www.ifc.org/thoughtleadership NOTE 76 • DEC 2019 Artificial Intelligence and 5G Mobile Technology Can Drive Investment Opportunities in Emerging Markets By Peter Mockel and Baloko Makala The intersection of artificial intelligence and 5G mobile technology has enormous potential to deliver dramatic improvements in productivity, efficiency, and cost across business sectors and broader society, delivering innovative products and services not previously possible. Though mainstream applications that combine AI and 5G have yet to emerge, key emerging markets sectors such as agribusiness, healthcare and education will be transformed by the combination of AI and 5G. While many mobile operators remain focused on recouping their investments in previous network standards, there is a growing interest in 5G networks globally. Artificial Intelligence (AI) refers to a growing body of to profoundly transform society as we know it. Although computational techniques relating to computer systems a majority of emerging markets have not yet rolled out 5G capable of performing tasks that would otherwise require networks, they can still reap some of the benefits of AI and human intelligence. Examples include the diagnosis of cellular communication technology. Even with a 2G cellular diseases, solving complex mathematical equations, and network, various successful business models have emerged analyzing electronic circuits.1 across emerging markets.4 For the purpose of this note, we follow the definition and What is 5G Cellular Technology? description of basic, advanced, and autonomous artificial The mobile communication industry is on the verge of intelligence put forward in previous EM Compass Notes. 2 yet another technological revolution: the fifth cellular Essentially, AI is the science and engineering of making technology generation, commonly referred to as 5G. Unlike machines intelligent, especially intelligent computer previous generations, 5G is expected to provide optimized programs.3 This also means that AI is not one type of support for a variety of different services, different traffic machine or robot, but a series of approaches, methods, and loads, and different end-user communities. 5G networks technologies that display intelligent behavior by analyzing are anticipated to impact virtually all processes of everyday their environments and taking actions—with some degree life and will be treated as critical national infrastructure.5 of autonomy—to achieve specific objectives. Higher spectral efficiency means that 5G networks can While AI alone offers innovations, it is the combination of transmit more bits of data per second and per hertz of AI and other technologies such as 5G that has the potential spectrum than previous cellular generations. This is About the Authors Peter Mockel, Principal Industry Specialist, Climate Strategy & Business Development, Economics and Private Sector Development, IFC. His email is pmockel@ifc.org. Baloko Makala, Consultant, Thought Leadership, Economics and Private Sector Development, IFC. Her email is bmakala@worldbank.org. 1 This publication may be reused for noncommercial purposes if the source is cited as IFC, a member of the World Bank Group. 50 use cases, for example home alarms or industrial machinery 45 monitoring. 5G introduces new network services dedicated Max Link Efficiency (Bits/s/Hz) 40 to IoT that can be tailored to lower-cost and lower- 35 bandwidth IoT applications, allowing efficient and low-cost 30 blanket coverage of large fleets of IoT objects. 25 20 The Industrial IoT in particular will be an important 15 factor in industrial competitiveness. There will be a direct 10 link between the availability of 5G network services and 5 economic development. Economies that lack 5G technology 0 will find themselves at a clear disadvantage. With the 1G 2G 3G 4G/SISO 4G/MIMO 5G expansion of 5G in urban core networks around the world, FIGURE 1 Spectral Efficiency of Mobile Systems consumers are already experiencing a significant increase Source: Ortenga; Ortenga Blog at ortenga.com/2016-2/. in their data rates as high as 1 Gigabit per second (Gbps) Note: SISO = single-input and single-output; MIMO = multiple-input and over the air. Below is a map of the roll-outs of 5G networks multiple-output. in cities around the world. Nonetheless, it is important to significant because spectrum is both scarce and expensive. mention that in real-world transformation, 5G will continue