1 00:00:00,050 --> 00:00:00,740 Case study. 2 00:00:00,740 --> 00:00:06,020 Navigating Global Ethical Standards for AI Omnivision's Facial Recognition deployment. 3 00:00:06,020 --> 00:00:11,750 Facial recognition technology is a compelling example of how global ethical guidelines for artificial 4 00:00:11,750 --> 00:00:14,000 intelligence diverge and converge. 5 00:00:14,000 --> 00:00:20,840 In 2023, a multinational tech firm, Omnivision, planned to deploy a new AI driven FRT system. 6 00:00:20,870 --> 00:00:26,810 Omnivision operates in the European Union, the United States, China and Japan, making it an excellent 7 00:00:26,810 --> 00:00:30,920 case for comparing the ethical frameworks governing AI in these regions. 8 00:00:33,050 --> 00:00:38,720 In the EU, the stringent General Data Protection Regulation requires that any use of biometric data, 9 00:00:38,750 --> 00:00:43,730 such as FRT, must comply with strict privacy and data protection standards. 10 00:00:44,270 --> 00:00:49,760 The EU's Ethics Guidelines for trustworthy AI, authored by the High Level Expert Group on Artificial 11 00:00:49,760 --> 00:00:54,410 Intelligence, emphasized transparency, accountability and fairness. 12 00:00:54,440 --> 00:01:00,110 Omnivision's European branch had to ensure that their FRT system could be easily understood by users, 13 00:01:00,110 --> 00:01:02,630 and included options for human oversight. 14 00:01:04,130 --> 00:01:09,300 How could Omnivision balance innovation while ensuring compliance with these strict regulations. 15 00:01:10,500 --> 00:01:16,650 In contrast, the US approach to AI ethics is more decentralized, relying on sector specific guidelines 16 00:01:16,650 --> 00:01:18,810 rather than a unified framework. 17 00:01:18,930 --> 00:01:24,240 For example, the National Institute of Standards and Technology released a draft framework for managing 18 00:01:24,270 --> 00:01:29,820 AI risks, focusing on the accuracy, reliability and security of AI systems. 19 00:01:30,030 --> 00:01:34,950 This framework is voluntary, aiming to foster innovation while addressing AI risks. 20 00:01:34,980 --> 00:01:40,620 Omnivision's US branch had more leeway but faced a fragmented regulatory environment. 21 00:01:41,070 --> 00:01:47,250 How can Omnivision ensure consistent ethical standards across different sectors and states in the US? 22 00:01:48,480 --> 00:01:54,060 China's ethical guidelines for AI, as outlined by the Beijing Academy of Artificial Intelligence, 23 00:01:54,060 --> 00:01:57,360 emphasized the collective good over individual rights. 24 00:01:57,720 --> 00:02:04,080 The Beijing AI principles focus on harmony, fairness, inclusivity, respect for privacy and security. 25 00:02:04,080 --> 00:02:05,220 Controllability. 26 00:02:05,580 --> 00:02:11,520 China's emphasis on societal stability and national security means that FRT is widely used in public 27 00:02:11,520 --> 00:02:13,870 surveillance and social credit systems. 28 00:02:13,900 --> 00:02:19,840 Omnivision's Chinese branch had to align its FR system with these principles, prioritizing societal 29 00:02:19,870 --> 00:02:22,000 benefits over individual privacy. 30 00:02:22,570 --> 00:02:28,450 Given this context, how should Omnivision address potential global criticisms regarding privacy violations 31 00:02:28,450 --> 00:02:29,200 in China? 32 00:02:30,550 --> 00:02:37,000 Japan takes a unique approach to AI ethics, prioritizing human centric AI and harmonious coexistence 33 00:02:37,000 --> 00:02:39,010 between humans and technology. 34 00:02:39,370 --> 00:02:44,770 The Japanese government's social principles of human centric AI emphasize education, literacy, and 35 00:02:44,770 --> 00:02:46,600 public understanding of AI. 36 00:02:47,230 --> 00:02:53,050 Omnivision's Japanese branch needed to ensure that the FRT system was designed to serve societal good 37 00:02:53,050 --> 00:02:54,940 and improve human well-being. 38 00:02:54,940 --> 00:03:00,850 How can Omnivision integrate public education and understanding into its deployment strategy in Japan? 39 00:03:02,380 --> 00:03:08,110 International organizations like the organization for Economic Cooperation and Development have also 40 00:03:08,110 --> 00:03:10,750 contributed to AI ethical guidelines. 41 00:03:11,200 --> 00:03:17,590 The OECD AI principles stress international cooperation, transparency and accountability. 