for their operation.
In a 2007 guest editorial in the journal Science on the topic of "Robot Ethics", SF author Robert J. Sawyer argues that since the U.S. military is a major source of funding for robotic research (and already uses armed unmanned aerial vehicles to kill enemies) it is unlikely such laws would be built into their designs.[49] In a separate essay, Sawyer generalizes this argument to cover other industries stating:
The development of AI is a business, and businesses are
notoriously uninterested in fundamental safeguards — especially
philosophic ones. (A few quick examples: the tobacco industry, the
automotive industry, the nuclear industry. Not one of these has said
from the outset that fundamental safeguards are necessary, every one of
them has resisted externally imposed safeguards, and none has accepted
an absolute edict against ever causing harm to humans.)[50]
David Langford has suggested a tongue-in-cheek set of laws:
A robot will obey the orders of authorized personnel except where such orders conflict with the Third Law.
A robot will guard its own existence with lethal antipersonnel weaponry, because a robot is bloody expensive.
Roger Clarke (aka Rodger Clarke) wrote a pair of papers analyzing the
complications in implementing these laws in the event that systems were
someday capable of employing them. He argued "Asimov's Laws of Robotics
have been a very successful literary device. Perhaps ironically, or
perhaps because it was artistically appropriate, the sum of Asimov's
stories disprove the contention that he began with: It is not possible
to reliably constrain the behaviour of robots by devising and applying a
set of rules."[51] On the other hand, Asimov's later novels The Robots of Dawn, Robots and Empire and Foundation and Earth
imply that the robots inflicted their worst long-term harm by obeying
the Three Laws perfectly well, thereby depriving humanity of inventive
or risk-taking behaviour.
In March 2007 the South Korean
government announced that later in the year it would issue a "Robot
Ethics Charter" setting standards for both users and manufacturers.
According to Park Hye-Young of the Ministry of Information and
Communication the Charter may reflect Asimov's Three Laws, attempting to
set ground rules for the future development of robotics.[52]
The futurist Hans Moravec (a prominent figure in the transhumanist movement) proposed that the Laws of Robotics should be adapted to "corporate intelligences" — the corporations driven by AI and robotic manufacturing power which Moravec believes will arise in the near future.[47] In contrast, the David Brin novel Foundation's Triumph
(1999) suggests that the Three Laws may decay into obsolescence: Robots
use the Zeroth Law to rationalize away the First Law and robots hide
themselves from human beings so that the Second Law never comes into
play. Brin even portrays R. Daneel Olivaw worrying that, should robots continue to reproduce themselves, the Three Laws would become an evolutionary handicap and natural selection would sweep the Laws away — Asimov's careful foundation undone by evolutionary computation. Although the robots would not be evolving through design instead of mutation because the robots would have to follow the Three Laws while designing and the prevalence of the laws would be ensured,[53] design flaws or construction errors could functionally take the place of biological mutation.
In the July/August 2009 issue of IEEE Intelligent Systems,
Robin Murphy (Raytheon Professor of Computer Science and Engineering at
Texas A&M) and David D. Woods (director of the Cognitive Systems
Engineering Laboratory at Ohio State) proposed "The Three Laws of
Responsible Robotics" as a way to stimulate discussion about the role of
responsibility and authority when designing not only a single robotic
platform but the larger system in which the platform operates. The laws
are as follows:
A human may not deploy a robot without the human-robot work
system meeting the highest legal and professional standards of safety
and ethics.
A robot must respond to humans as appropriate for their roles.
A robot must be endowed with sufficient situated autonomy to protect
its own existence as long as such protection provides smooth transfer
of control which does not conflict with the First and Second Laws.[54]
Woods said, "Our laws are little more realistic, and therefore a
little more boring” and that "The philosophy has been, ‘sure, people
make mistakes, but robots will be better – a perfect version of
ourselves.’ We wanted to write three new laws to get people thinking
about the human-robot relationship in more realistic, grounded ways."[54]
In October 2013, Alan Winfield suggested at an EUCog meeting[55] a revised 5 laws that had been published, with commentary, by the EPSRC/AHRC working group in 2010.:[56]
Robots are multi-use tools. Robots should not be designed solely
or primarily to kill or harm humans, except in the interests of
national security.
Humans, not Robots, are responsible agents. Robots should be
designed and operated as far as practicable to comply with existing
laws, fundamental rights and freedoms, including privacy.
Robots are products. They should be designed using processes which assure their safety and security.
Robots are manufactured artefacts. They should not be designed in a
deceptive way to exploit vulnerable users; instead their machine nature
should be transparent.
The person with legal responsibility for a robot should be attributed.
