Artificial intelligence national security is an important part of the overall national security，Current，In terms of institutional and academic reasoning, it is urgent to carry out an overall plan for the rule of law system for the national security risk of artificial intelligence。For this，On the basis of the common value goals of human beings that need to be regulated by security risks in the field of artificial intelligence，Build an artificial intelligence national security rule of law with Chinese characteristics。

　　bet365 Play online games

The modern usage of "National Security" first appeared in the book "American Foreign Policy" published by American writer Li Pan in 1943。1947，The United States promulgated the world's first "National Security Law"，and formed the "National Security Council"。my country also passed the new "National Security Law" in 2015。Generally speaking，The national security in the sense of traditional law refers to the objective state of various threats and dangers that the national interests confirmed and protected by the law of a country。Internal threats from their own citizens and organizations that threaten national security，External threats from other countries or organizations that threaten national security。Traditional national security is an objective security in a physical security sense，It mainly includes military security and political security。

Compared with the context of traditional law，The national security in the era of artificial intelligence has two major characteristics。First, national security in the era of artificial intelligence is a kind of security，instead of Safety。"Safety" focuses on personal safety and property security，Three principles such as Asimov'bet365 Play online games s robot，It aims to ensure that the robot will not harm humans; "Security" means politics、Economy、Culture、Overall safety in ecology and other aspects，instead of threatening the physical objects such as the country and the land of the country，Challenges brought by data cross -border flow to national security。Second, the national security risks of the artificial intelligence era are in the three -party structure instead of the two parties。Traditional national security is in the two parties of the "national -individual" or "national -country"，The threat to national security comes from individuals or other countries。But in the era of artificial intelligence，"Artificial Intelligence" is added to these two structures。Artificial intelligence may be intended by individuals or other countries to threaten national security，You can also be out of control or misuse of technology、Abuse and generate overflow risks，additional threat to national security。

Generally speaking，In law，National Security in the era of artificial intelligence refers to the misuse of national interests due to artificial intelligence、Overall safety caused by the risk caused by abuse and out of control。This kind of security looks at political security from the outer extension、Economic Security、Military security、Cultural Security、Social Security and other aspects，From the perspective of categories, internal safety and external security、Traditional security and non -traditional safety。There are three requirements: First, the security risks that artificial intelligence is not mature and malicious applications bring to the national society through the law; Intelligent security rule of law，Guarantee of artificial intelligence national security and steady development。

　　The theoretical basis of the rule of law of the national security risk regulation

There is a structural interaction between artificial intelligence technology and national security。On the one hand，Artificial intelligence technology can be based on evolutionary practical application，From traditional safety、bet365 Play online games Non -traditional security and overlapping areas between the two，Provide a more effective maintenance and guarantee mechanism for national security。On the other hand，The application of artificial intelligence technology also has the possibility that beyond expected vision，Existing spillover risk caused by technical abuse and endogenous risk caused by the deficiency of the technical itself。To deal with these two types of risks，It is necessary to first consolidate the theoretical basis of the rule of law of the rule of law of the national security risk regulation: First, the principle of rule of law in the risk society，Second, the principle of change in the innovation model of artificial intelligence technology。

Artificial intelligence security risk regulation is an integral part of risk and social governance。Different from the traditional social model，The most scarce value in the risk society is the pursuit of certainty，Law as a certainty value as the best regulatory mode in the operation of a risk society。At the same time，The governance of risk society also puts forward new requirements that are different from traditional rule of law: First, the law must maintain a balance between stability and flexibility。The era of artificial intelligence brings "Soft Revolution"，Develop legal norms with flexibility and flexibility to bring more good effects to emergency and treatment of major risks。Second, it is necessary to combine laws to combine rules and principles。In risk society，The development prospects of new technologies are not completely determined，The results of social relations brought about by artificial intelligence have not been completely fixed，Principles of the law should be enhanced。On the contrary，Once the social relationship in the relevant field is solidified，or some practices are proven to be valid、It can be stable and unchanged for a long time，can be risen to specific and clear rules。Third, it is required to rely on the way to prevent advance to deal with risks。Scientific and technological risks Bet365 lotto review will cause large -scale、Artificial catastrophic consequences，Principles of prevention should become the main principle of the government with legal means in response to scientific and technological risks。

