El Salvador

Innovation in Artificial Intelligence: Illustrations in Academia, Apparel, and the Arts

Innovation in Artificial Intelligence: Illustrations in Academia, Apparel, and the Arts

USAR LOS ARTICULOS DE ABAJO PARA PONER MAS POSTS DE ESTE, MEJOR SI NO ES DE KAPLAN

Summary

Artificial intelligence (AI), commonly defined as “a system’s ability to correctly interpret external data, to learn from such data, and to use those learnings to achieve specific goals and tasks through flexible adaptation,” can be classified into analytical, human-inspired, and humanized AI depending upon its application of cognitive, emotional, and social intelligence. AI’s foundations took place in the 1950s. A sequence of vicissitudes of funding, interest in, and support for AI followed subsequently. In 2015 AlphaGo, Google’s AI-driven system, won against the human grandmaster in the highly complex board game Go. This is considered one of the most significant milestones in the development of AI and marks the starting of a new period, enabling several AI innovations in a variety of sectors and industries.

Higher education, the fashion industry, and the arts serve as illustrations of areas wherein ample innovation based on AI occurs. Using these domains, various angles of innovation in AI can be presented and decrypted. AI innovation in higher education, for example, indicates that at some point, AI-powered robots might take over the role of human teachers. For the moment, however, AI in academia is solely used to support human beings, not to replace them. The apparel industry, specifically fast fashion—one of the planet’s biggest polluters—shows how innovation in AI can help the sector move toward sustainability and eco-responsibility through, among other ways, improved forecasting, increased customer satisfaction, and more efficient supply chain management. An analysis of AI-driven novelty in the arts, notably in museums, shows that developing highly innovative, AI-based solutions might be a necessity for the survival of a strongly declining cultural sector.

These examples all show the role AI already plays in these sectors and its likely importance in their respective futures. While AI applications imply many improvements for academia, the apparel industry, and the arts, it should come as no surprise that it also has several drawbacks. Enforcing laws and regulations concerning AI is critical in order to avoid its adverse effects. Ethics and the ethical behavior of managers and leaders in various sectors and industries is likewise crucial. Education will play an additional significant role in helping AI positively influence economies and societies worldwide. Finally, international entente (i.e., the cooperation of the world’s biggest economies and nations) must take place to ensure AI’s benefit to humanity and civilization. Therefore, these challenges and areas (i.e., enforcement, ethics, education, and entente) can be summarized as the four summons of AI.

Keywords

Subjects

  • Business Education
  • Entrepreneurship
  • Ethics
  • Information Systems
  • Technology and Innovation Management

Introduction: AI’s Historical Sketch

The birth of artificial intelligence (AI) can be pinpointed to a series of apocryphal and actual events, beginning with the publication of Isaac Asimov’s short story “Runaround” (1950). In it, he described the well-known Three Laws of Robotics, which still inspire AI experts and specialists worldwide:

1.

A robot may not injure a human being or, through inaction, allow a human being to come to harm.

2.

A robot must obey the orders given it by human beings except where such orders would conflict with the First Law.

3.

A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

In the same year, Alan Turing (1950) wrote his now-famous article, a scientific paper titled “Computing Machinery and Intelligence,” which sets forth a way of examining a machine’s (artificial) intelligence, the so-called Turing test. Essentially, the Turing test has a human assess if they chat with a machine or another human being. The human evaluator would know that one of the two chat partners was an AI system. If the assessor were incapable of distinguishing the machine from the human, the machine would be considered to have passed the Turing test; in other words, it was able to display intelligent behavior equivalent to that of a human being.

In 1956, John McCarthy and Marvin Minsky, together with others, organized the Dartmouth Summer Research Project on Artificial Intelligence, at which the term artificial intelligence was coined. This workshop, considered by many to be the founding event of AI as a field, brought together leading scientists and experts. The event’s initial proposal is often credited with having introduced the term “artificial intelligence”:

“We propose that a 2-month, 10-man study of artificial intelligence be carried out during the summer of 1956 at Dartmouth College in Hanover, New Hampshire. The study is to proceed on the basis of the conjecture that every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it. An attempt will be made to find how to make machines use language, form abstractions and concepts, solve kinds of problems now reserved for humans, and improve themselves. We think that a significant advance can be made in one or more of these problems if a carefully selected group of scientists work on it together for a summer.”

 

What followed was a period of high investment and strong enthusiasm for AI progress which spanned nearly 20 years. Early examples of success stories were the General Problem Solver program created in 1959, able to automatically solve simple problems such as the Towers of Hanoi; or the Eliza program, finalized in 1966, a natural language processing tool able to simulate a conversation with a human and said to have passed the Turing test. The strong optimism that prevailed therein can best be illustrated by a 1970 interview by Marvin Minsky in Life Magazine in which he stated that AI could attain average human intelligence in at most 8 years.

Yet, with time, initial optimism turned into pessimism due to an obstacle with regard to AI’s feasibility: computational power was just not big enough, which represented a substantial barrier to AI’s evolution. This resulted in a period, also called AI winter, characterized by scarce funding and vehement skepticism toward AI development.

While funding briefly returned in the 1980s, overall financing stayed relatively low, although some highly publicized breakthroughs in AI evolution took place in the 1990s and 2000s. One of these events occurred in 1997, when IBM’s Deep Blue, a chess-playing computer program, defeated Russian world chess champion Gary Kasparov. This was the first time a machine won a match against a reigning world champion and therefore was intensively mediatized across the world.

