Abstract
Artificial intelligence (AI) is a burgeoning technology poised to revolutionize the world and bring significant change to global society. AI offers new capabilities and efficiency for humanity in the areas of finance, governance, healthcare, criminal justice, energy, and countless other fields and industries. However, similar to a double-edged sword, the technology also generates serious risks, particularly for the most vulnerable in society, by potentially furthering inequality and possibly violating intrinsic rights. This paper seeks to analyze and evaluate this dual nature of AI and its consequences for humanity. Focusing on the use of narrow AI, combined with an emphasis on human rights, this paper dives into the effects of AI in the fields of the economy, the healthcare system, the environment, and the government, determining the benefits, areas for monitoring and improvement, and solutions to possible threats stemming from AI. To achieve a deep and complex discussion of these aspects of AI, this paper employs a dialectical approach, viewing the technology from the lens of multiple, often conflicting perspectives, to fully examine AI.
Introduction
Artificial intelligence (AI) is a rapidly growing sector that is on the verge of revolutionizing the world. Already, the technology is deeply integrated in global industries like finance, healthcare, and criminal justice (Pilat, 2019). Google CEO Sundar Pichai recently described AI as “more profound than fire or electricity” (Petroff, 2018), and the technology has an already expansive footprint, “impacting… the way we shop, communicate… [and] research” (Franke, 2019).
Due to the nature of AI, this technology “can be incredibly helpful, but… can also be used to wreak havoc” (Sydell, 2020). Unfortunately, the black-box opacity of AI algorithms – which makes it challenging to follow their decision-making processes – presents credible risks to human rights, with the potential to exacerbate prejudices, worsen societal disparities, and provide tools for the subjugation of people and their rights (Alang, 2017; Bleicher, 2017).
Knowledge grows and become more accurate from debate, discussion, and the expression of free speech, particularly the kind of speech that opposes and criticizes long-held, typical beliefs. Based on this principle, this paper utilizes a dialectical approach, observing conflicting ideas regarding AI in order to obtain a more accurate assessment of the technology. Specifically, the following essay seeks to analyze the materialized and potential impacts of AI in the fields of the economy, healthcare, the environment, and governance and evaluate its benefits, harms, and possible optimal implementations.
What is Artificial Intelligence?
While the surrounding public perception focuses on futuristic technology, the term “artificial intelligence” has been around for decades and was first used by prominent computer scientist and Turing Award winner John McCarthy in a 1956 Dartmouth Conference (Joshi & Mishra, 2010). Despite this history, there is “no single definition of AI that is universally accepted” (Holdren et al., 2016). Instead, artificial intelligence is “an umbrella term that includes a variety of computational techniques and associated processes dedicated to improving the ability of machines to do things requiring intelligence, such as pattern recognition, computer vision, and language processing” (Raso et al., 2018).
At its core, AI can be classified into “narrow” and “general” types (Holdren et al., 2016; Kurzweil, 2005). Most of the AI algorithms developed and utilized currently are examples of narrow AI, focusing on “addressing specific application areas,” such as self-driving cars, maps, language processing, and image recognition (Holdren et al., 2016). Importantly, narrow AI needs “some level of human reprogramming or reconfiguration… to enable the system to retain its level of intelligence;” in other words, a narrow AI system has a limited range of capabilities and typically requires human intervention (Goertzel, 2014).
The most widely used form of narrow AI is a technique known as “machine learning,” which is “a statistical process that starts with a body of data and tries to derive a rule or procedure that explains the data or can predict future data” (Holdren et al., 2016). A more advanced subfield of machine learning is “deep learning,” which uses “structures loosely inspired by the human brain, consisting of a set of units (‘neurons’)” to recognize complex patterns in large sets of data (Holdren et al., 2016).
On the other hand, general AI, also known as “artificial general intelligence” (AGI), is more speculative and has yet to be achieved (Holdren et al. 2016; Goertzel, 2014). An AGI system seeks to reach general intelligence, human-like or not, “across the full range of cognitive tasks” (Holdren et al. 2016), allowing it to “achieve a variety of goals, and carry out a variety of tasks, in a variety of different contexts and environments” (Goertzel, 2014).
In this paper, we adopt the stance of Dr. David Tuffley: Currently, “the correct perspective is to see AI as a capability extending adjunct to human intelligence, allowing us to do things that could not be done unaided” (Tuffley, 2019). Thus, this paper will limit the discussion of AI to applications of narrow AI that are currently in use or are far along the development process to ground our dialogue.
A Human Rights Perspective
This paper examines the consequences of AI from the perspective of human rights, which provide a fundamental framework of norms to assess its effects on society while also building on an already-established and shared system of evaluation that can be engaged by several parties (Pielemeier, 2019). The basis of human rights lies in the International Bill of Rights, the combination of three guiding international documents: the United Declaration of Human Rights (UDHR); the International Covenant on Civil and Political Rights (ICCPR); and the International Covenant on Economic, Social, and Cultural Rights (ICESCR) (Latonero, 2018; Raso et al., 2018). These three documents, the latter two of which have been ratified by nearly 170 countries, outline the framework for human rights in the international community (Latonero, 2018).
The UDHR, a non-binding United Nations General Assembly resolution, is “the leading statement of the rights that every human being enjoys by virtue of their birth” (Raso et al., 2018; Universal Declaration of Human Rights, 1948). On the other hand, the ICCPR and ICESCR are binding international treaties that “elaborate upon the human rights that were first articulated by the UDHR at the international level, and clarify the duties of states in relation to two categories of rights” (Raso et al., 2018; International Covenant on Civil and Political Rights, 1966; International Covenant on Economic, Social, and Cultural Rights, 1966).
Notably, these documents focus on the role of governments in protecting human rights (Raso et al., 2018); however, private businesses and citizens hold a similar obligation to preserve these rights as well, as outlined in the United Nations Guiding Principles on Business and Human Rights (UNGP) (Raso et al., 2018; Guiding Principles on Business and Human Rights, 2011). In this paper, these documents and the rights articulated within – including civil, political, social, cultural, and economic rights – will provide the foundation on which AI’s human rights obligations and implications are to be evaluated.