The idea of ​​regulating the rule of law of artificial intelligence security risks is also related to the changes of artificial intelligence technology innovation model。For more than half a century，Western capitalist legal system mainly reflects Xiong Peter's technology innovation mode。Xiong Peter mode is essentially a benefit innovation mode，Its core connotation is the commercial combination of market elements。It has four main characteristics: there is enough business benefits、Meet user needs、Taxes、High efficiency。Effective innovation model does not require products to meet ethical requirements other than functional factors，Therefore, ethical risks are inevitable。Obviously，This innovation model is no longer suitable for the application of artificial intelligence technology in risk society，In particular, it is not suitable for the application of artificial intelligence technology related to national security。The innovative model protected by laws in the era of artificial intelligence should be upgraded from the innovation model of benefit to a righteous innovation mode。The main feature of the latter mode is: first，It is significantly different from the pursuit of economic benefits with benefit innovation，Righteous and innovative pursuit of collaborative improvement of economic and social benefits。2，Different products/services that meet the specific products/services that meet the needs of user needs，The target object of justice and innovation is the artificial/artificial system that meets the needs of social development。third，Continuing with benefit innovation、Strengthen the tradition of industrial civilization，Justice and innovation exploration more tolerant civilization path。In this new mode，Law's attention to people will be fundamentally bet365 live casino games strengthened，The understanding of people and machines will be greatly updated，The relationship between people and machines will be redefined in law。The law will be under the guidance of the principle of well -being，Promotion、More tolerance of machines and the environment with more inclusive development。

　　The rule of law framework of the national security risk regulation of artificial intelligence

The legal operation concept of the era of artificial intelligence will shift from "how to repair and restore society" to "how to train and shape society"。Artificial Intelligence National Security Risk Regulations should take a focus on prevention、focusing on shaping the relevant subject behavior logic、Mode focusing on risk regulation。Based on this mode，It should be constructed with legal regulations as the core、Artificial Intelligence National Security Risk Regulations of the Law of the rule of law。Specifically，This system should be composed of a triple framework。

First is the value -standard framework，It consists of three levels from high to low: the top layer is the "basic principle"，With the "National Welfare Principles" as the core，supported by the "Trust and Controlling Principles" and "Human Regulatory Principles";，Including legal compliance、Risk prevention、Safe and transparency、Agile governance、Stability (robustness)、Confidentiality、Integration、usability、Fairness and accountability, etc.; The bottom layer is "specific implementation rules"，Including product standards (corporate standards)、Technical Standards (International Standards/National Standards)、Industrial Policy、Industrial Policy、Laws and regulations and ethical codes such as specific requirements and specifications。

The second is to distinguish and evaluate the framework，Including the separation of the regulatory framework of the national security risk of artificial intelligence and the assessment framework of the national security risk of artificial intelligence。The former includes three aspects: the first is Bet365 app download the classification of national security risks of artificial intelligence，Including online national security risks、Data national security risks and algorithm national security risks; followed by the grading of artificial intelligence national security risks，The national security risk grading of the application scenario of the artificial intelligence system，and online national security risk grading and data national security risk grading; Finally, the staging supervision of the national security risk of artificial intelligence，In the process of the development of the artificial intelligence system、Before using it、Risk supervision during use and the response to system vulnerabilities and accident disposal。The latter contains the overall requirements of the national security risk assessment of artificial intelligence and the basic list of feasible assessment。

Third is the framework of the governance system，Including a governance system for national security risks of artificial intelligence、Governance measures and governance mechanisms。The governance system of the national security risk of artificial intelligence includes decision -making and negotiation coordination agencies、The review and regulatory agencies at the central and provincial levels (including standardized and standardized agencies)，and the innovative subject with a safety risk review and reporting obligation (Enterprise、Scientific research institution)。Governance measures include two types of technical and non -technical methods，The governance mechanism contains a passive defense mechanism、Active defense mechanism、Intelligence information mechanism、Counter -offensive mechanism and other levels。

　　(this article is the 2022 ministerial -level legal research topic "Artificial Intelligence National Security Risks and its governance system research" (CLS (2022) C15) phase results)

(Author Unit: Law School of Tianjin University of Commerce)