Higher spectral efficiency means more concurrent users can to coexist with previous technologies for many years.6 be served at lower cost. The transition between network generations is less sharp Another key difference between 5G and earlier mobile than it seems. As existing 4th generation networks are network generations is the higher priority of Internet of pushed to their limit in what is referred to as 4G LTE-A Things, or IoT, applications. Earlier generations served the (Long Term Evolution – Advanced), the step change to the IoT market through mobile modems that provided full- first 5G networks is more gradual. service access for machine-to-machine applications. That Indeed, the adoption of 5G in emerging markets will take provided an adequate but expensive solution for high-value time. In Sub-Saharan Africa, for example, 5G adoption FIGURE 2 5G Roll-out in Cities Around the World Source: Ookla Interactive 5G Map. www.ookla.com. Note: Red indicates commercial availability, i.e. a 5G network is present and devices are available for consumers to purchase and use; purple indicates limited availability, i.e. a 5G network is present but devices are limited to select consumers; pink indicates pre-release, i.e. 5G network hardware is in place but is currently in testing and/or not yet accessible to consumers; dark blue dots indicate the number of multiple networks of any combination of red, purple, or pink that are revealed when zooming in in the interactive map showing, for example, the high number of such networks in Kuwait and in and around Kuwait City. 2 This publication may be reused for noncommercial purposes if the source is cited as IFC, a member of the World Bank Group. GENERATION 2G EDGE 3G HSPA 3G HPSA+ 4G LTE Cat 4 4G+ LTE-A Cat 6–16 5G MAXIMAL SPEED 0.3 Mbps 7.2 Mbps 41 Mbps 150 Mbps 300 Mbps–1 Gbps 1–10 Gbps AVERAGE SPEED 0.1 Mbps 1.5 Mbps 4 Mbps 15 Mbps 30–90 Mbps 150–200 Mbps LATENCY IN SECONDS 0.5 0.1 0.05 0.001 FIGURE 3 Comparing Speeds of Different Generations From 2G to 5G Source: Authors. Note: Mbps = Megabit per second; Gbps = Gigabit per second; HPSA = high speed packet access; LTE-A = Long Term Evolution - Advanced; Cat = Category. For detailed and accessible explanations see: https://kenstechtips.com/index.php/download-speeds-2g-3g-and-4g-actual-meaning. is not imminent.7 Mobile operators are still recouping multi-party video calls are just about practical with the their investment in previous technology standards. Thus, current 4G networks, and they will also be one of the initial it is expected that there will not be any large-scale 5G key applications for 5G, albeit with improved quality. deployment before 2025, although 5G pilot testing is Beyond that, the first set of applications that are firmly outside already underway in many markets.8 In Egypt, it was the performance envelope of even the best 4G networks, and reported that 5G would be used during the African Cup of therefore require 5G, include autonomous driving, augmented Nations.9 Cellular provider Rain South Africa launched the reality, and online gaming with low latency. country’s first 5G network in September 2019.10 Across the Latin America region, 18 5G trials were Potential of AI and 5G in Emerging Markets: identified across the region as of August 2019.11 However, The world is embracing a new era characterized by it is anticipated that Latin America will trail the rest of the breakthroughs in emerging technologies, the rise of artificial world in terms of 5G adoption with an 8 percent adoption intelligence, and faster data rates, the combination of which rate by 2025. The future of 5G in the region is dependent is set to transform virtually all facets of everyday life through on macroeconomic stability, government policies, and the automation and machine-enabled decision making. use of 4G.12 Undoubtedly, AI and 5G are set to become the engines of In the Asia-Pacific region, on the other hand, the lack of this new technological revolution. However, the majority clear business cases and lack of demand were identified as of use cases that have been proposed for 5G involving AI the main impediments to 5G adoption.13 To illustrate this, cannot be considered true 5G use cases as they do not necessarily require a generational change. What will truly BOX 1 5G and Latency make the difference are applications that would leverage an improvement in latency reduction, that is, response time In networks, latency describes