The real milestone and game changer, however, did not happen until 2015, when Google’s AI-driven system, AlphaGo, beat the human grandmaster in the highly complex abstract strategy board game Go (also known as Weiqi or Weichi). Until that point, computational power was too weak to solve the complex algorithms such as those underlying Go, rendering this match so particular and groundbreaking: What previously was impossible became reality. This match marked the beginning of a new era of AI with a plethora of current and future innovations enabled across all sectors and industries, finally leading to returns on investment of years-long research and development.

In this article, three sectors—higher education, the (fast) fashion industry, and the cultural sector as applied in museums—will be described and analyzed vis-à-vis AI innovation and progress. Prior to this, because AI is a quite vague concept, it will be explained, categorized, and delimited from other similar concepts, such as machine learning, the Internet-of-things, big data, and expert systems. A final section discusses four important notions, the four “Es” for society and human civilization to be properly prepared for upcoming AI-based innovation: enforcement of laws and regulations, ethical conduct, education, and international cooperation, or in diplomatic jargon, “entente.”

AI: Definition and Delimitation

Artificial intelligence (AI) pioneer Marvin Minsky (1968) defined AI as “the science of making machines do things that would require intelligence if done by men” (p. v). John McCarthy (2007) later expanded upon Minsky’s definition, describing AI as “the science and engineering of making intelligent machines, especially intelligent computer programs” (p. 2). For Dartmouth professor Praveen Kopalle et al. (2022), AI characterizes “programs, algorithms, systems, and machines that mimic intelligent human behavior” (p. 522). A more detailed definition of AI describes it as a system’s ability to correctly interpret external data, to learn from such data, and to use those learnings to achieve specific goals and tasks through flexible adaptation.

Such diversity in defining AI only confirms the fuzziness of the concept and the consequent difficulty in coming up with a clear, straightforward definition thereof . There are at least two explanations therefor: Firstly, it is already quite complicated to define (human) intelligence, which is often described in various ways, such as the ability for creativity, critical thinking, learning, planning, problem-solving, reasoning, and understanding. Secondly, AI is a moving target, termed “the AI effect”, which occurs when onlookers discount the behavior of an AI program by claiming that it does not reflect actual intelligence.It’s part of the history of the field of artificial intelligence that every time somebody figured out how to make a computer do something—play good checkers, solve simple but relatively informal problems—there was a chorus of critics to say, “that’s not thinking.”

A further reason rendering a definition for AI elusive is that there are various types of AI building upon each other and reflecting various combinations of cognitive, emotional, and social competencies, which are analytical, human-inspired, and humanized AI . Analytical AI exhibits traits consistent with strictly cognitive competencies by producing a cognitive representation of a phenomenon based on past learning to inform future decision-making. Both cognitive and emotional competencies are necessary to be considered human-inspired AI in which such AI-driven systems understand and detect human emotions in addition to cognitive elements. Finally, humanized AI is characterized by a combination of cognitive, emotional, and social competencies, enabling such systems to be self-aware and self-conscious when interacting with humans.

 

To illustrate the differences between the three types, it is instructive to look at the apparel industry. Amazon, for example, the largest online apparel seller worldwide, applies analytical AI in its Amazon Go Store, its recommendation engine, and its voice assistant, Alexa. Based on a shopper’s previous purchases, past purchasing behavior, and browsing history, a Go Store shopper could be provided with AI-driven, personalized fashion tips from Alexa or suggestions for further purchases by Amazon’s recommendation engine . Within the scope of human-inspired AI, these cognitive elements would be supplemented with capabilities of emotional intelligence. Walmart, known for its affordable apparel, has developed facial recognition software able to identify their in-store shoppers’ emotions. Able to detect shoppers’ emotional states with the help of in-store cameras, this system could, for example, decide to send a sales associate to assist a shopper in the clothing department identified as being highly dissatisfied. Finally, in the currently still-fictional domain of humanized AI, such an employee could, instead of being human, be an AI-powered sales robot. Capable of self-awareness and socially intelligent, such a robotic salesforce would, for example, when advising a couple shopping, sense whenever they might need a private moment to discuss their purchase, or conversely, the robot’s presence might be welcome if it might release some tension between the couple disagreeing on a purchase .

Most AI systems belong to the category of analytical AI applied to a multitude of sectors such as banking (e.g., credit risk assessments), human resources (HR) (e.g., recruitment decisions), medicine (e.g., surgery allocation), and public administration (e.g., immigration status decisions) increasingly relying on AI-based algorithms for decision-making and judgement. More and more human decisions are replaced or augmented by AI, which demands that the human workforce acquire new capabilities and skills in order to remain relevant in the job market . With an increase in computational power and technological advances, AI progresses from solving rather narrow and simple tasks to more complex and broader ones in such multifaceted areas as the arts, creativity, and innovation. AI has been said to be at the core of the so-called “fourth industrial revolution” , characterized by the transfer of control and influence from humans to technology and thus redefining human–technology entanglements.

