Artificial intelligence is a advanced technology, which drivevariationof economy and society ominously in the proximate future. It tinbe employed to switch human labour ship effecting various perilous then dreary chores, providing us through additional convenient and competent life. It can be a slice from the wide-ranging application of the emerging technology. In the paper, we make discussions on the smart security, privacy, safety and innovative issues in artificial intelligence applications and plug out the possiblehazards and threats. Securityactions in research and supervision are proposed and our anticipation for artificial development will be clearly explained below.
AI techniques, present drive likely stand new robots or brainy programs that can support as human's assistants also ensure a slice for us, such as reading email, housework, or even driving cars. This one will also fetch us control on privacy, security and ethic. Several uncontrolled significances might ascend from AI applications if we be unsuccessful to find and avoid related threats in advance. In this study, we deliberate the latent hazards and threats of artificial intelligence, elevation warnings and stretch suggestions.
Artificial Intelligence and Applications
Artificial intelligence can be categorized by two types. Weak AI and strong AI. In the category of strong AI, AI system must be considered as human-like high level perception ability, such as common sense, self-awareness and creativity, while weak AI simulates human intelligent processes passively without physical understanding. Weak AI is designed to finish a particular task, while strong AI is usually understood a general AI system, which takes the skill to achieve several kinds of intelligent tasks. Current AI systems are entirely at the stage of weak AI and strong AI sort out not yet exist so far. It is imaginary that it would take spans for human to realize strong Atypical AI techniques take in machine learning, speech Recognition, natural languagedispensation,robot,etc. artificial intelligence has remained used universally in our life, from speech text input and tailored network shopping, to several smart answering systems. AI applications in these fields such as education, science, engineering, business, medical care and manufacturing etc.
In AI techniques, smart home-based system tin monitor in our daily life, including sleep time, isometrics, diet predilections, and seizure signs of their changes. Future lavatory perhaps can detect through excreta to regulate if body is healthy, and provide relevant information to doctor. Such system can assist doctors in making decision without consulting specialists which are unusual resources in rural areas and in many developing countries. Finance: Manufacturing and administration can use artificial intelligence to grasp early detection of anomalous financial hazards and to reduce spiteful actions such as manipulating markets, fraud and unusual transactions. There are many AI techniques, like artificial neural networkand support path machine, used by commercial bankers and business consultants to make bankruptcy prediction and company financial distress prediction.
I can ranking and assess students in an intelligent way and help them learn at their own pace.AI technique can build a new education system that includes intelligent evaluation and interactive learning. Through the help of smart education assistant, tutor can be provided with additional support by knowing the status and learning progress of each student at any time.
Smart Security, Privacy in AI Applications
In artificial intelligence systems are facing these types of problems.
There are different types of security problems of AI including security threats of technology abuse, security problem made by technical flaws and self-aware intelligence induced security problems.
- Security threats of technology exploitation: It is believed that artificial intelligence is a neutral technique. If it is harmed by spiteful people, the technique may fetch us problems of security, discretion and tenet. Research shows that attackers can blastoff a large-scale attack with only a slight resource using intelligent methods. Invaders may also use AI technology to access remote information illegally.AI systems may do something harmful to people in control of criminals.
- Security problems induced by technical faults: In recent artificial intelligence system is future from perfect. Sometimes AI system could not be secure as it seems due to some technical faults. Solitary is technical imperfection as robots, tools, controllers and related mechanisms are not correctly designed or tested. One more reason is improper management; affecting many robotic fates occur under unusual working conditions such as software design, maintenance, testing, installation or alteration.
In current centuries, big data focused prototype has conquered artificial intelligence research and has ran to a new tint of AI development. present machine learning, the number and quality of data sets will affect in high degree the training results, so maximum of successful AI applications depend on heavily onbig data. As secrecy problem is a main threat of data analysis, inevitably, there will also be privacy problems in the applications of artificial intelligence. data acquirement: In AI,wide use of smart home devices, multiplicity of data can be kept for years or even decades. These data, if used correctly, will type life better. But some of private information would also be used illegally by technology businesses for commercial purposes. cloud computing: In cloud computing, many corporations and government organizations are drifting the data into cloud, because it is cheap, easy to use, and suitable in getting on-demand system access to a communal tarn. our private information is also stored in cloud, our secrecy need to be make sure.
the most special problem that rapidly changing AI technology may bring to us, almost all scientific and industrial personnel will believe tenet is the focus of attention, due to the human like brainpower ability of this powerful technology. Decent problems might be induced by following issues.