AI and the Economy
The economic consequences of AI are expected to be staggering. A report by PricewaterhouseCoopers estimated that, by 2030, AI’s productivity improvements will raise global GDP by 14 percent—nearly $15.7 trillion (Rao et al., 2017). By 2035, AI may also boost profits by nearly 40 percent and “increase economic growth rates by a weighted average of 1.7 percentage points… across 16 industries” (Purdy et al., 2017). Unfortunately, the effects of such growth will likely be complex and unequitable; because of the disruptive consequences of automation, AI may create short-term frictional unemployment & exacerbate economic inequality, but the technology may also provide new job opportunities and a higher standard of living in the long term (Holdren et al., 2016).
Research from Oxford University found that “about 40% of jobs in Europe…, almost half of jobs in the USA, and an even greater share in developing countries” will be threatened by AI “over the coming decades” (Goldin, 2019; Frey & Osborne, 2013). Specifically, AI will likely replace low-skill, repetitive jobs, such as truckers and telemarketers, that are filled disproportionately by lower income laborers (Goldin, 2019; Frey & Osborne, 2013). Critics of AI believe that such labor displacement targeted toward low-skill workers presents a serious risk of increasing global inequality, both within nations and between developed and developing nations, entrenching many in poverty (Goldin, 2019; Bartlett, 2019; Lee, 2017).
However, as seen during the era of 19th Century industrialization, some workforce disruption is inevitable with technological advancements (Thompson, 2017), and there is optimism that AI will create new economic opportunities to overcome these challenges.
According to a report by the McKinsey Global Institute, AI automation may actually increase the size of the workforce for three reasons. First, it creates new industries and categories of jobs; for example, AI requires low-skill workers to annotate and develop its data sets and high-skill workers to create and maintain its models (Manyika et al., 2017; Kande & Sönmez, 2020). Indeed, the World Economic Forum estimated that AI will create a total net gain of 58 million jobs, signifying a benefit for workers (Hanspal, 2021; Leopold et al., 2018). Second, the McKinsey report finds that AI enables greater productivity, increasing the incomes of people, and, as many spend their higher incomes, it creates demand for new goods and services across the economy, further increasing employment (Manyika et al., 2017). Third, AI’s productivity advancements provide more time for recreational and leisure activities, generating jobs in these sectors, such as tourism and entertainment; already, tourism accounts for nearly 300 million jobs globally or “one in every ten jobs on the planet” (Manyika et al., 2017).
Critically, governments and business will be vital to maintaining a stable workforce in a tumultuous transition period of frictional unemployment and workforce realignment; policy makers “[e]nsuring robust demand growth and economic dynamism is a priority” for continuous job growth, and “[m]idcareer job training will be essential” to workers transitioning careers (Manyika et al., 2017; Holdren et al., 2016).
Additionally, some believe that, by increasing competition from automation and forcing laid-off workers to flood into service jobs, AI will depress wages (Schlogl & Sumner, 2020). The Industrial Revolution provides empirical data to the contrary; MIT economist Erik Brynjolfsson noted that “[a]verage wages have been increasing for the past 200 years,” since industrialization has been “creating wealth” for all (Thompson 2017). Moreover, AI innovation, as measured by patent grants, benefits companies and raises wages; one study found that, “regardless of the industrial sector of the firm, every AI patent granted contributes to higher TFP [a measure of productivity] by 3.2%…” and “increases a firm’s wages by 1.5%” (Bassetti et al., 2020).
Overall, because productivity is the driver of economic growth, AI will be crucial to the economic prosperity and stability of many across the world. As Associate Professor Betsey Stevenson of the University of Michigan explains, “[p]roductivity growth ultimately gives us better lives and more options” (Stevenson, 2019).
AI in Healthcare
Healthcare represents the industry with the greatest potential for significant advancement due to AI. First, Dr. Bertalan Meskó notes that there is a shortage of “about 17.4 million” healthcare professionals globally, resulting in gaps “in access to care” and “differing [treatment] quality… worldwide” (Meskó et al., 2018). Moreover, the Future Advocacy think tank writes that nurses spend “2.5 million hours a week on clerical tasks” (Fenech et al. 2018) instead of patient care.
Thankfully, an Accenture report finds that AI addresses this “healthcare workforce crisis” (Meskó et al., 2018) by giving “workers tools to do their jobs better” (Collier et al.,2017). Specifically, “AI can help remove or minimize time spent on routine, administrative tasks, which can take up to 70 percent of a healthcare practitioner’s time” (Spatharou et al.); by “analys[ing] medical notes, automatedly complet[ing] forms” (Fenech et al. 2018), and performing other clerical tasks, AI decreases wasted time by “17 percent for doctors [and] 51 percent for… nurses” (Collier et al., 2017), allowing them to focus on patients’ wellbeing, increasing “the quality and the safety of outcomes” (Meskó et al. 2018). Additionally, a healthcare study found that AI robotic surgeries generated “a 21% reduction in patients’ stay” (Kalis et al., 2018), translating to “speed[ier] patient recoveries” as patients were discharged “three days earlier” (Salzman, 2019). Subsequently, Accenture concludes that AI may “address an estimated 20 percent of unmet clinical demand” (Collier et al., 2017).