Further confusing the issue of defining AI is that terminology such as big data, the Internet of things, machine learning, and expert systems is often applied incorrectly, albeit closely connected to AI. To be able to learn from data, AI needs lots of it (i.e., big data), defined as datasets made up by huge quantities (volume) of frequently updated data (velocity) in various formats, such as numeric, textual, or images/videos (variety). The Internet of things provides one way of obtaining big data, and is described as an extension of Internet connectivity into physical devices and everyday objects such as a large home appliances, which are equipped with sensors and software to collect and exchange data. Machine learning, in contrast, is defined as methods that help computers learn without being explicitly programmed to do so, thus enabling the detection of underlying patterns within (big) data.

Likely the most difficult delimitation of AI is with respect to expert systems, which are collections of rules programmed by humans in the form of ‘if > then’ statements and theorize human intelligence in a top-down manner; in other words, they assume that human brain activity can be systematized as per predefined standards and rules. Expert systems are the clear predecessors of AI systems but which do not represent actual AI, at least not according to the defining trait of being able to learn autonomously. In fact for a long time, existing AI was categorized with expert systems, (i.e., it should not have been called AI stricto sensu). Conversely, genuine AI follows a bottom-up approach and reconstructs human brain activity by applying big data to derive learnings autonomously. Probably the best known expert system is Deep Blue, IBM’s chess supercomputer, which defeated world champion Garry Kasparov in 1997. This victory of a machine over a human has been the subject of several books and films and was considered a landmark in the history of AI. Strictly speaking, Deep Blue represented an expert system, not capable of autonomous learning as is AI.

Implementations in Academia: AI and Higher Education

Much of early artificial intelligence (AI) progress had its origins in academia as mentioned in the section “Introduction: AI’s Historical Sketch.” The term artificial intelligence was defined and coined at Dartmouth College in 1956. DeepMind, the human Go champion-defeating AI system, was also developed by three scientists, two of them students at University College London. True to its beginnings, AI has definitely begun affecting higher education. Concerning research, AI is applied to writing academic literature reviews, conducting thorough plagiarism checks, or detecting fake data and the misuse of data analysis and statistics. In teaching, AI-powered chatbots are already used at Georgia Tech, Carnegie Mellon, and the Technical University of Berlin to answer students’ queries, thus freeing up professors’ time for more value-added activities. By now, AI is also increasingly used for grading, admissions, and even instruction—applications which have become controversial. Thinking ahead, one might ask if AI-driven robots might replace human professors altogether, prompting to ask whether universities have not sowed the seeds of their own disruption with their innovation in AI.

Jill Watson, likely the best known AI-based course assistant, is Georgia Tech’s chatbot, having begun her career in 2016. Two situations motivated Jill’s creation: Firstly, the team of human teaching assistants (TAs) was no longer capable of responding to 10,000 emails and queries per semester. Secondly, it was hoped that Jill would augment student engagement, which seemed to suffer due to increasingly high numbers of students enrolled in the course. The bot easily responds to all queries that have clear, straightforward answers. Examples are office hours, when the next class session takes place, or the reading list. Whenever questions are too demanding or complex, the human TAs take over, a rare occurrence as Jill is constantly improving her capacities due to her autonomous learning ability. From the outset, results were so persuasive, with Jill accurately responding to 97% of student queries, that some students could not detect when they were corresponding with a machine rather than a human.

Besides facilitating administrative processes and information seeking, AI has successfully been applied in instruction in higher education. Using so-called learning analytics, which are described as the measurement, collection, and analysis of (big) data about learners and their contexts, aiming at understanding and optimizing learning and teaching, faculty can more easily monitor students’ progress and evaluate their own pedagogical approach, course content, teaching resources, and so forth. Big data in higher education furthermore enables adaptive learning (i.e., the application of AI-based pedagogy to adapt the level and speed of instruction to the individual learner). Adaptive learning is not only consistent with student-centricity, but has been shown to be superior to nonadaptive learning. Moreover, students themselves may use such analytics for self-evaluation purposes. While highly useful for teachers and students alike, such analytics also prompt reflection about students’ data protection and right to privacy.

Grading is also increasingly done using AI, to the delight of many professors. Machines have been grading multiple-choice exams since the 1950s. Today, AI-driven machines can evaluate student essays and exam papers. AI can imitate an instructor’s grading style within less than one percentage point disparity. With more than 140,000 online students, Southern New Hampshire University (SNHU) extensively works on ways to use AI to grade such huge student numbers as quickly as possible. SNHU even tests AI-powered machines that monitor and analyze a student’s voice, body language, and response time within video communications to see how such indicators might be integrated into student evaluations.

Finally, many universities and higher education institutions also apply AI for admission and financial aid decisions. Indicators integrated into AI-driven data analysis are taken from works, a candidate’s way of interacting with the university’s website or email exchange, assessing their likelihood of enrollment and successfully graduating from the respective program, and the likely level of financial aid obtained. Kira Talent, a Canadian EdTech company, has developed an AI system that is able to rate a candidate’s soft skills, perceived motivation, and personality traits by simply reviewing video content submitted by the applicant. While AI helps to narrow the pool of candidates, humans—for now—are still making final admissions decisions. Despite this, some institutions, such as the University of Texas at Austin, have decided to stop using AI in admissions after detecting that such a system potentially strongly reinforces and perpetuates bias in acceptance decisions.

While AI is applied in higher education institutions around the globe, China is particularly advanced therein, demonstrating clearly how the sector could evolve and what innovations can be expected. Supported by the Chinese Education Ministry’s Smart Campuses campaign, Xian University, for example, has deployed facial recognition cameras and AI-powered gates across campus. Students must scan their faces to enter and to prove class attendance. Furthermore, AI technology supported by hundreds of cameras deployed across campus can detect seven different emotions: surprise, sadness, neutral, happiness, fear, disappointment, and anger. These emotional states are used to analyze students’ attentiveness, among other traits. Such an approach raises concerning questions about data privacy and the protection of personal information.