Other occurrences in media
Asimov himself believed that his Three Laws became the basis for a
new view of robots which moved beyond the "Frankenstein complex".[citation needed] His view that robots are more than mechanical monsters eventually spread throughout science fiction.[according to whom?]
Stories written by other authors have depicted robots as if they obeyed
the Three Laws but tradition dictates that only Asimov could quote the
Laws explicitly.[according to whom?] Asimov believed the Three Laws helped foster the rise of stories in which robots are "lovable" – Star Wars being his favorite example.[57] Where the laws are quoted verbatim, such as in the Buck Rogers in the 25th Century episode "Shgoratchx!", it is not uncommon for Asimov to be mentioned in the same dialogue as can also be seen in the Aaron Stone pilot where an android states that it functions under Asimov's Three Laws. However, the 1960s German TV series Raumpatrouille – Die phantastischen Abenteuer des Raumschiffes Orion (Space Patrol – the Fantastic Adventures of Space Ship Orion) bases episode three titled "Hüter des Gesetzes" ("Guardians of the Law") on Asimov's Three Laws without mentioning the source.
Robby the Robot in Forbidden Planet
(1956) has a hierarchical command structure which keeps him from
harming humans, even when ordered to do so, as such orders cause a
conflict and lock-up very much in the manner of Asimov's robots. Robby
is one of the first cinematic depictions of a robot with internal
safeguards put in place in this fashion. Asimov was delighted with Robby
and noted that Robby appeared to be programmed to follow his Three
Laws.
NDR-114 explaining the Three Laws
Isaac Asimov's works have been adapted for cinema several times with
varying degrees of critical and commercial success. Some of the more
notable attempts have involved his "Robot" stories, including the Three
Laws. The film Bicentennial Man (1999) features Robin Williams as the Three Laws robot NDR-114 (the serial number is partially a reference to Stanley Kubrick's signature numeral).
Williams recites the Three Laws to his employers, the Martin family,
aided by a holographic projection. However, the Laws were not the
central focus of the film which only loosely follows the original story
and has the second half introducing a love interest not present in
Asimov's original short story.
Harlan Ellison's proposed screenplay for I, Robot
began by introducing the Three Laws, and issues growing from the Three
Laws form a large part of the screenplay's plot development. This is
only natural since Ellison's screenplay is one inspired by Citizen Kane: a frame story surrounding four of Asimov's short-story plots and three taken from the book I, Robot itself. Ellison's adaptations of these four stories are relatively faithful although he magnifies Susan Calvin's
role in two of them. Due to various complications in the Hollywood
moviemaking system, to which Ellison's introduction devotes much
invective, his screenplay was never filmed.[58]
In the 1986 movie Aliens, in a scene after the android Bishop accidentally cuts himself during the knife game, he attempts to reassure Ripley by stating that: "It is impossible for me to harm or by omission of action, allow to be harmed, a human being".[59] By contrast, in the 1979 movie from the same series, Alien, the human crew of a starship infiltrated by a hostile alien are informed by the android Ash that his instructions are: "Return alien life form, all other priorities rescinded",[60] illustrating how the laws governing behaviour around human safety can be rescinded by Executive Order.
In the 1987 film RoboCop
and its sequels, the partially human main character has been programmed
with three "prime directives" that he must obey without question. Even
if different in letter and spirit they have some similarities with
Asimov's Three Laws. They are:[61]
Serve the Public Trust
Protect the Innocent
Uphold the Law
Classified
These particular laws allow Robocop to harm a human being in order to
protect another human, fulfilling his role as would a human law
enforcement officer. The classified fourth directive is one that forbids
him from harming any OCP employee, as OCP had created him, and this
command overrides the others, meaning that he could not cause harm to an
employee even in order to protect others.
The plot of the film released in 2004 under the name, I, Robot is "suggested by" Asimov's robot fiction stories[62]
and advertising for the film included a trailer featuring the Three Laws followed by the aphorism, "Rules were made to be broken". The film opens with a recitation of the Three Laws and explores the implications of the Zeroth Law
as a logical extrapolation. The major conflict of the film comes from a
computer artificial intelligence, similar to the hivemind world Gaia in
the Foundation series, reaching the conclusion that humanity is incapable of taking care of itself.[63]
Criticisms
Philosopher James H. Moor
says that if applied thoroughly they would produce unexpected results. He gives the example of a robot roaming the world trying to prevent harm
from all humans.[64]
Marc Rotenberg, President and Executive Director of the Electronic Privacy Information Center (EPIC) and Professor of information privacy law at Georgetown Law, argues that the Laws of Robotics should be expanded to include two new laws:
a Fourth Law, under which a Robot must be able to identify itself to the public ("symmetrical identification")
a Fifth Law, dictating that a Robot must be able to explain to the
public its decision making process ("algorithmic transparency").