Second, Future Advocacy found that AI is playing “an expanded role in… diagnostics and treatment” due to “the reliance of modern medicine on ever-increasing… data” (Fenech et al. 2018). Crucially, because AI handles large amounts of data effectively, it allows for earlier and more accurate diagnoses of conditions such as coronary aneurysms, brain bleeds, osteoporosis, and cardiac arrests (Bernaert and Akpakwu, 2018; Salzman, 2019). Empirically, AI systems in California and North Carolina have helped reduce ICU deaths by 58 percent and predicted cardiac arrest well in advance (Salzman, 2019). AI systems have also detected lung cancer with 94 percent accuracy and breast cancer with 99 percent accuracy—and at speeds 30 times faster than an average medical professional (Powell, 2020). Consequently, the UK government forecasted that AI will “prevent 22,000 deaths from cancer each year by 2033 and give patients an additional five years of healthy, independent life” through early diagnosis (Perkins, 2018).
Moreover, AI has already been playing a key role during the COVID-19 pandemic. AI models detected warning signs of the crisis as early as the end of 2019 (Heaven, 2020). During the COVID-19 pandemic, AI has been a valuable tool in helping with diagnosis, treatment, and vaccine development (Arora et al., 2021). In the future, AI may be “a giant firewall against… pandemics… [since] AI has the capacity to quickly search enormous databases for an existing drug… or develop a new one in… months” (Chandler, 2020).
Third, AI helps develop more effective and more accessible prostheses and assistance for people with disabilities. For example, AI wheelchairs and walking aids greatly help people with disabilities to “perform daily activities,” which “enhances [their] quality of life with ease in performance” (Nayak & Das, 2020). Other researchers found that AI prosthetic hands and limbs can “fulfill the same functions as natural ones” and “stimulate the arm to move as close as possible with the human one” (Kuts and Bozhok, 2020).
AI’s impact on people with disabilities isn’t limited to prostheses. Google’s Project Euphonia uses AI to enhance speech recognition technology for those with speech impediments, looking to make these effective solutions more accessible (United Spinal, 2020). Other tools, ranging from smart assistants like Amazon’s Alexa to apps like Ava, have helped thousands of hearing-and-visually-impaired people, improving their lives (Snow, 2019). Moreover, AI applications in digital ophthalmology have been used for eye screening to detect cataracts and diabetic retinopathy – leading causes of blindness that have affected millions worldwide (Gunasekeran & Wong, 2020; Goh et al., 2020). Critically, AI-enhanced eye screening, which can be attached to smartphones, may “provide better outreach for cataract screening, especially in rural or less-resourced areas” (Goh et al., 2020), which may “facilitate more widespread and cost-effective screening” against blindness (Wong & Sabanayagam, 2019; Martiniello et al., 2019). In addition to healthcare quality and accessibility improvements, these AI healthcare tools “could help integrate a segment of our population that has often been left out of routine daily life activities and job opportunities,” reducing a disproportionately high unemployment rate among people with disabilities (Snow, 2019).
Unfortunately, AI improvements in healthcare are predicated on data that may be entrenched in significant bias, worsening disparities (Kent, 2020; Ross et al., 2020). A recent study revealed AI’s widespread potential for perpetuating discrimination in healthcare, with one algorithm prioritizing the treatment of white patients over equally sick African American ones (Ledford, 2019; McCullom, 2020). In fact, although African Americans have disproportionately higher rates of pre-existing conditions, they received, on average, $1,800 less in provided healthcare than white patients with similar conditions (Ledford, 2019; McCullom, 2020). Worryingly, AI algorithms like these manage the healthcare of 200 million Americans (Ledford, 2019; Ross et al., 2020), revealing deep inequalities within the current system.
Thankfully, the solution to these disparities may lie in AI itself: algorithms and models are being developed today to detect and mitigate bias in medical data. For example, one AI model was able to detect and reduce racial disparities in data for the National Institute of Health, combatting biases (Hao, 2021). Also, with a greater sense of awareness on this issue in the industry, companies like IBM are now trying to use larger, more representative data sets and are reworking AI models to reduce biases in an effort to “make decisions that don’t unfairly discriminate” (Totty, 2020),
Another concern is that the substantial amount of medical data needed to build and train AI models infringes on patients’ right to privacy (Rao et al. 2017; Tom et al., 2020). Worse, there exist legitimate risks that patient data used for AI algorithms can be breached to access private personal and financial information or impact one’s employment or insurance coverage (Tom et al., 2020). Many fear that the current American law regulating healthcare privacy – the Health Insurance Portability and Accountability Act (HIPAA) – has “significant gaps” in protecting patients’ data with respect to AI and should be expanded to a model similar to the European General Data Protection Regulation (GDPR) instead (Gerke et al., 2020). Thankfully, there are new approaches to AI currently being developed to protect patient privacy, such as differential privacy, federated learning, and distributed AI models (Tom et al., 2020), and updates to regulatory frameworks with consideration of recent technological advances can create a climate in which privacy is valued and secured (Gerke et al., 2020).
Ultimately, healthcare is a field ripe for significant breakthroughs due to AI, preserving the right to life and ensuring access to better and cheaper healthcare for millions. Accordingly, research firm Frost & Sullivan found that the previous advancements and AI’s “added layer of decision support… helping [to] mitigate… errors” can “improve [health] outcomes by 30 to 40 percent” (Frost & Sullivan, 2016), which Biundo et al., 2020 explains may save “annually 380,000 to 403,000 lives” in Europe alone (Biundo et al., 2020).
AI and the Environment
Threatening “the premature extinction of Earth-originating intelligent life or the permanent destruction of its potential for desirable future development,” climate change is an “existential national security risk” and may be the greatest challenge humanity will face (Spratt & Dunlop, 2019; Rolnick et al., 2019). Without intervention, this imminent climate crisis will “escalate cycles of humanitarian and socio-political crises, conflict and forced migration” (Spratt & Dunlop, 2019; Carleton & Hsiang, 2016; openDemocracy, 2019). Thankfully, AI may play a significant role in addressing climate change in several ways.