42 00:03:17,620 --> 00:03:23,290 Omnivision operating globally, had to consider these international principles to ensure that its free 43 00:03:23,320 --> 00:03:26,350 system was aligned with broader global standards. 44 00:03:26,380 --> 00:03:32,350 How can Omnivision foster international cooperation and adhere to the diverse guidelines of the regions 45 00:03:32,350 --> 00:03:33,340 it operates in? 46 00:03:34,660 --> 00:03:40,810 A significant challenge for Omnivision was ensuring the ethical deployment of its FR system while navigating 47 00:03:40,810 --> 00:03:42,910 different regulatory landscapes. 48 00:03:43,120 --> 00:03:49,000 For instance, in the EU, transparency and data protection were paramount to comply. 49 00:03:49,030 --> 00:03:54,520 Omnivision implemented advanced encryption techniques and offered clear, user friendly explanations 50 00:03:54,520 --> 00:04:01,300 of how their FR system processed data comparative to the US, where sector specific guidelines prevailed. 51 00:04:01,330 --> 00:04:07,000 Omnivision adopted a flexible design allowing customization to meet varying state regulations. 52 00:04:07,360 --> 00:04:13,660 In China, Omnivision emphasized the societal benefits of its FR system, such as enhancing public safety 53 00:04:13,660 --> 00:04:16,510 while ensuring compliance with local regulations. 54 00:04:16,630 --> 00:04:22,270 In Japan, Omnivision focused on public education campaigns, highlighting the benefits and ethical 55 00:04:22,270 --> 00:04:26,960 considerations of their FR system to align with human centric principles. 56 00:04:28,730 --> 00:04:34,670 Omnivision's case illustrates the complexities and necessities of balancing innovation with ethical 57 00:04:34,670 --> 00:04:36,710 considerations in AI deployment. 58 00:04:37,220 --> 00:04:42,620 The company had to navigate diverse ethical guidelines and regulatory requirements while maintaining 59 00:04:42,620 --> 00:04:46,490 a commitment to transparency, fairness, and societal well-being. 60 00:04:47,150 --> 00:04:52,760 By examining this case, students can better understand the practical implications of global AI ethical 61 00:04:52,760 --> 00:04:56,840 guidelines and develop strategies for responsible AI deployment. 62 00:04:58,880 --> 00:05:04,700 Addressing the questions posed throughout this case study, Omnivision adopted a multifaceted approach 63 00:05:05,210 --> 00:05:06,200 in the EU. 64 00:05:06,200 --> 00:05:12,320 Balancing innovation and compliance required investing in robust data protection measures and transparent 65 00:05:12,320 --> 00:05:13,880 communication with users. 66 00:05:14,270 --> 00:05:20,390 In the US, ensuring consistent ethical standards involved creating a flexible system design adaptable 67 00:05:20,390 --> 00:05:23,060 to different sectors and state regulations. 68 00:05:23,570 --> 00:05:29,750 In China, addressing global criticisms about privacy violations meant emphasizing societal benefits 69 00:05:29,750 --> 00:05:35,370 and adhering to local ethical principles while transparently communicating the system's purpose and 70 00:05:35,370 --> 00:05:36,300 safeguards. 71 00:05:37,020 --> 00:05:42,660 In Japan, integrating public education into the deployment strategy involved collaborating with local 72 00:05:42,660 --> 00:05:48,810 educational institutions and media to raise awareness about AI ethics and the benefits of free. 73 00:05:49,440 --> 00:05:55,110 Finally, fostering international cooperation and adherence to global standards required Omnivision 74 00:05:55,110 --> 00:06:01,140 to actively participate in international forums and align its practices with OECD principles promoting 75 00:06:01,170 --> 00:06:04,560 transparency, accountability and cooperation. 76 00:06:06,090 --> 00:06:12,060 In conclusion, the case of Omnivision's first system deployment highlights the importance of understanding 77 00:06:12,060 --> 00:06:18,210 and navigating the ethical guidelines for AI across different regions by addressing the unique cultural, 78 00:06:18,210 --> 00:06:23,130 social and political contexts of the EU, US, China and Japan. 79 00:06:23,160 --> 00:06:27,300 Omnivision was able to deploy its technology responsibly and ethically. 80 00:06:28,110 --> 00:06:33,720 This case underscores the need for ongoing dialogue, collaboration, and a commitment to transparency, 81 00:06:33,750 --> 00:06:38,040 fairness, and societal well-being in the global governance of AI.