the time it takes for a corresponding server to reply when a packet of less than 1 millisecond. data is sent. It is therefore different from the speed The AI and 5G journey in emerging markets will most of downloading data packets. Why does reduced likely involve enhancing existing use cases14 and the latency matter? Take an example of an automobile travelling at a speed of 50 mph using 4G LTE Cat 4 development of new ones that are yet to be addressed by with a latency of 0.05 seconds. At this speed the current technologies. There are many applications that can vehicle would travel another 1 meter before the be made possible as a result of the combination of AI and corresponding server would respond (the distance faster data rates supported by 5G networks. These include: traveled in 0.05 seconds). With 5G, that latency would be reduced by 98 percent, to just one Enhanced Mobile Communication Network and Services. millisecond (0.001 seconds). During this reduced The provision of mobile communication services is time the car would only travel 2 centimeters becoming increasingly complex. The roll-out of 5G networks before the server responds. It can be argued that is much more challenging than previous generations as it the reduced latency is at least as powerful as the increased speed of downloading in opening up requires upgrades in radio, edge, transport, core, and cloud new applications, as counterparts are reached infrastructure that would ideally be optimally managed with faster and can reply faster. AI to address the complexity associated with 5G networks and the billions of IoT devices these networks can support. 3 This publication may be reused for noncommercial purposes if the source is cited as IFC, a member of the World Bank Group. In addition, the combination of 5G and AI will bring about a for potential improvements. These include cost reductions, host of new or improved technology applications and services. network optimization, and increased revenue streams, Entertainment is expected to become richer and more social which stand at 48 percent, 35 percent, and 41 percent thanks to high-speed connectivity capable of delivering improvement, respectively.18 holograms, augmented reality, and other virtual applications. Sub-Saharan Africa is the region with the highest mobile AI and 5G represent a symbiotic combination and enable subscription growth rate globally. Subscription rates in the each other. 5G relies on automated optimization based region are expected to grow at a compound annual growth on AI algorithms. 5G is too complex to work with the rate (CAGR) of 5 percent, to over 900 million subscriptions static, manual planning optimization of earlier mobile between 2017 and 2023. That will lead to a penetration network generations. 5G networks change their topology rate of 105 percent,19 which indicates availability on average dynamically, responding to changes in traffic. The better the of more than one mobile phone per user. Digitalization optimization works, the more efficiently the network will is rising across the region and represents $10.5 billion in perform in terms of spectrum and energy use. For example, potential revenue by 2026 across key industries in Africa.20 5G relies on Self-Organizing Network (SON) technology. Affordability remains an important factor in the adoption SON measures radio parameters such as path loss and of AI and 5G-enabled services in emerging markets. Most of throughput constantly for different frequencies, drawing the potential applications for these technologies in emerging conclusions about optimal performance and automatically markets hinge on the development of affordable network adapting network settings. technologies in industries such as agriculture, education, As a result, mobile network operators are currently testing health, mining, and others.21 AI due to increased network complexity that requires an Other examples for the pivotal role of AI in 5G network intelligent and automated approach.15 It is expected that operations include:22 more than half of mobile operators will incorporate AI • Support Vector Machines (SVM) for path loss prediction into their networks by the end of 2020.16 According to the models in urban environments. key findings of a study conducted by Ericsson, with 132 service providers globally as respondents,17 AI is already • Machine learning for dynamic frequency and being incorporated into cellular networks primarily