First, AI improves production from renewable sources; it “can both reduce emissions from… generators and enable the transition to carbon-free systems by helping improve… forecasting, scheduling, and control… and… create advanced electricity markets that accommodate… variable electricity and flexible demand” (Donti, 2019). AI boosts production through forecasting the generation and demand fluctuations “to inform real-time electricity scheduling and longer-term system planning,” which can “reduce… reliance on polluting standby plants and… manage increasing amounts of variable sources” (Donti, 2019). AI also increases access to renewables by allowing sustainable microgrids in rural areas “through accurate forecasts of demand and power production” within data-intensive networks, establishing long term clean energy systems and reducing costs of using non-fossil fuel energy (Donti, 2019).
Electricity generation is responsible for “a quarter of human-caused greenhouse gas emissions each year,” but the AI-led shift from fossil fuels to renewables impacts every sector, resulting in a global energy transition (Donti, 2019). “The use of AI in the power sector…may have a critical impact as…[it] has the potential to cut energy waste, lower energy costs, and facilitate and accelerate the use of clean renewable energy sources in power grids worldwide” (Makala & Bakovic, 2020). Finally, “with the integration of [AI] in renewable energy sources, an increase in energy efficiency does not seem far off” (Kumar, 2018).
Second, AI is a force multiplier for increasing energy efficiency in industries, cities, and transportation, resulting in significant reductions in greenhouse gas (GHG) emissions. A report by two AI and climate researchers quantifies those buildings, accounting for “a quarter of global energy-related emissions,” can experience energy usage reductions “by up to 90%.” By “modeling energy consumption” of systems like HVAC and “allowing devices and systems to adapt to usage patterns” in a more efficient manner, AI makes building systems “radically more [energy] efficient” (Milojevic-Dupont & Kaack, 2019). When applied on a city-wide scale, AI “can reduce GHG emissions by coordinating between infrastructure sectors and better adapting services” (Milojevic-Dupont & Kaack, 2019).
AI also makes industrial production and logistics, the “leading causes of difficult-to-eliminate GHG emissions,” more efficient (Waldman-Brown, 2019). Specifically, AI optimizes industries by “streamlin[ing] supply chains, impro[ving] production quality, predict[ing] machine breakdowns, optimiz[ing] heating and cooling systems, and prioritiz[ing] the use of clean electricity” (Waldman-Brown, 2019).
Third, in order for “global emissions … [to] become net-negative,” AI “may help with many aspects of [carbon] sequestration,” such as “identify[ing]… potential storage locations” for CO2, “monitor[ing] and maintain[ing] active sequestration sites… [and] monitor[ing] for CO2 leaks” (Ross & Sherwin, 2019). Additionally, “up to 0.9 billion hectares” of trees can be planted, “a means of sequestering CO2 over the long term.” However, “to ensure a positive impact,” such afforestation must consider “local conditions and native species” (Lacoste, 2019). AI “can [help] in automating large-scale afforestation by locating appropriate planting sites, monitoring plant health, assessing weeds, and analyzing trends,” and AI startups can already plant “more quickly and cheaply than traditional methods” (Lacoste, 2019).
AI also allows humans to adapt to the changing climate. With rising temperatures and frequent natural disasters, “AI can make climate models more accurate, predicting more exact future temperature increases” and “extreme weather” (Brightman, 2019). Specifically, AI models can “simulate weather events and natural disasters to find vulnerabilities in disaster planning, determine which strategies for disaster response are most effective, and provide real-time disaster response coordination” (Cho et al., 2018). Recent analysis of NOAA’s AI algorithm “in predicting high weather events” was preferred over human intuition in “75% of cases,” which is critical since more accurate forecasting from AI “is essential to aid preparation” (Huntingford et al., 2019).
It should be noted that the use of AI has a relatively significant carbon footprint: “Modern AI models consume a massive amount of energy, and these energy requirements are growing at a breathtaking rate” (Toews, 2020); there has been a “300,000x increase between 2012 and 2018” in the computational and energy resources to develop models (Toews, 2020; Amodei & Hernandez, 2020). Moreover, recent estimates of AI models found that “[t]raining a single… model can generate up to 626,155 pounds of CO2 emissions—roughly equal to the total lifetime carbon footprint of five cars” (Toews, 2020; Strubell et al., 2019; Hao, 2019).
However, “large energy savings” arise with “smaller model[s],” since they “will be used many times in [their] lifetime” (Saenko, 2020). Furthermore, new developments are seeking to address this issue. MIT researchers recently developed an “automated AI system for training… neural networks” that improved “the computational efficiency of the system” and “reduced [its] footprint by orders of magnitude,” expending “roughly 1/1,300 the carbon emissions” of current systems (Matheson, 2020). Google used “DeepMind’s machine learning… to reduce the amount of energy used for cooling by up to 40 percent, a huge improvement” for such a “large scale energy-consuming environment” (Evans & Gao, 2016). Another system at the University College London “uses memristors to create artificial neural networks” that are “at least 1,000 times more energy efficient,” making “[e]xtremely energy-efficient artificial intelligence… now closer to reality” (Greaves, 2020).
Ultimately, solving or even reducing climate change is critical, since its consequences on human life and rights are detrimental. Experts believe that, due to rising sea levels and more frequent natural disasters, more than half of the global population will face lethal conditions, ecosystems will collapse, two billion people will face water scarcity, more than one billion will be displaced, the risk of armed conflicts will increase by 50 percent in Africa alone, and nuclear war is more likely (Spratt & Dunlop, 2019; Kelley et al., 2015; McDonald, 2018; Carleton & Hsiang, 2016; openDemocracy, 2019). Worse, these crises spur the rise of autocracies “in poorer countries … [and] wealthy industrialized nations” (McDonald, 2018; openDemocracy, 2019; Kelley et al., 2015), with regional resource wars generating crises fueling far-right movements and the infringement of human rights worldwide (McDonald, 2018; Carleton & Hsiang, 2016; Kelley et al., 2015). Clearly, a climate revolution is in dire need, and AI may lead the way.
AI and Governance
AI algorithms generate significant risks to people’s rights to fair and free governance; however, new developments may help in the fight for democratic principles and against authoritarian states.