Presently, while AI mostly frees up faculty from strictly administrative tasks, in the future, human teachers might be replaced by AI-driven robotics, bringing one back to the question of whether universities indeed have sowed the seeds of their own disruption with their use of AI. The Chinese government makes no secret of the fact that its extensive development and use of AI has the clear objective of surmounting its deficit in competent teaching staff, possibly complementing them. AI-driven machine instructors most likely will, at some point, be less costly than human professors, making the former an appealing option. Only time will tell what AI developments will occur in higher education. Ultimately, the test will be if students prefer AI-powered robots to human instructors, the answer to which might be more complicated than might be expected (or hope for). To provide food for thought, a recent study by the Center for the Governance of Change at Spain’s IE University demonstrated that one-fourth of Europeans would prefer that politics be conducted by AI-driven systems instead of human politicians, who are often extreme ideologically or tempted by corruption.

Applications in Apparel: AI and Fast Fashion

Another arena of innovation in artificial intelligence (AI) is fast fashion, which is one of the world’s biggest polluters, emitting 10% of global carbon emissions and 20% of global wastewater from textile dyeing . Aware of an increasing number of eco-responsible consumers, several brands are seeking to become more sustainable in their approach and operations, at least as much as possible, without completely leaving their current successful business model behind. AI and its innovations are one of the sector’s avenues toward sustainability.

The first group of innovations aims at increasing customer satisfaction. AI-driven assistants or recommendation systems make personalized styling suggestions to clients. A customer’s body measurements and shape, personal taste, and client history serve as the basis for such recommendations. Similarly, smart mirrors enable clients to try on clothes virtually, quickly compare several clothing options, alternate styles, and colors, and combine various fashion items. Again, mirrors can issue recommendations for which color or style works best with a customer’s complexion or body shape. More suitable and better fitting clothes result in higher customer satisfaction with purchased apparel, and consequently fewer returns, in turn significantly decreasing the industry’s carbon footprint. Currently, the sector prefers to destroy such items rather than trying to resell them, the latter unfortunately costing more than the former. Such an AI-driven approach would even be more beneficial in countries like China, where shipping is cheap and highly efficient, leading Chinese consumers to place online orders without much reflection, as items can easily be returned.

Moreover, again to increase customer satisfaction, AI can help to increase a product’s emotional value to its customers, providing them with the possibility of purchasing tailor-made, customized, and personalized apparel. Via advanced three-dimensional modeling technologies, a customer’s measurements are taken in dedicated in-store fitting rooms or via a mobile device in the case of online shopping. Digital models, which serve as the template for actual production, are altered accordingly. In a final step, the tailored item is directly produced and either handed to the customer in-store or shipped to their chosen address. Due to their higher emotional value, customers usually keep tailor-made clothes longer and wear them more.

The second group of AI innovations in fast fashion addresses improving supply chain efficiency due to advanced data analytics, highly optimized sales forecasting, the prediction of fashion trends and consumer behavior. More efficient supply chains equal savings and higher eco-responsibility and sustainability. Suboptimal supply chains are the cause of overstocking, overproduction, wasteful energy consumption, and huge quantities of waste. At least in part, AI can overcome these practices and lower the industry’s carbon footprint. In this respect, AI can even help find and source more sustainable textiles. Comparing materials and their properties and analyzing suppliers’ locations and production methods, AI can find the environmentally optimal manufacturer who produces textiles adapted to a specific design pattern, resulting in the least polluting supply chain.

In the context of rendering the supply chain more efficient, one needs to briefly mention robotics and automation. According to several reports analyzing the (fast) fashion industry and its future, robots most likely will replace a significant share of the human workforce in production and transportation, sorting and packing in warehouses, and marketing and sales. On the positive side, increased automation results in significant cost reductions and permits reshoring of manufacturing, potentially abolishing the sector’s much-maligned labor conditions. On the negative side, (AI-powered) automation also results in unemployment. While improved supply chains increase environmental sustainability, societies worldwide might face the challenge of increased of layoffs and consequent social instability.

A third and final group of AI innovations in fast fashion represents opportunities that would lead to a more profound change—altering the industry’s entire business model. This would mean incentivizing longer usage of fewer fashion items at a most likely higher price. To incentivize more eco-responsible purchasing behavior, AI could foster more transparency regarding where apparel or textiles are sourced and under what labor conditions. Potential buyers could then choose more sustainable options. AI could even determine an item’s optimal price by calculating material cost, energy consumption, and finally, the cost of a product’s carbon emissions. Extending priority to second-hand markets would also fall into the category of AI innovations. AI-powered websites and mobile apps could be a powerful way of simplifying the purchase and selling of second-hand garments, recommending products based on personal data, suggesting trustworthy and adapted prices, and so forth.

Environmentally speaking, while AI creates many possibilities for it to become more sustainable, such as encouraging more eco-responsible consumption and improved supply chain and inventory management, fast fashion might need to rethink its business model altogether to tap into AI’s potential fully. Increasing a product’s emotional value and incentivizing second-hand purchases or more transparency run counter to fast fashion’s current policy of promoting overconsumption and short product life cycles. Whether apparel brands will go in this direction (the circular economy approach) or stick to their old habits represents an ethical turning point for the sector.