bandwidth allocation in self-organizing dense small cell because of a potential reduction in capital expenditures. deployments. Service providers testing AI are focused on several areas • Unsupervised learning for cooperative spectrum sensing. Delay Autonomous Tactile Services that can be delivered 1ms Augmented by legacy networks driving Reality internet Virtual Services that could be Reality enabled by 5G 10ms Disaster Real time Multi-person alert gaming video call Fixed Automotive Bi-directional Nomadic ecall remote controlling 100ms Device On the go remote First responder controlling connectivity M2M connectivity 1,000ms Personal Wireless cloud cloud-based Monitoring Video office sensor networks streaming Bandwidth <1 Mbps 1 Mbps 10 Mbps 100 Mbps >1 Gbps throughput FIGURE 4 Bandwidth and Latency Requirements of Potential 5G Use Cases Source: GSMA 4 This publication may be reused for noncommercial purposes if the source is cited as IFC, a member of the World Bank Group. And AI applications are also dependent on 5G technology, it is used in combination with industrial IoT. That is, IoT requiring its ubiquitous, high-bandwidth, low-latency used in a non-manufacturing environment, for example connectivity. While 5G improves on bandwidth, the from windows and doors to air conditioning control hub technology’s ability to cut response times represents a units, and almost all objects and tools we use on a daily qualitative improvement for AI applications: Lower latency basis. Although 4G LTE could deliver on some of these is not just a qualitative improvement for AI applications but requirements, only 5G can offer them all. a qualitative step change. In addition to radically reducing Drones are another important application. They represent latency, 5G makes latency definable as a Quality of Service a $100 billion market and have multiple applications, in (QoS) parameter. That is, the network can guarantee particular public safety through information gathering and response times, whereas earlier network generations inspection in remote regions. 24 Drones also have been used delivered on a best-effort basis. across emerging markets to address different transport and Taken together, higher bandwidth, ubiquitous availability, logistical challenges. and guaranteed latency performance make it possible to In Argentina, for example, drones have been used in forest run critical AI applications in real time. For example, management to offer cost-effective, high-resolution imagery.25 vehicle-to-vehicle communication (V2V) for accident The World Economic Forum highlighted how Africa helped avoidance requires communication between fast-moving the drone industry get off the ground as the continent is cars at guaranteed maximum latency. experiencing a drone revolution with pioneers using drones If one car warns another car of an obstacle on the road, the to address various challenges.26 For example, the U.S. startup system needs to rely on the message not being delayed by the Zipline, in partnership with the Rwandan government, made network. In general, any application that requires acquisition headlines with the first delivery of blood supplies to remote of continuous high-volume data (for example, video) and the medical facilities. real-time control of assets based on their analysis will require a 5G network. Drones require high-speed connectivity in order to perform command and control, media sharing, and autonomous Furthermore, AI is deemed to be vital for improved customer flying. 5G networks are well placed to provide machine-to- service and enhanced customer experience. Because of AI, machine connectivity while meeting stringent requirements service providers anticipate recouping their investments in 5G. of latency, throughput, capacity, and availability. 27 Potential 5G and AI Uses HAWK 30 drone, an Alphabet-backed project developed at From crisis-related challenges such as responses to natural the University of Hawaii, is a solar-powered drone beaming or man-made catastrophes, or using drone capability for a 5G signal that can fly non-stop for six months at a time. economic empowerment through use of low-altitude sensors The project is currently in testing phase. 