China is one of the leaders in AI-led authoritarianism and oppression (Romaniuk & Burgers, 2018). China’s advanced system of AI surveillance employs services to “track the local Uyghur population…” (Roberts et al., 2020), who have been detained and sent to imprisonment camps (Maizland, 2021). Worse, China has become a “model police state” to countries in across the world who have begun implementing similarly oppressive tactics (Romaniuk & Burgers, 2018), and, unfortunately, AI has helped with “abetting repressive regimes and…accelerating a global resurgence of authoritarianism” (Feldstein, 2019).
Thankfully, while AI may be a tool for authoritarian regimes, it also has the potential to support democracy and protect political rights: “The algorithmic tools that are used to mislead, misinform and confuse could equally be repurposed to support democracy” (Polonski, 2017). AI systems have been developed to fact-check misinformation, promote digital accountability, and detect deep fake content (Johnson, 2020). AI can also bring civic information into voters’ hands and make governments more aware of their people’s needs, supporting political rights and counteracting propaganda (Polonski, 2017). In addition to these AI-centric solutions, combatting AI authoritarianism and infringement of political rights will require major democracies to resist oppression in the international forum and establish global norms for transparent, rights-respecting AI development (Wright, 2020). Overall, AI’s influence on political rights is substantial and potentially devastating, but continued pressure from leading democracies and new tools to combat authoritarianism will be essential in preserving these rights and safeguarding freedom.
Conclusion
The imminent AI and robotics revolution has been dubbed the “Fourth Industrial Revolution” (Schwab, 2016), an apt moniker due to the similarities between the First & Second Industrial Revolutions and the Fourth. The First and Second Industrial Revolutions resulted in significant technological disruption and difficulties, especially for the poorest and most vulnerable, yet most modern historians believe the subsequent advancements in society and the overall standard of living far outweighed the harms (Thompson, 2017); similarly, the effects of AI come with substantial risk, but the capabilities and benefits of the technology should outweigh, at least when underpinned by a steady hand from governments, AI developers, and consumers (Schwab, 2016).
Ultimately, Dr. David Tuffley explains the potentially advantageous inevitability of AI: “all technologies embody some degree of risk … [but] from a utilitarian perspective… banning technology because it poses a manageable risk is unreasonable” (Tuffley, 2019). Therefore, rather than resisting the technology out of fear, embracing AI and focusing on developing it in a rights-respecting manner may be the best way to improve the lives of millions without trampling on human rights.
References
Alang, N., 2017. Turns out algorithms are racist. The New Republic. Available at: https://newrepublic.com/article/144644/turns-algorithms-racist
Amodei, D. & Hernandez, D., 2020. AI and compute. OpenAI. Available at: https://openai.com/blog/ai-and-compute/
Bartlett, M., 2019. The AI link between global inequality and your bank balance. Medium. Available at: https://towardsdatascience.com/the-surprisingly-simple-relationship-between-ai-global-inequality-and-your-bank-balance
Bassetti, T. et al., 2020. Artificial Intelligence – impact on total factor productivity, e-commerce & fintech. ec.europa.eu. Available at: https://publications.jrc.ec.europa.eu/repository/bitstream/JRC122268/kjna30428enn.pdf
Bernaert, A. & Akpakwu, E., 2018. Four ways AI can make healthcare more efficient and affordable. World Economic Forum. Available at: https://www.weforum.org/agenda/2018/05/four-ways-ai-is-bringing-down-the-cost-of-healthcare/
Biundo, E.B. et al., 2020. The socio-economic impact of AI in healthcare. https://www.medtecheurope.org. Available at: https://www.medtecheurope.org/wp-content/uploads/2020/10/medtecheurope_digital-health_report-on-socioeconomic-impact-of-ai-pr-27-oct-2020.pdf
Bleicher, A., 2017. Demystifying the black box that is ai. Scientific American. Available at: https://www.scientificamerican.com/article/demystifying-the-black-box-that-is-ai/
Brightman, B., 2019. Ai can help lessen dangers of climate change, industry leaders say. NBC Bay Area. Available at: http://www.nbcbayarea.com/news/national-international/ai-can-help-mitigate-climate-change-effects-industry-leaders-say/153207
Carleton, T. & Hsiang, S., 2016. Social and economic impacts of climate. Science (New York, N.Y.). Available at: https://pubmed.ncbi.nlm.nih.gov/27609899/
Chandler, S., 2020. How AI may prevent the next coronavirus outbreak. Forbes. Available at: https://www.forbes.com/sites/simonchandler/2020/03/05/how-ai-will-prevent-the-next-coronavirus-pandemic/
Cho, R. et al., 2018. Artificial intelligence-a game changer for climate change and the environment. State of the Planet. Available at: https://news.climate.columbia.edu/2018/06/05/artificial-intelligence-climate-environment/
Collier, M. et al., 2017. ARTIFICIAL INTELLIGENCE: Healthcare’s New Nervous System. www.accenture.com. Available at: https://www.accenture.com/_acnmedia/PDF-49/Accenture-Health-Artificial-Intelligence.pdf
Donti, P.L., 2019. “1 Electricity Systems.” Tackling Climate Change with Machine Learning. arXiv.org. Available at: https://arxiv.org/abs/1906.05433
Evans, R. & Gao, J., 2016. DeepMind AI reduces energy used for Cooling Google data centers by 40%. The Keyword. Available at: https://blog.google/outreach-initiatives/environment/deepmind-ai-reduces-energy-used-for/
Feldstein, S., 2019. The road to digital unfreedom: How artificial intelligence is reshaping repression. Journal of Democracy. Available at: https://www.journalofdemocracy.org/articles/the-road-to-digital-unfreedom-how-artificial-intelligence-is-reshaping-repression/