Illustrations in the Arts: AI and Modern Museums

Whereas artificial intelligence (AI) innovation in academia might lead to the sector’s disruption, AI-driven innovation in the arts sector might actually assure its survival. Museums are under threat in an era characterized by digital transformation, social media, and the (future) metaverse. In particular, younger visitors are increasingly absent from museums, which no longer represent primary attractions to them. Museums will need to be innovative to stay appealing to survive over time. As an illustration of innovation in AI in museums, New York City’s Metropolitan Museum of Art, more familiarly called “the Met,” will serve as an example.

With its permanent collection of more than two million works of art, the Met is the largest U.S. art museum. The Met’s collection includes pieces from ancient Egypt, classical antiquity, art by all of the major European artists, and a significant share of American artwork. Known for its innovative approach, especially concerning technology, the Met has implemented an open-access policy that enables the sharing and posting—free of any charges and possible copyright infringements—of hundreds of thousands of high-quality images of its artwork across all social media. In addition, the application of AI is increasingly present in the museum. Due to the multitude of current and planned innovations based on AI, the Met could easily earn the title of the Metropolitan Museum of ARTificial Intelligence.

As in many sectors, the COVID-19 pandemic accelerated AI’s applications in the arts, with most exhibits closed for several months due to lockdowns. Many affected institutions used digital means to bring the cultural experience directly to art consumers. The Met responded by partnering with telecommunications provider Verizon to deploy highly innovative augmented and virtual reality technologies, which enabled art lovers to escape lockdown boredom by digitally visiting the Met’s collection with a surprising degree of realism.

The Met’s “Art Explorer,” developed in collaboration with Microsoft, is a further use of its AI innovations. Essentially “Art Explorer” is a powerful AI-driven search engine that is part of the Met’s virtual tour. The system suggests art pieces to visitors that might be of interest. For example, if a visitor liked Van Gogh’s Olive Trees, then the system will search similar artwork and might suggest looking at Van Gogh’s Cypresses. It might also suggest paintings that do not have much in common, at least not at first glance. While such subtler similarities might have remained undetected ordinarily—even by experts— simply due to the sheer quantity of pieces inside (and outside) the Met’s collection, AI can detect connections, commonalities, and similarities. Moreover, AI can even detect visitors’ emotions when looking at certain works, categorizing them into various emotional groups. Visitors then could choose to be scared, amused, or disgusted—if they so wish—during both their virtual and actual visit put together by the AI system specifically for them.

Another collaboration of the Met and Microsoft, joined by the Massachusetts Institute of Technology (MIT) is the Met x Microsoft x MIT initiative, which represents another example of AI in art. According to the Microsoft vice president in charge of AI, Mitra Azizirad, The close partnership between the Met, MIT, and Microsoft is a great example of how AI is empowering curators and technologists to make art and human history accessible and relevant to everyone on the planet. This partnership objective is twofold: first, to enable the Met’s visitors to post, share, and communicate about their visit in innovative ways, mostly involving their social media; and second, to increase their engagement with the artwork and the museum generally (Kessler, 2019). Three of the initiative’s projects—“Storyteller,” “Tag, That’s It!,” and “Artwork of the Day”—are briefly presented:

“Storyteller” entails an AI-driven voice recognition system that listens to what visitors talk about while walking through the Met. Capable of listening to conversations in more than 64 languages, the system consequently proposes artwork related to the visitors’ discussions and exchanges that they have with each other. “Storyteller” puts together an entire museum tour based on whatever visitors’ conversations are prompted by the exhibits they are seeing. Additionally, visitors are provided with a complete art catalog, preserving for posterity their highly engaging, personalized, and customized tour of the Met.

“Tag, That’s It!” is another Met x Microsoft x MIT initiative projects, which could be described as “museum visitors becoming museum staff.” Through a gamification approach, visitors are asked to tag pictures that they take during their Met visit as precisely as possible. This tagging and describing of paintings and other works are aided via a dedicated app and, step by step, will transform the Met’s collection into a huge database containing highly valuable information, adding tremendous benefit to what is known about art in general and the Met’s collection in particular. Museum visitors do the work—joyfully so—of art curators, historians, and other personnel.

With the help of “Artwork of the Day,” each visitor receives a personalized suggestion of their visit’s artwork. Suggestions are issued by an AI-based system considering, for example, current news and events of interest, the weather and time of day, a visitor’s past behavior (if existent), and current location. Based on this data, the system chooses a piece of art tailored to the museum visitor, for example, one of particular interest and appeal to that individual or group visiting the Met. For the moment, in an effort to provide an ultra-personal recommendation as well as to manage the flow of visitors and prevent crowding, “Artwork of the Day” is programed such that no two visitors receive the same suggestion on a given day. Obviously, as an autonomously learning AI system, these application parameters definitely could change in the future.

These are just a few examples of how AI can make museums more exciting, be perhaps more relevant to digital natives, and hopefully secure sufficient visitor numbers in the future. Besides increasing visitor engagement, these AI innovations can also render museum operations more efficient and less costly through better energy management, stronger security systems, and more granular forecasting of visitor numbers and behavior. In addition to increasing visitor numbers and keeping them constant, applying AI to museum operations could significantly reduce costs, thereby being a supplementary lever to survival.