28 to improve farming yields, AI-powered applications could In South Korea, drone manufacturer Percepto developed leverage 5G networks and by doing so deliver real societal drone-in-a-box systems that are designed for industrial and benefits and impact.23 enterprise applications in areas such as security optimization or Other potential and impactful uses include: business operations using aerial visual insights while reducing The automotive industry, which is betting big on 5G risk and operational costs. Percepto recently demonstrated connectivity and its potential when combined with AI. autonomous drones using SK Telecom’s 5G trial network.29 Today, virtually every major car manufacturer has developed Public safety and security will benefit from the proliferation its own autonomous vehicle. Manufacturers such as Tesla of IoT devices. The main impediments to the emergence of a and Toyota are testing self-driving vehicles. These vehicles “smart city” have been speed and bandwidth. The combination rely on sensors to continuously detect their surroundings, first of IoT devices and 5G through advances in analytics and by identifying and classifying the information (perception), AI could enable applications around public transit law followed by acting on the information through autonomous enforcement. In 2017, U.S. fire and police agencies acquired control of the vehicle. The success of this new transportation drones for proactive policing. In China, at the 2019 “5G is business model will depend on the availability of network ON” summit, the China Mobile Industrial Research Institute coverage, low latency, and fast connection speeds. project, in which 5G drones provide firefighting support, was However, one of the most promising aspects of 5G is when bestowed a Mobile Internet Innovation Pioneer award. 5 This publication may be reused for noncommercial purposes if the source is cited as IFC, a member of the World Bank Group. Utilities and Energy. Smart meters are already sector is in full expansion with a market size of $530 commonplace in many households around the world. The million in 2017, which is expected to reach $1.4 billion by modernization of smart grids is becoming an imperative 2022. The use of AI is limited in this sector, but there is in order to reduce hydropower generation inefficiencies, considerable potential. fault prediction, decision making, theft prevention, and The combination of AI and 5G can help make immersive load balancing. Data communication has been critical education methods such as virtual reality possible, as many to efficient power generation and consumption. AI is such methods require high bandwidth and low latency expected to be an important component in this effort, as to perform optimally. With 5G, download speeds could the increasing use of sensors means a constant demand be reduced to seconds, freeing teachers to use their time for faster data speeds. This could be a possible application in other areas. In addition, IoT devices can be used to for AI and 5G, although the energy sector appears to automate administrative tasks.34 be following a modest adoption of 5G because of the reluctance of utilities to test new technologies.30 Challenges and Next Steps In agriculture, smart farming is not an option but a necessity AI and 5G represent a powerful combination of if the industry is to keep pace with a growing world technologies that will be the engine of the Fourth Industrial population, particularly at a time when crop yields are being Revolution. However, true applications for the combination affected by climate change. Sensor technology is already of AI and 5G that would trigger massive adoption have being used in agriculture through IoT devices that allow yet to emerge. Also, the adoption of both technologies has farmers to better measure critical factors such as moisture, associated challenges. fertilization, and weather patterns. Global technology Artificial intelligence is already having a profound impact on company XAG develops drones, IoT, AI, and other digital business and industrial processes where machines are taking tools to help farmers effectively grow high-quality produce over tasks previously performed by humans. The ongoing without excessive pressure on the environment.31 High speed ethical debate around AI aside, there is now an urgency to connectivity in rural areas remains a challenge, however, prepare and/or reskill current and future labor forces in which is an obstacle for 5G to address. preparation for the job displacement that will result from AI in healthcare offers new avenues to solve health the intersection of AI and 5G, as together they offer a level challenges. For example, Ada