Fenech, M. et al., 2018. ETHICAL, SOCIAL, AND POLITICAL CHALLENGES OF ARTIFICIAL INTELLIGENCE IN HEALTH. wellcome.org. Available at: https://wellcome.org/sites/default/files/ai-in-health-ethical-social-political-challenges.pdf
Franke, U., 2019. Harnessing artificial intelligence. ECFR. Available at: https://ecfr.eu/publication/harnessing_artificial_intelligence/
Frey, C.B. & Osborne, M., 2013. The future of employment: How susceptible are jobs to computerisation? Oxford Martin School. Available at: https://www.oxfordmartin.ox.ac.uk/publications/the-future-of-employment/
Gerke, S., et al., 2020. Ethical and legal challenges of artificial intelligence-driven healthcare A. Bohr & K. Memarzadeh, eds. Artificial Intelligence in Healthcare. Available at: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7332220/
Goertzel, B., 2014. Artificial general intelligence: Concept, state of the art, and future prospects. Journal of Artificial General Intelligence. Available at: https://www.sciendo.com/article/10.2478/jagi-2014-0001
Goh, J.H.L. et al., 2020. Artificial intelligence for Cataract Detection and… : The ASIA-PACIFIC Journal of Ophthalmology. LWW. Available at: https://journals.lww.com/apjoo/Fulltext/2020/04000/Artificial_Intelligence_for_Cataract_Detection_and.6.aspx
Goldin, I., 2019. Will AI kill developing world growth? BBC News. Available at: https://www.bbc.com/news/business-47852589
Greaves, M., 2020. Brain-inspired electronic system could vastly reduce ai’s carbon footprint. UCL News. Available at: https://www.ucl.ac.uk/news/2020/aug/brain-inspired-electronic-system-could-vastly-reduce-ais-carbon-footprint
Guiding Principles on Business and Human Rights. 2011. Human Rights Council Res. 17/4, Rep. of the Hum. Rts. Council, 17th Sess., June 16, 2011, U.N. Doc. A/HRC/RES/17/4 (July 6, 2011); Special Rep. of the Sec’y Gen., Guiding Principles on Business and Human Rights: Implementing the United Nations “Protect, Respect and Remedy” Framework, Hum. Rts. Council, U.N. Doc. A/HRC/17/31. Available at: https://www.ohchr.org/documents/publications/guidingprinciplesbusinesshr_en.pdf
Gunasekeran, D.V. & Wong, T.Y., 2020. Artificial Intelligence in Ophthalmology in 2020: A Technology on the Cusp for Translation and Implementation. LWW. Available at: https://journals.lww.com/apjoo/Fulltext/2020/04000/Artificial_Intelligence_in_Ophthalmology_in_2020_.2.aspx
Hanspal, A., 2021. Here’s why robots are actually going to increase human employment. World Economic Forum. Available at: https://www.weforum.org/agenda/2021/02/world-economic-forum-automation-create-jobs-employment-robots/
Hao, K., 2019. Training a single AI model can emit as much carbon as five cars in their lifetimes. MIT Technology Review. Available at:
Hao, K., 2021. Ai could make health care fairer-by helping us believe what patients say. MIT Technology Review. Available at:
https://www.technologyreview.com/2021/01/22/1016577/ai-fairer-healthcare-patient-outcomes/
Heaven, W.D., 2020. AI could help with the next pandemic-but not with this one. MIT Technology Review. Available at: https://www.technologyreview.com/2020/03/12/905352/ai-could-help-with-the-next-pandemicbut-not-with-this-one
Holdren, J.P. et al., 2016. Preparing for the future of AI- whitehouse.gov. Available at: https://obamawhitehouse.archives.gov/sites/default/files/whitehouse_files/microsites/ostp/NSTC/preparing_for_the_future_of_ai.pdf
Huntingford, C. et al., 2019. Machine learning and artificial intelligence to aid climate change research and preparedness. Environmental Research Letters. Available at: https://iopscience.iop.org/article/10.1088/1748-9326/ab4e55
International Covenant on Civil and Political Rights. 1966. 999 U.N.T.S. 171 and 1057 U.N.T.S. 407, opened for signature 16 Dec. 1966, entered into force 23 Mar. 1976, art. 2. Available at: https://www.ohchr.org/en/professionalinterest/pages/ccpr.aspx
International Covenant on Economic, Social and Cultural Rights. 1966. 993 U.N.T.S. 3, opened for signature 16 Dec. 1966, entered into force 3 Jan. 1976, art. 2(1). Available at: https://www.ohchr.org/en/professionalinterest/pages/cescr.aspx
Johnson, K., 2020. How AI can empower communities and strengthen democracy. VentureBeat. Available at: https://venturebeat.com/2020/07/04/how-ai-can-empower-communities-and-strengthen-democracy/
Joshi, A. & Mishra, G., 2010. Artificial intelligence. Artificial intelligence | Proceedings of the International Conference and Workshop on Emerging Trends in Technology. Available at: https://dl.acm.org/doi/abs/10.1145/1741906.1742236
Kaack, L.H. & Milojevic-Dupont, N., 2019. “3 Buildings & Cities.” Tackling Climate Change with Machine Learning. arXiv.org. Available at: https://arxiv.org/abs/1906.05433
Kalis, B. et al., 2018. 10 promising AI applications in health care. Harvard Business Review. Available at: https://hbr.org/2018/05/10-promising-ai-applications-in-health-care
Kande, M. & Sönmez, M., 2020. Don’t fear ai. the Tech will lead to long-term job growth. World Economic Forum. Available at: https://www.weforum.org/agenda/2020/10/dont-fear-ai-it-will-lead-to-long-term-job-growth
Kelley, C.P. et al., 2015. Climate change in the Fertile Crescent and implications of the recent syrian drought. PNAS. Available at: https://www.pnas.org/content/112/11/3241
Kent, C., 2020. A race to the bottom: How AI Encodes racial discrimination within medicine – medical Technology: Issue 31: September 2020. Medical Technology | Issue 31 | September 2020. Available at:
https://medicaltechnology.nridigital.com/medical_technology_sep20/ai_racial_discrimination_medicine
Kumar, V., 2018. How artificial intelligence can increase energy efficiency. Blue and Green Tomorrow. Available at: https://blueandgreentomorrow.com/energy/artificial-intelligence-increase-energy-efficiency/
Kurzweil, R., 2006. The singularity is near: When humans transcend biology, New York: Penguin Books.
