공공부문에서 인공지능 활용에 관한 연구
A Study on Utilization of Artificial Intelligence in Public Sector
0
조회
0
다운로드
21
인용
0
참고문헌
키워드
초록
1990년대 인터넷이, 2000년대에 스마트폰이 우리의 생활에 들어왔듯이 2010년대 중반부터 인공지능이 활용되고 있다. 인공지능이 공공부문에 적용되면 가장 파괴적인 혁신수단이 될 것이라는 전망도 있지만 아직은 기술 개발단계이라서 파일럿 프로젝트 수준에 머물고 있으며, 챗봇서비스의 경우에도 정확도가 70% 수준에 머물고 있다. 인공지능이 공공부문에 본격적으로 적용되었을 때 나타날 수 있는 결과에 대해 과도하게 기대도, 염려를 하지 않아도 된다. 다만 몇 가기 원칙을 제시할 수 있다. 첫째는 부담과 편익을 모든 집단이 동일하게 공유해야 한다. 둘째, 시민중심으로 프로그램을 만들어야 한다. 셋째, 윤리적인 위험부담은 피하고 인공지능에 의한 의사결정은 줄여야 한다. 마지막으로 거버넌스를 구축해야 한다. 공공분야에서 인공지능을 활용하는데 있어서 약으로 작용할 것인가 아니면 독으로 작용할 것인가의 문제는 인간의 선택에 달려 있다고 할 수 있다.
Abstract
Artificial intelligence technology has been developing rapidly since the mid-2010s. Artificial intelligence technologies such as machine learning, computer vision, natural language processing, and robotics are naturally integrated into various fields to help human life. The private sector has achieved remarkable results, including the introduction of artificial intelligent speakers and the introduction of self-propelled drivers, but the public sector is still in its early stages as a pilot project in chatbots and waste disposal Although it is expected that artificial intelligence will be the most destructive innovation tool when it will be applied to the public sector, it is still in the pilot project level because it is in the technology development stage, and the accuracy of the chatbot service is about 70%. There is no need to worry too much about the consequences of artificial intelligence being applied to the public sector in earnest. Only a few principles can be suggested. First, all groups should share the burden and benefits equally. Second, we should create programs based on citizens. Artificial intelligence should not be implemented unilaterally by the government because its impact on each area of the people is very large. Artificial intelligence should be provided as a tool to solve a given problem. Third, we avoid ethical risks and reduce decision making by artificial intelligence. It should be used to avoid ethical problems that may be caused by bias, privacy invasion, and lack of explanatory power of artificial intelligence. Finally, since the development of artificial intelligence technology is very fast and its impact on human life is very large, it is necessary to establish governance in which participate the public sector, industry, research institute related to it is established to understand the technology trend. The question of whether to act as a medicine or a poison in the use of artificial intelligence in the public sector depends on human choice.
참고문헌 (37)
- "경기도"[URL]
- 과학기술정보통신부, "I-Korea 4.0 실현을 위한인공지능(AI) R&D 전략" (2018)
- 김성우, "빅데이터의 정책적 활용과 과제 - 빅데이터 추진 현황과 정책적 적용 쟁점을 중심으로 -", 한국비교정부학보, 18(3), pp.309~324 (2014), 한국비교정부학회[DOI]
- 김의중, "인공지능, 머신러닝, 딥러닝 입문" (2017), 위키북스
- "대한민국 정책브리핑" (2018)[URL]
- "보안뉴스" (2018)[URL]
- "오마이뉴스" (2017)[URL]
- 윤상오, "인공지능 기반 공공서비스의 주요 쟁점에 관한 연구: 챗봇(ChatBot) 서비스를 중심으로", 한국공공관리학보, 32(2), pp.83~104 (2018), 한국공공관리학회[DOI]
- 윤상오, "인공지능을 활용한 정책결정의 유형과 쟁점에 관한 시론", 한국지역정보화학회지, 21(1), pp.31~59 (2018), 한국지역정보화학회[DOI]
- "딥러닝 기반 객체 인식 기술동향" (2018)[URL]
- 정규환, "인공지능 기술의 현주소 및 향후 전망", KISO Journal, 33, pp.11~15 (2017)
- 정준화, "챗봇(chatbot)의 현황과 향후 과제", 이슈와 논점, 1456, 국회입법조사처
- 최윤섭, "IBM 왓슨 포 온콜로지의 의학적 검증에 관한 고찰", Hanyang Med Rev, 37, pp.49~60 (2017)
- 기계학습 기반 사회조장 빅데이터 분석 및 예측모형 연구, "한국보건사회연구원", 한국보건사회연구원, ~, pp.연구보고서. 2017-46 (2017)
- "인공지능시대의 정부 : 인공지능이 어떻게 정부를 변화시킬 것인가" (2017), 한국정보화진흥원[URL]
- Bessen, James, "Learning by Doing: The Real Connection between Innovation, Wages, and Wealth" (2016), Yale University Press
- "Intro to AI for Policymakers: Understanding the Shift" (2018)[URL]
- "Unleashing the Potential of Artificial Intelligence in the Public Sector" (2017)[URL]
- Carl Benedikt Frey, "The Future of Employment: How Susceptible Are Jobs to Computerization" (2013), Oxford Martin School, University of Oxford
- "Colorado Child Welfare County Workload Study" (2014)[URL]
- "AI-Augmented Government : Using Cognitive Technologies to Redesign Public Sector Work" (2017)[URL]
- "Explainable Artificial Intelligence" (2016)[URL]
- "AI in the UK: Ready, Willing and Able" (2017)[URL]
- "Artificial Intelligence and Machine Learning: Policy Paper" (2017)[URL]
- "Digital Decision-Making: The Building Blocks of Machine Learning and Artificial Intelligence" (2017)[URL]
- "The Bright Side of Patents, USPTO Economic Working Paper No. 2015-5" (2015)[URL]
- "Identifying and Prioritizing Fire Inspections: A Case Study of Predicting Fire Risk in Atlanta" (2015)[URL]
- "NHS illegally Handed Google Firm 1.6m Patient Records" (2017)[URL]
- "Machine Intelligence: Technology Mimics Human Cognition to Create Value" (2017)[URL]
- "First Robot-Written Stories from Press Association Make It into Print in ‘World-First’ for Journalism" (2017)[URL]
- "Artificial Intelligence: Potential Benefits and Ethical Considerations Policy Department C: Citizens' Rights and Constitutional Affairs European Parliament" (2018)[URL]
- Sadilek, Adam, "Deploying nEmesis: Preventing Food Borne Illness by Data Mining Social Media", Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence, pp.3982~3989 (2016), AAAI Press
- "Initial Application of the SURTRAC Adaptive Traffic Signal Control System" (2013)[URL]
- "Artificial Intelligence Life in 2030 : One Hundred Year Study on Artificial Intelligence" (2016)[URL]
- "Artificial Intelligence and Public Policy" (2017)[URL]
- "How Much Time and Money Can AI Save Government? : Cognitive Technologies Could Free Up Hundreds of Millions of Public Sector Worker Hours" (2017)[URL]