Health has developed of productivity and efficiency that humans cannot match. a chatbot in Swahili that helps patients and doctors Therefore, the focus of retraining and reskilling should be on diagnose diseases ranging from malaria to diabetes.32 human skills that machines cannot model. The combination of fast data speeds and AI could bolster 5G is part of a succession of cellular technology standards healthcare quality, particularly when AI is already used that have transformed the way we communicate—and in the detection of diseases or in the reduction of costs. indeed the way we live. Each of these generations brought Other applications in healthcare that would require high- about a key differentiator that has had a transformative speed connectivity include remote-control surgery, as well impact on consumers’ communication experiences, yet as downloading large data files, as the healthcare industry each also exhibited weakness that would be addressed by generates massive amount of data daily, including heavy the following generation. Some of these standards are still imagery such as MRI, CAT, and PET scans. The resulting relevant today. For example, primary 2G mobile brought data in turn could be processed using AI for faster diagnosis digital telephony and messaging. Successful business models and treatment. Real-time remote monitoring and sensor were built on the back of this standard. M-Pesa, a Kenyan innovation aimed at developing do-it-yourself innovation mobile fintech application, is arguably one of the most would place medical devices in the hand of patients who successful business models leveraging 2G cellular technology. would be able to monitor their health at home.33 However, 2G network data rates are very low and can The global market for AI in education is expected to reach barely support internet connectivity. The advent of 3G $2 billion globally by 2023, which represents a 38 percent and 4G connectivity enabled faster data transmission, and annual growth rate from 2018 to 2023. While emerging both standards will remain dominant in emerging markets markets constitute just a fraction of this market, they are until at least 2025, which will delay the development and seeing growth, too. In Africa, for example, the eLearning adoption of 5G in those regions. 6 This publication may be reused for noncommercial purposes if the source is cited as IFC, a member of the World Bank Group. Spectrum availability is currently a challenge to the growth Currently, AI and 5G are at the center of geopolitical of 5G. The power and speed of 5G rest on the frequency tensions. These tensions have given momentum to a race for used. One way to circumvent spectrum-related challenges technological innovation dominance in a 5G-enabled world. is to use frequencies with lower bandwidths that may have Regardless of who wins that race, there is a call for close advantageous properties. For instance, a frequency of 600 collaboration between developers and other actors in the MHz may not lose power quickly and therefore can easily cellular communication ecosystem to come up with solid 5G reach 5G phones while overcoming physical obstacles such use cases and appropriate implementation models. Previous as thick walls. network generations will not disappear overnight because of the advent of 5G. It is therefore important for operators Also, the 5G standard is very expensive to implement. Many to devise a clear strategy toward a 5G-enabled digital mobile operators in Africa and elsewhere have yet to make ecosystem and paths to maximize returns on investment. an investment in infrastructure supporting 5G because of the significant capital costs involved and the lack of a clear ACKNOWLEDGMENTS business model that would recoup the investment. Operators The authors would like to thank the following colleagues would need to upgrade base stations, add new base stations, for their review and suggestions: German Cufre, Manager, and upgrade the backhaul high-speed, high-capacity mobile Telecom, Media, and Technology – Global Infrastructure, access networks (4G, 5G etc.) that are often not available IFC; Ferdinand Van Ingen, Senior Industry Specialist, in emerging markets. That renders the investment hurdle Telecom, Media, and Technology – Global Infrastructure, even higher, as not only will operators have to implement IFC; Georges Houngbonon, Economist, Telecom, Media, 5G radio technology, they will also have to upgrade their