Kuts, M. & Bozhok, N., 2020. The effect of artificial intelligence on the improvement of prosthetic hands. eNUFTIR: Home. Available at: http://dspace.nuft.edu.ua/handle/123456789/31626
Lacoste, A., 2019. “5 Farms & Forests.” Tackling Climate Change with Machine Learning. arXiv.org. Available at: https://arxiv.org/abs/1906.05433
Latonero, M., 2018. Upholding human rights & dignity – data & society. Available at: https://datasociety.net/wp-content/uploads/2018/10/DataSociety_Governing_Artificial_Intelligence_Upholding_Human_Rights.pdf
Ledford, H., 2019. Millions of black people affected by racial bias in Health-care algorithms. Nature News. Available at: https://www.nature.com/articles/d41586-019-03228-6
Leopold, T.A. et al., 2018. The Future of Jobs Report. http://www3.weforum.org. Available at: http://www3.weforum.org/docs/WEF_Future_of_Jobs_2018.pdf
Maizland, L.S., 2021. China’s repression of Uyghurs in Xinjiang. Council on Foreign Relations. Available at: https://www.cfr.org/backgrounder/chinas-repression-uyghurs-xinjiang
Makala, B. & Bakovic, T.B., 2020. Artificial intelligence in the power sector. Available at: https://www.ifc.org/wps/wcm/connect/bd3a196d-a88f-45af-bbc6-e0b00790fba8/EMCompass_Note_81-05-web.pdf?MOD=AJPERES&CVID=n72pj5g
Manyika, J. et al., 2021. Jobs Lost, jobs gained: What the future of work will mean for jobs, skills, and wages. McKinsey & Company. Available at: https://www.mckinsey.com/featured-insights/future-of-work/jobs-lost-jobs-gained-what-the-future-of-work-will-mean-for-jobs-skills-and-wages
Martiniello, N. et al., 2019. Exploring the use of smartphones and tablets among people with visual impairments: Are mainstream devices replacing the use of traditional visual aids? Assistive technology: the official journal of RESNA. Available at:
https://www.ncbi.nlm.nih.gov/pubmed/31697612
Matheson, R., 2020. Reducing the carbon footprint of artificial intelligence. MIT News. Available at: https://news.mit.edu/2020/artificial-intelligence-ai-carbon-footprint-0423
McCullom, R., 2020. Artificial intelligence, health disparities, AND COVID-19. Undark Magazine. Available at: https://undark.org/2020/07/27/ai-medicine-racial-bias-covid-19/
McDonald, S.M., 2018. This may be the most Overlooked threat of climate change. The New Republic. Available at: https://newrepublic.com/article/148861/climate-change-authoritarian-leaders
Meskó, B. et al., 2018. Will artificial intelligence solve the human resource crisis in healthcare? BMC Health Services Research. Available at:
https://bmchealthservres.biomedcentral.com/articles/10.1186/s12913-018-3359-4
Nayak, S. & Das, R.K., 2020. Application of artificial Intelligence (AI) in prosthetic And Orthotic rehabilitation. IntechOpen. Available at: https://www.intechopen.com/chapters/73486.
openDemocracy, 2019. Democracy, authoritarianism and the climate emergency. openDemocracy. Available at:
Perkins, A., 2018. May to PLEDGE millions to AI Research assisting early cancer diagnosis. The Guardian. Available at: https://www.theguardian.com/technology/2018/may/20/may-to-pledge-millions-to-ai-research-assisting-early-cancer-diagnosis
Petroff, A., 2018. Artificial intelligence More ‘PROFOUND than electricity or fire. CNNMoney. Available at: https://money.cnn.com/2018/01/24/technology/sundar-pichai-google-ai-artificial-intelligence/index.html
Pielemeier, J., 2019. Ai & global governance: The advantages of applying the international human rights framework to artificial intelligence. United Nations University Centre for Policy Research. Available at: https://cpr.unu.edu/publications/articles/ai-global-governance-the-advantages-of-applying-the-international-human-rights-framework-to-artificial-intelligence.html
Pilat, D., 2019. OECD AI PRINCIPLES – THE ROLE OF MPS IN LEVERAGING THE BENEFITS OF AI. OECD Global Parliamentary Network. Available at: https://www.oecd.org/parliamentarians/meetings/gpn-meeting-october-2019/Dirk-Pilat-OECD-AI-principles-11-Oct-2019.pdf
Polonski, V., 2017. How artificial intelligence silently took over democracy. World Economic Forum. Available at: https://www.weforum.org/agenda/2017/08/artificial-intelligence-can-save-democracy-unless-it-destroys-it-first/
Powell, A., 2020. Risks and benefits of an ai revolution in medicine. Harvard Gazette. Available at: https://news.harvard.edu/gazette/story/2020/11/risks-and-benefits-of-an-ai-revolution-in-medicine/
Purdy, M. et al., 2017. Industry profits and Innovation – Accenture. Available at: https://www.accenture.com/fr-fr/_acnmedia/36dc7f76eab444cab6a7f44017cc3997.pdf
Rao, A.S. et al., 2017. Sizing the prize What’s the real value of AI for your business and how can you capitalise? Available at: https://www.pwc.com/gx/en/news-room/docs/report-pwc-ai-analysis-sizing-the-prize.pdf.