Conduct rule: Artificial intelligence robots must study rules formerly making decisions. If the design of intelligent mediator is not unified with the human restraints, it is likely to track a dissimilar logic with human actions and prime to rotten values. To make new systems help the whole communal, not fair the regulator of the system, through compelling actions of AI systems to fulfill with predefined communal tenet rules.
Destroy of robots. There is a stimulating Query around the tenet of AI system can kill an intelligent robot to grasp that robots are risky for us. The first tricky in this subject is that can kill robots as we famine. The killing of robots in the earlier may be an fate. But, it may be planned someday in the future. If these robots are not pleasant or don't comply with human beings, must control them or kill them.
All these glitches must be careful before in design process of such AI systems.
Hostage measures And Deliberations
Artificial intelligence is able to make human life more powerful, so far with security, privacy, ethic and additional risks at the same time..
Highlighting Safety, Secrecy and Tenet Research
Owing to latent abilities and difficulty of AI, and its close connections with human users, research on the safety of artificial intelligence technology is particularly important. Scholars and academics need to highlight additional on safety protection and try their best to make artificial intelligence more secure.
Implanting tenet rule: the artificial intelligence systems take movements by their own,their performance will be gifted to fulfill with strict and casual rules that humans need to follow, counting ethical, legal rules. These rules should be measured and drive in in the AI system during expansion period.
Refining safety and heftiness: In many gears, artificial intelligence system is intended to operate in a compound situation. AI system should be healthy to contract with surprising conditions and must be safe sufficient to manage with a wide range of deliberate attacks. Beforehand putting an artificial intelligence system into a wide range of applications, it is necessary to make sure that the system is safe, steadfast and controllable.
The artificial intelligence is also significant for conduct the safety, secrecy and decent problems of AI besides the technology research itself. Research on tuning, lawmaking and rule should be accepted out to make sure that the application of artificial intelligence is in control.
Law & Policy creation: artificial intelligence may bring potential terrorizations and risks, the administrations can make laws and related policies to define artificial intelligence can do or what is not permissible. Organization needs to recover terms of laws and lay down rules for AI industry and products usage, such as the concern for the accident of unmanned vehicles, air UAV's assault of special privacy and so on.
Artificial intelligence is developing at an stirring speed. It can fetch us adeptness and suitability, but we essential to evade damage to humans. While wide range AI applications and their influential impact have not seemed in our lives, it is needed to discourse the public and decent issues potency be elevated by AI in advance. Highlighting on security, secrecy and tenet issues had better waged enough care by AI researchers. It is not advisable to carry out severe observation on artificial intelligence especially at present application stage, so as not to build difficulties to technology revolution.
- D. Crevier, 'AI: The tumultuous history of the search for artificial intelligence,' Basic Books, London and New York, 1994.
- M. Flasinski, 'History of artificial intelligence,' Introduction to the History of Computing, Springer-Verlag New York Inc, New York, 2016.
- Executive office of the president of the United States, 'The National Artificial Intelligence Research and Development Strategic Plan,' Washington, October 2016.
- Pannu, M. T. Student, 'Artificial intelligence and its application in different areas,' International Journal of Engineering and Innovative Technology (UEIT), Vol. 4(10), April 20lS, pp. 79-84.
- S. S. Sikchi, S. Sikchi, and M. Ali, 'Artificial intelligence in medical diagnosis,' International Journal of Applied Engineering Research, Vol. 7, Jan. 2012, pp. IS39-1543.
- X. F. Hui, J. Sun, 'An application of support vector machine to companies' Financial Distress Prediction,' Third International Conference on Modeling Decisions for Artificial Intelligence (MDAI 2006), Tarragona, Spain, April 2006, pp. 274-282, doi: 10.1007/11681960 27.
- M. Kandlhofer, G. Steinbauer, S. Hirschmugl-Gaisch and P. Huber, 'Artificial intelligence and computer science in education: From kindergarten to university,' 2016 IEEE Frontiers in Education Conference (FIE2016), Eire, PA, Oct. 2016, pp. 1-9.
- P. Patil. 'Artificial intelligence in cyber security,' International Journal of Research in Computer Applications and Robotics, VoI.4(S), 2016, pp. 1-5.