Technology, Venture Capital and Funds – Sector Economics networks end to end. In contrast, operators in advanced and Development Impact, Economics and Private Sector Development, IFC; Ariana Batori, Investment Officer, Telecom, markets will have already made extensive progress. Media, and Technology – Global Infrastructure, IFC; and 5G also poses unique security challenges because the role of Thomas Rehermann, Senior Economist, Thought Leadership, equipment integrators will be much larger than for previous Economics and Private Sector Development, IFC. generations of mobile technologies. The security risks will Please see the following additional reports and increase significantly as 5G architecture pushes functionality EM Compass Notes about technology and its role in that was once at the core of networks out to the “edges” emerging markets: Reinventing Business Through Disruptive where it is more difficult to control and more vulnerable to Technologies - Sector Trends and Investment Opportunities tampering. With the sheer number of connected devices, the for Firms in Emerging Markets (March 2019); Blockchain: number of unsecured and compromised devices will increase. Opportunities for Private Enterprises in Emerging Markets (January 2019); How Artificial Intelligence is Making Transport These could be used for nefarious purposes, for example in Safer, Cleaner, More Reliable and Efficient in Emerging Markets the case of a distributed denial-of-service, or DDoS, attack. (Note 75, November 2019); Bridging the Trust Gap: Blockchain’s In addition, the volume of data generated will increase the Potential to Restore Trust in Artificial Intelligence in Support of difficulty of detecting abnormal or malicious traffic. Aside New Business Models (Note 74, Oct 2019); Artificial Intelligence: from these technical concerns, a greater share of the global Investment Trends and Selected Industry Uses (Note 71, Sept economic output that would depend on the data generated 2019); The Role of Artificial Intelligence in Supporting Development will be at greater risk.35 in Emerging Markets (Note 69, July 2019). 1 Nilsson, Nils. 1980. “Principles of Artificial Intelligence.” Berlin: Springer Berlin. 2 EMCompass Note 69. 2019. The Role of Artificial Intelligence in Supporting Development in Emerging Markets. 3 Strusani, Davide and Georges Vivien Houngbonon. 2019. “The Role of Artificial Intelligence in Supporting Development in Emerging Markets.” EM Compass Note 69, IFC, July 2019, pp. 1-2. This note defines AI as “the science and engineering of making machines intelligent, especially intelligent computer programs.” This definition is also guided by the AI100 Panel at Stanford University, which defined intelligence as “that quality that enables an entity to function appropriately and with foresight in its environment.” See also “One Hundred Year Study on Artificial Intelligence (AI100).” 2016. Stanford University. https://ai100.stanford.edu/. See also Nilsson, Nils. 2010. “The Quest for Artificial Intelligence: A History of Ideas and Achievements.” Cambridge University Press; OECD. 2019. “Recommendation of the Council on Artificial Intelligence.” OECD Legal 0449 as adopted on May 21, 2019. PwC. 2018. “The Macroeconomic Impact of Artificial Intelligence.” February 2018. 4 Different business models have emerged from the advent of 2G including mobility patterns, offline social networks used in building credit score for microlending, mobile money etc. 5 GSM Association. 2017. “The 5G Era: Age of Boundless Connectivity and Intelligent Automation.” 6 GSM Association. 2017. 7 GSM Association. 2019. “5G in Sub-Saharan Africa: Laying the Foundations.” 8 GSM Association. 2019. 7 This publication may be reused for noncommercial purposes if the source is cited as IFC, a member of the World Bank Group. 9 Telecomreview.com. 2019. “Egypt Set to Experience 5G as of June 2019.” Telecomreview.com, April 25, 2019. 10 Villiers, James de. 2019. “South Africa’s First 5G Network is Now Live in Parts of Johannesburg and Tshwane – Here’s What You’ll Pay.” Businessinsider.co.za, September 18, 2019. 11 Mobile World Live. 2019. “Intelligence Brief: What Will Make 5G take Off in LatAm?” mobileworldlive.com, August 28, 2019. 12 Mobile World Live. 2019. 13 The Economic Times. 2019.“5G Adoption Will Be Slow Across Asia-Pacific: Moody’s.” economictimes.indiatimes.com, July 1, 2019. 