Raso, F. et al., 2018. Artificial Intelligence & Human Rights: Opportunities & Risks. Berkman Klein Center for Internet & Society Research Publication. Berkman Klein Center for Internet & Society. Available at: https://cyber.harvard.edu/sites/default/files/2018-09/2018-09_AIHumanRightsSmall.pdf
Roberts, H. et al., 2020. The Chinese approach to artificial INTELLIGENCE: An analysis of Policy, ethics, and regulation. AI & SOCIETY. Available at: https://link.springer.com/article/10.1007/s00146-020-00992-2
Romaniuk, S.N. & Burgers, T., 2018. How China’s AI technology exports are Seeding Surveillance Societies Globally. The Diplomat. Available at: https://thediplomat.com/2018/10/how-chinas-ai-technology-exports-are-seeding-surveillance-societies-globally/
Ross, A.S. & Sherwin, E.D., 2019. “6 Carbon Dioxide Removal.” Tackling Climate Change with Machine Learning. arXiv.org. Available at: https://arxiv.org/abs/1906.05433
Ross, C. et al., 2020. How software infuses racism into U.S. health care. STAT. Available at: https://www.statnews.com/2020/10/13/how-software-infuses-racism-into-us-health-care/
Saenko, K., 2020. Ai has a huge carbon footprint. here’s what we can do to reduce it. Fast Company. Available at: http://www.fastcompany.com/90587393/ai-has-a-huge-carbon-footprint-heres-what-we-can-do-to-reduce-it
Salzman, S., 2019. How Hospitals Are Using AI to Save Their Sickest Patients, Curb ‘Alarm Fatigue’. ACM. Available at: https://www.nbcnews.com/mach/science/how-hospitals-are-using-ai-save-their-sickest-patients-curb-ncna1032861
Schwab, C., 2016. The Fourth Industrial Revolution: What It Means and how to respond. World Economic Forum. Available at: https://www.weforum.org/agenda/2016/01/the-fourth-industrial-revolution-what-it-means-and-how-to-respond/
Schlogl, L. & Sumner, A., 2020. Disrupted development and the future of inequality in the age of automation, Basingstoke: Palgrave Pivot. Available at: https://link.springer.com/book/10.1007/978-3-030-30131-6
Snow, J., 2019. How people with disabilities are using AI to improve their lives. PBS. Available at: https://www.pbs.org/wgbh/nova/article/people-with-disabilities-use-ai-to-improve-their-lives/
Spratt, D. & Dunlop, I., 2019. Existential climate-related security risk: A scenario approach. Available at:
https://docs.wixstatic.com/ugd/148cb0_90dc2a2637f348edae45943a88da04d4.pdf
Stevenson, B., 2019. Artificial intelligence, income, employment, and meaning. NBER. Available at: https://www.nber.org/books-and-chapters/economics-artificial-intelligence-agenda/artificial-intelligence-income-employment-and-meaning
Strubell, E. et al., 2019. Energy and policy considerations for deep learning in nlp. arXiv.org. Available at: https://arxiv.org/abs/1906.02243
Frost & Sullivan, 2016. From $600 M to $6 BILLION, artificial intelligence SYSTEMS poised for dramatic market expansion in healthcare. Frost & Sullivan. Available at: https://www.frost.com/news/press-releases/600-m-6-billion-artificial-intelligence-systems-poised-dramatic-market-expansion-healthcare/
Sydell, E., 2020. Don’t blame AI for bad outcomes – blame incompetence. ERE. Available at: https://www.ere.net/dont-blame-ai-for-bad-outcomes-blame-incompetence/
Thompson, C., 2017. When robots take all of our jobs, remember the Luddites. Smithsonian.com. Available at: https://www.smithsonianmag.com/innovation/when-robots-take-jobs-remember-luddites-180961423/
Toews, R., 2020. Deep Learning’s carbon emissions problem. Forbes. Available at: https://www.forbes.com/sites/robtoews/2020/06/17/deep-learnings-climate-change-problem
Tom, E. et al., 2020. Protecting data privacy in the age of ai-enabled ophthalmology. Translational Vision Science & Technology. Available at: https://tvst.arvojournals.org/article.aspx?articleid=2770246#247052185
Totty, M., 2020. How to make artificial intelligence less biased. The Wall Street Journal. Available at: https://www.wsj.com/articles/how-to-make-artificial-intelligence-less-biased-11604415654
Tuffley, D., 2019. Human intelligence + artificial intelligence = human potential. Griffith Journal of Law & Human Dignity. Available at: https://research-repository.griffith.edu.au/handle/10072/385881
United Spinal, 2020. How AI is changing the lives of people with disabilities. United Spinal Association. Available at: https://unitedspinal.org/ai-changing-lives/
Universal Declaration of Human Rights. 1948. U.N.G.A. Res. 217 A (III). Available at: https://www.un.org/en/udhrbook/pdf/udhr_booklet_en_web.pdf
Waldman-Brown, A., 2019. “4 Industry.” Tackling Climate Change with Machine Learning. arXiv.org. Available at: https://arxiv.org/abs/1906.05433
Wong, T.Y. & Sabanayagam, C., 2019. Strategies to tackle the global burden of diabetic retinopathy: From epidemiology to artificial intelligence. International Journal of Ophthalmology. Available at: https://pubmed.ncbi.nlm.nih.gov/31408872
Wright, N.D., 2020. Artificial Intelligence and Democratic Norms -Meeting the Authoritarian Challenge. www.ned.org. Available at: https://www.ned.org/wp-content/uploads/2020/07/Artificial-Intelligence-Democratic-Norms-Meeting-Authoritarian-Challenge-Wright.pdf
Omer Mujawar is a final year student of Johns Creek High School, Johns Creek, Georgia, Atlanta, USA. This paper is based on his participation in an international written debate competition in the State of Georgia on the motion: “The benefits of artificial intelligence outweigh the harms”. The author is grateful to Professor Abdulrahim Vijapur of Aligarh Muslim University, Aligarh, India, for his crucial mentoring and guidance through the process of creating, writing, and publishing this paper. The author would also like to thank his parents for supporting his intellectual endeavours. He can be reached at: [email protected]