Afterword: AI’s Four Summons

Several articles have been published about ethical concerns with regard to the application of artificial intelligence. Regulation is needed, as called for by several scientists in the field. However, such ethical and regulatory frameworks are scarce, if not entirely absent (Scherer, 2016). A variety of solutions have been suggested, especially with regard to accountability, for example, have opted for a self-governance approach while others have called for laws to be imposed by governments and supranational regulators.

Looking at AI innovations across various sectors, it becomes clear that its application demands some reflection and the overcoming of challenges. The use of AI at the Metropolitan Museum of Art raises several questions about privacy and personal data. AI listening to a tour group’s conversations during a museum visit, for example, might result in a highly interesting walk-through; however, suppose visitors forget “Big Brother” is watching/listening? Suppose they regret things they said during the tour? Who will be accountable if some of this information falls into the wrong hands?

To reflect upon such questions, the apparel company H&M, addressing similar concerns in the fast fashion sector, established the Ethical AI Debate Club, where potentially nefarious AI innovations under consideration by the Swedish fashion brand are thoroughly discussed before implementation. H&M evaluates each new AI project in terms of potential ethical and societal risks via a nine-principle index: AI applied therein must be beneficial, collaborative, fair, focused, governed, reliable, secure, transparent, and respectful of human agency.

While such initiatives by individual organizations are more than welcome, a more structured approach to AI challenges is necessary, and accordingly is the subject of these concluding remarks. Summarized under the heading “AI’s four summons,” these appeals can be structured along four Es: enforcement, ethics, education, and entente (see Figure 2). The first summons addresses the enforcement of state regulation and legislation with respect to AI. Therein, several questions will need to be addressed, especially potentially increased unemployment rates due to automation. This task is anything but easy, as regulation often constitutes a trade-off and a barrier to creativity, innovation, and invention.

While regulation and laws are essential, it is not easy to control and oversee everything that AI encompasses, as it is constantly changing and evolving rapidly. Therefore, the second summons addresses ethics, or more precisely, the ethical behavior of leaders, managers, and decision-makers in a given society. It has been established that AI can be used either for bad or for good, to prevent or inflict harm. Ultimately, AI does whatever activity it has been programed to do. Those with weak ethics can take advantage of this by explicitly attempting harmful and/or illegal activities.

Via education, the third summons—ethical behavior, specifically regarding AI’s progress—must be taught from an early stage onward. Therein, individuals can be educated to become more aware of AI’s potential misuse or possible manipulative action. Moreover, end users/consumers must be better educated and informed about privacy questions and data security, as well as AI and big data’s opportunities and challenges. Accordingly, AI and associated ethical conduct should be part of both dedicated course syllabi and entire program curricula in primary, secondary, and higher education worldwide.

Finally, international cooperation being crucial, entente is the fourth summons. In some of the examples discussed in this article, some countries are less restrictive concerning data privacy and protection than are others. This regulatory difference can represent a competitive advantage for corporations in the less restrictive locales concerning the development of AI. As a reaction thereto, companies in AI-restrictive locales might be tempted to offshore their business. Consequently, employees and workers might suffer on the one side, while consumer protection and privacy suffer on the other side. Therefore, the call for diplomacy and cooperation is imperative.

To summarize, this article provided a general picture of AI, its history, what it is and what it is not, and its innovations in three specific sectors: academia, apparel, and the arts. Various industries highlight differing angles to innovations in AI. In the case of museums, AI might ensure their survival. With respect to fast fashion, AI might lead to more sustainability. In academia, AI might ultimately result in higher education’s disruption. AI obviously has many positive and several negative aspects. Or, to put it into physicist Stephen Hawking’s words, AI will be either the best, or the worst thing, ever to happen to humanity.

Further Reading

  • Haenlein, M., & Kaplan, A. (2019). A brief history of AI: On the past, present, and future of artificial intelligence. California Management Review, 61(4), 5–14.
  • Libai, B., Bart, Y., Gensler, S., Hofacker, C., Kaplan, A., Köttenheinrich, K., & Kroll, E. (2020). A brave new world? On AI and the management of customer relationships. Journal of Interactive Marketing, 51(C), 44–56.