14 Please see for several examples of artificial intelligence-based business models: Xiaomin Mou. 2019. “Artificial Intelligence: Investment Trends and Selected Industry Uses”, EM Compass Note 71, IFC, September 2019. 15 “5G and AI: Increased Network Complexity Requires An Intelligent and Automated Approach.” rcrwireless.com, May 28, 2019. 16 Ayvazian, B. 2019. “Employing AI to Enhance Returns on 5G Network Investments - AI Trends.” aitrends.com, September 12, 2019. www.aitrends.com/ai-and-5g/employing-ai-to-enhance-returns-on-5g-network-investments/. 17 Ericsson. 2019. “Employing AI Techniques to Enhance Returns on 5G Network Investments.” 18 Ericsson, R. 2019. 19 Coetzee, Wydeman, Fisseha Mekuria and Zaaid du Toit. 2018. “Making 5G a Reality for Africa.” CSIR & Ericsson. 20 Coetzee, Wydeman et al. 2018. 21 Coetzee, Wydeman et al. 2018. 22 M. E. Morocho Cayamcela and W. Lim. 2018. “Artificial Intelligence in 5G Technology: A Survey.” 2018 International Conference on Information and Communication Technology Convergence (ICTC), Jeju, 2018, pp. 860-865. 23 Chui, M., Harrysson, M., Manyika, J., Roberts, R., Chung, R., Nel, P. and van Heteren, A. 2018. Applying artificial intelligence for social good. [online] McKinsey & Company. Available at: https://www.mckinsey.com/featured-insights/artificial-intelligence/applying-artificial-intelligence-for-social-good. 24 Takacs, Attila, Xingqin Lin, Stephen Hayes and Erika Tejedor. 2018.“Drones & Networks: Safe & Secure Operations.” Ericsson.com. www.ericsson.com/en/white-papers/drones-and-networks-ensuring-safe-and-secure-operations. 25 World Bank. 2016.“Experimenting with Drones to Improve Forestry Management.” June 14, 2016. 26 Berkley, Seth. 2018.“Africa is Helping the Drone Industry Get Off the Ground. Here’s How.” World Economic Forum, October 19, 2018. www.weforum.org/agenda/2018/10/drones-for-development-can-deliver-innovation-for-the-whole-industry-heres-how/. 27 Patel, Nikunj. 2017. “The Sky Is the Limit in 5G Game of Drones.” Blog, einfochips.com, October 27, 2017. 28 Blum, Sam. 2019. “New Gargantuan Internet-Beaming Drone Aims to Succeed Where All Others Have Failed.” Popularmechanics.com, March 7, 2019. www.popularmechanics.com/flight/drones/a26751066/aeroenvironment-hawk-30/. 29 Bausch, Jeffrey. 2019. “Successful Demonstration of Autonomous Drones Flying Over 5G Trial Network.” iot.eetimes.com, May 22, 2019. https://iot.eetimes.com/successful-demonstration-of-autonomous-drones-flying-over-5g-trial-network/. 30 Ericsson. 2019. “5G For Business: A 2030 Market Compass - Setting a Direction for 5G-Powered B2B Opportunities” Ericsson, October 2019. www.ericsson.com/assets/local/5g/the-5g-for-business-a-2030-compass-report-2019.pdf. 31 Xag Website. https://www.xa.com/en. 32 Newey, Sarah. 2019. World’s First AI Health App in Swahili Launches to Tackle Doctor Shortages. telegraph.co.uk, November 19, 2019. 33 Stracuzzi, Marco. 2019. “4 Revolutionary Use Cases of 5G in Healthcare.” telit.com, July 24, 2019. 34 Nocchi, Paul. 2019. “5 Ways 5G Will Make Classrooms Smarter.” Getting Smart.com, April 21, 2019. 35 Triolo, Paul, Kevin Allison. 2018. “The Geopolitics of 5G”. Eurasia Group, November 2018. Additional Selected EM Compass Notes Previously Published by IFC Thought Leadership NOVEMBER 2019 JULY 2019 Note 75: How Artificial Intelligence is Making Transport Note 69: The Role of Artificial Intelligence in Supporting Safer, Cleaner, More Reliable and Efficient in Emerging Development in Emerging Markets Markets JUNE 2019 OCTOBER 2019 Note 68: Basic Business Models for Banks Providing Note 74: Bridging the Trust Gap: Blockchain’s Potential to Digital Financial Services in Africa Restore Trust in Artificial Intelligence in Support of New Business Models APRIL 2019 Note 73: Closing the SDG Financing Gap—Trends and Data Note 67: The Case for Responsible Investing in Digital Financial Services SEPTEMBER 2019 MARCH 2019 Note 72: Blended Concessional Finance: The Rise of Returnable Capital Contributions Note 66: Blended Concessional Finance: Governance Matters for Impact Note 71: Artificial Intelligence: Investment Trends and Selected Industry Uses Note 65: Natural Gas and the Clean Energy Transition AUGUST 2019 FEBRUARY 2019 Note 70: How Insurtech Can Close the Protection Gap in Note 64: Institutional Investing: A New Investor Forum Emerging Markets and Growing Interest in Sustainable Emerging Markets Investments 8 This publication may be reused for noncommercial purposes if the source is cited as IFC, a member of the World Bank Group.