References

  • Alsadoon, E. (2020). The impact of an adaptive e-course on students’ achievements based on the students’ prior knowledge. Education and Information Technologies, 25(5), 3541–3551.
  • Amabile, T. M. (2020). Creativity, artificial intelligence, and a world of surprises. Academy of Management Discoveries, 6(3), 351–354.
  • Asimov, I. (1950). Runaround. In I. Robot (Ed.), TheIsaac Asimov collection (pp. 40–58). Doubleday.
  • Baer, L., & Carmean, C. (2019). An analytics handbook: Moving from evidence to impact. Society for College and University Planning.
  • Brock, J. K. U., & Von Wangenheim, F. (2019). Demystifying AI: What digital transformation leaders can teach you about realistic artificial intelligence. California Management Review, 61(4), 110–134.
  • Byers, G. (2020, February 28). Artificial intelligence is restyling the fashion industry. Towards Data Science.
  • Camargo, L. R., Pereira, S. C. F., & Scarpin, M. R. S. (2020). Fast and ultra-fast fashion supply chain management: An exploratory research. International Journal of Retail & Distribution Management, 48(6), 537–553.
  • Carbonell, J. R. (1970). AI in CAI: An artificial intelligence approach to computer-aided instruction. IEEE Transactions on Man-Machine Systems, 11(4), 190–202.
  • Cates, M. (2019, February 4). The Met, Microsoft, and MIT explore the impact of artificial intelligence on how global audiences connect with art. The Met.
  • Chen, A. (2021, December 29). 5 Ways AI is transforming the fashion industry for sustainability. Towards Data Science.
  • Crawford, K., & Calo, R. (2016). There is a blind spot in AI research. Nature, 538, 311–313.
  • Darrach, B. (1970, November 20). Meet Shaky, the first electronic person. Life Magazine.
  • Del Prete, M. (2022). Mindful sustainable consumption and sustainability chatbots in fast fashion retailing during and after the COVID-19 pandemic. Journal of Management and Sustainability, 12(1), 19–36.
  • Ellen MacArthur Foundation. (2017). A new textiles economy.
  • Fletcher, K., & Tham, M. (2015). Routledge handbook of sustainability and fashion. Routledge.
  • Glikson, E., & Woolley, A. W. (2020). Human trust in artificial intelligence: Review of empirical research. Academy of Management Annals, 14(2), 627–660.
  • Haenlein, M., Huang, M.-H., & Kaplan, A. (2022). Business ethics in the era of artificial intelligence. Journal of Business Ethics, 178, 867–869.
  • Haenlein, M., & Kaplan, A. (2021). Artificial intelligence and robotics: Shaking up the business world and society at large. Journal of Business Research, 124, 405–407.
  • Haenlein, M., Kaplan, A., Tan, C.-W., & Zhang, P. (2019). Artificial intelligence (AI) and management analytics. Journal of Management Analytics, 6(4), 341–343.
  • Hern, A. (2016, October 19). Stephen Hawking: AI will be “either best or worst thing” for humanity. The Guardian.
  • John-Mathews, J.-M., Cardon, D., & Balagué, C. H. (2022). From reality to world—A critical perspective on AI fairness. Journal of Business Ethics, 178(4), 945–959.
  • Kaplan, A. (2009). Virtual worlds and business schools: The case of INSEAD. In C. Wankel & J. Kingsley (Eds.), Higher education in virtual worlds: Teaching and learning in second life (pp. 83–100). Emerald Group.
  • Kaplan, A. (2018). A school is a building that has 4 walls—With tomorrow inside: Toward the reinvention of the business school. Business Horizons, 61(4), 599–608.
  • Kaplan, A. (2020a). Artificial intelligence, marketing, and the Fourth Industrial Revolution: Clarifications, challenges, concerns. In B. Christiansen & T. Škrinjarić (Eds.), Handbook of research on applied AI for international business and marketing applications (pp. 1–13). IGI Global.
  • Kaplan, A. (2020b). Artificial intelligence, social media, and fake news: Is this the end of democracy? In A. A. Gül, Y. D. Ertürk, & P. Elmer (Eds.), Digital transformation in media and society (pp. 149–161). Istanbul University Press Books.
  • Kaplan, A. (2020c). Artificial intelligence: Emphasis on ethics & education. InternationalJournal of Swarm Intelligence and Evolutionary Computation, 9(3), 1–2.
  • Kaplan, A. (2020d). Marrying cultural heritage and high tech: The use of ARTificial Intelligence in museums (Beijing’s Palace Museum, NYC’s Met, and Paris’ Louvre). Case Centre.
  • Kaplan, A. (2020e). Retailing and the ethical challenges and dilemmas behind artificial intelligence. In P. Eleonora (Ed.), Retail futures: The good, the bad and the ugly of the digital transformation (pp. 181–191). Emerald.
  • Kaplan, A. (2021a). Artificial intelligence (AI): When humans and machines might have to coexist. In P. Verdegem (Ed.), AI for everyone: Critical perspectives (pp. 21–32). University of Westminster Press.
  • Kaplan, A. (2021b). Higher education at the crossroads of disruption: The university of the 21st century. Emerald.
  • Kaplan, A. (2022a). Artificial intelligence, business, and civilization: Our fate made in machines. Routledge.
  • Kaplan, A. (2022b). Digital transformation and disruption of higher education. Cambridge University Press.
  • Kaplan, A. (2022c). Fast fashion goes green? AI and the apparel industry (H&M, Shein, and Zara). Case Centre.
  • Kaplan, A. (2023a). Business schools post-covid-19: A blueprint to survival. Taylor & Francis.
  • Kaplan, A. (2023b). Fast fashion’s fate: Artificial intelligence, sustainability, and the apparel industry. Palgrave Macmillan.
  • Kaplan, A., & Haenlein, M. (2009). The fairyland of second life: About virtual social worlds and how to use them. Business Horizons, 52(6), 563–572.
  • Kaplan, A., & Haenlein, M. (2010). Mondes virtuels: Retour au réalisme. Expansion Management Review, 138, 90–102.
  • Kaplan, A., & Haenlein, M. (2016). Higher education and the digital revolution: About MOOCs, SPOCs, social media and the cookie monster. Business Horizons, 59(4), 441–450.
  • Kaplan, A., & Haenlein, M. (2019). Siri, Siri in my hand, who is the fairest in the land? On the interpretations, illustrations, and implications of artificial intelligence. Business Horizons, 62(1), 15–25.
  • Kaplan, A., & Haenlein, M. (2020). Rulers of the world, unite! The challenges and opportunities of artificial intelligence. Business Horizons, 63(1), 37–50.
  • Kelley, S. (2022). Employee perceptions of the effective adoption of AI principles. Journal of Business Ethics, 178(4), 871–893.
  • Kessler, M. (2019, February 21). The Met x Microsoft x MIT: A closer look at the collaboration. The Met.
  • Khalil, O. (1993). Artificial decision-making and artificial ethics: A management concern. Journal of Business Ethics, 12(4), 313–321.
  • Kopalle, P. K., Gangwar, M., Kaplan, A., Ramachandran, D., Reinartz, W., & Rindfleisch, A. (2022). Examining artificial intelligence (AI) technologies in marketing via a global lens: Current trends and future research opportunities. International Journal of Research in Marketing, 39(2), 522–540.
  • Krakowski, S., Luger, J., & Raisch, S. (2022). Artificial intelligence and the changing sources of competitive advantage. Strategic Management Journal.
  • Krotov, V. (2017). The Internet of Things and new business opportunities. Business Horizons, 60(6), 831–841.
  • Li, G., Li, G., & Kambele, Z. (2012). Luxury fashion brand consumers in China: Perceived value, fashion lifestyle, and willingness to pay. Journal of Business Research, 65, 1516–1522.
  • López, T., Riedler, T., Köhnen, H., & Fütterer, M. (2021). Digital value chain restructuring and labour process transformations in the fast-fashion sector: Evidence from the value chains of Zara & H&M. Global Networks, 22(4), 684–700.
  • McCarthy, J. (2007). What is artificial intelligence? Stanford University.
  • McCarthy, J., Minsky, M. L., Rochester, N., & Shannon, C. E. (2006). A proposal for the Dartmouth summer research project on artificial intelligence—August 31, 1955. AI Magazine, 27(4), 12–14. (Original work published 1955)
  • McCorduck, P. (2004). Machines who think: A personal inquiry into the history and prospects of artificial intelligence. A K Peters.
  • Minsky, M. L. (1968). Semantic information processing. MIT Press.
  • Nassauer, S. (2017, July 19). Robots are replacing workers where you shop: Walmart and other large retailers, under pressure from Amazon, turn to technology to do workers’ rote tasks. Wall Street Journal.
  • Paschen, U., Pitt, C., & Kietzmann, J. (2020). Artificial intelligence: Building blocks and an innovation typology. Business Horizons, 63(2), 147–155.
  • Pereira, A. M., Moura, J. A. B., Costa, E. D. B., Vieira, T., Landim, A. R. D. B., Bazaki, E., & Wanick, V. (2022). Customer models for artificial intelligence-based decision support in fashion online retail supply chains. Decision Support Systems, 158(C), 113795.
  • Roberts, H., Cowls, J., Morley, J., Taddeo, M., Wang, V., & Floridi, L. (2020). The Chinese approach to artificial intelligence: An analysis of policy, ethics, and regulation. AI & Society, 36, 59–77.
  • Rubel, A., & Jones, K. M. L. (2016). Student privacy in learning analytics: An information ethics perspective. The Information Society: An International Journal, 32(2), 143–159.
  • Rubio, D., Lastra, C., Frey, C. B., Colclough, C., Jonsson, O., de Tena, C. L., & Menéndez, I. (2019). European tech insights 2020. IE University.
  • Saarikko, T., Westergren, U. H., & Blomquist, T. (2017). The Internet of Things: Are you ready for what’s coming? Business Horizons, 60(5), 667–676.
  • Scherer, M. (2016). Regulating artificial intelligence systems: Risks, challenges, competencies, and strategies. Harvard Journal of Law & Technology, 29(2), 354–400.
  • Schwab, K. (2017). The fourth industrial revolution. Penguin.
  • Shrestha, Y. R., Ben-Menahem, S. M., & Von Krogh, G. (2019). Organizational decision-making structures in the age of artificial intelligence. California Management Review, 61(4), 66–83.
  • Sullivan, Y., & Wamba, F. S. (2022). Moral judgments in the age of artificial intelligence. Journal of Business Ethics, 178(4), 917–943.
  • Taddeo, M., & Floridi, L. (2018). How AI can be a force for good. Science, 361, 751–752.
  • Todd, L. (2017, March 22). A professor built an AI teaching assistant for his courses—And it could shape the future of education. Business Insider.
  • Toth, Z., Caruana, R., Gruber, T., & Loebbecke, C. (2022). The dawn of the AI robots—Towards a new framework of AI robot accountability. Journal of Business Ethics, 178(4), 895–916.
  • Turing, A. (1950). Computing machinery and intelligence. Mind, 59(236), 433–460.
  • United Nations Environment Programme. (2019). UN alliance for sustainable fashion addresses damage of fast fashion.
  • Zawacki-Richter, O., Marín, V. I., Bond, M., & Gouverneur, F. (2019). Systematic review of research on artificial intelligence applications in higher education: Where are the educators? InternationalJournal of Educational Technology in Higher Education, 16(39), 1–27.
  • Zelaya, I. (2021, January 1). This Verizon-powered virtual art experience brings the Met to you. Adweek.
  • Zwakman, D. S., Pal, D., & Arpnikanondt, C. (2021). Usability evaluation of artificial intelligence-based voice assistants: The case of Amazon Alexa. SN Computer Science, 2(1), 1–16.

Related Articles

  •  

Posted

in

by

Salva Trios

Tags:

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *