Chatgpt technology potential political security risk
March 29, 2023 10:01 Source: "Chinese Social Sciences" March 29, 2023 Total 2620 Author: Ji Lei Li Dajiao

At the end of 2022,Openai launched a new generation of artificial intelligence large language model ChatGPT,Become a phenomenon -level application of rapidly breaking 100 million users around the world。Compared to traditional artificial intelligence technology,Big model represented by ChatGPT shows a strong language understanding and interaction ability、Multi -field content generating ability、The ability to evolve rapidly and the prospects of continuous development of weak artificial intelligence towards strong artificial intelligence,attracted strong attention from all walks of life。With the continuous deep integration of the digital world and the physical world in the intelligent era,The continuous evolutionary large model technology deeply embeds domestic and international politics、Economy、The fields of society become the general trend,Thinking about the potential politics and security risks of this high -singing technology is also increasingly urgent。

  bet365 best casino games

The problem of scattered and fragmented in public space in the digital era is increasingly prominent。Big language model is based on Internet data processing and deep learning technology,In the future political process, we can fully summarize information、Determine the target collection、Design reasonable solution, etc.,Perfectly assisted the country to systematically integrate a diverse society,The convenience of strengthening political decision -making、Scientific and inclusive。But,Big model technology enhances national governance Bet365 app download capabilities,It also brings some implicit risks to political decisions。

First, the political decision -making mechanism is reshaped。In -depth with the popularity of the big model and the degree of application,Big models with unparalleled information collection processing capabilities,With advanced algorithms,The decision -making suggestions given after processing information are integrated and convenient to a considerable extent,It will inevitably make certain policy makers rely on large models to make political decisions。In this case,Big models have actually partially undertaked policy suggestions and even political decision -making functions。If policy makers rely on large model technology for a long time without paying attention to the improvement of their own political literacy,So there will be a problem of lack of policy innovation,Then reduce the level of decision -making of the entire country。

Second, political decision -making ignore preferences。Big model is just a technology,When collecting information, it is only for pure collection purposes。Although the large model processs massive data resources,and the model sensitivity will be improved by relying on future training and data filling,But the object of political decision -making is a life body with emotional preferences,The policy they like may not be the most in line with the policy calculation。This has a very difficult problem -the artificial intelligence system may never fully understand the emotional appeal of the political decision -making object。Therefore,The political recommendations or decisions made by the big model based on the risk of inadequate response。

Third is the risk of reduced political trust。After the big model enters political life,Those who are political decision makers or political demands Bet365 lotto review can use this model to engage in political activities。In this case,Due to the high degree of humanization in the input language of large models,It is difficult for political decision makers to judge whether the political appeal received from the real "human" mouth,Political appeal expressions are difficult to determine whether the receiving policy response really originated from the "official",At this time, the two -way trust crisis of political decision makers and demand expressions may occur。

  Public opinion spread facing new challenges

Big models as revolutionary artificial intelligence technology,It will play a significant social authority function,Further enhances the right to speak on public issues,Promote the spread and influence of public opinion,Promote the organizational ability of social groups and social organizations。However,It also brings a lot of challenges to the spread and governance of public opinion in the country and the international community。

First,Error information and viewpoints will subtly guide public opinion。The algorithm and data of the big model are robust、Transparency、Explanable、Lack of reliability,"Big Data+Mass Parameters+Macau" improved the overall anthropomorphic performance of the model,But it is not understood and thought like a person,Sometimes it is unable to engage in the authenticity of the real distinction information,It is difficult to identify the prejudice and toxicity of the hidden data,So as to produce false、misleading or low -quality results。These errors and prejudices often affect people's political judgment and political choice,Even shaping and misleading public opinion,impact mainstream value and concept。

Second,Big language model may weaken the right to interpret the country。Bet365 lotto review Explanation represents the country's guidance of public opinion、Promote the authority of social ideology,It is an important component of the national soft power。As the big model is fully embedded in society and participates in the interpretation of some major concepts and events,Big models due to their own popularity and convenience,It may make the public get used to seeking a big model explanation path for the first time。This preconceived as the main superposition of the model output system is highly anthropomorphic and complete,It will seriously weaken the persuasiveness of the official explanation,Weaken the authority explained by the country。

third,Big Language Model is reduced to ideological weapons。Big model technology in computing power、Algorithm、Data and other high thresholds can easily lead to technical monopoly in a few countries,Tools that may be regarded as ideological war or color revolution。One country can provide content products with political purposes and ideological preferences for the public's psychological weakness for the public's public psychological weakness,Then affect the political judgment and political selection of the target country's public,Indirect control of political direction,Invisible penetration of the target national ideological security position。

  Technical supervision faces new difficulties

Big language model is widely used,Data pressure that the country needs to deal with is geometric growth,And with the high degree of humanization of the output language,It is extremely difficult to distinguish the authenticity of this information。The right to speak in society with a technology company that master the big language model technology is getting greater and greater,Makes the country's difficulty in digital governance and technical supervision bet365 Play online games in the era of big models。

First,There is a gap between regulatory needs and supervision capabilities。With the gradual development of large model technology,Discourse expressions that have an output error or have obscure ideological propaganda,It is difficult to identify by non -professional people in most cases。,When large models such as CHATGPT participate in public affairs in the form of network waters,and deliberately output the wrong facts of the output of a certain political event,The mass of the huge data will cause a lot of regulatory pressure。If the government department cannot properly supervise the network environment in time,Differential processing of error information,This kind of error message widely spread among the public due to regulatory gaps may cause serious ideological risks。

Next,Technical monopoly erosion national governance capabilities。Creation of Big Language Model Technology、Development and popularization are inseparable from professional technology companies,Especially technology giants。These enterprises participate in the implementation of national governance goals by commissioning governance。But the algorithm necessary for the big language model、Computing Power、Data three major numbers are mastered by a few head technology companies,Expansion of the scale effect,These companies will inevitably form a technological monopoly trend。The power and operational efficiency of the oligopoly technology giant dependency algorithm,Vague power boundary of the country,Even replacing some of the country's functions,Bringing the danger of "empty nationalization"。

  Deeply affects the international system of the turbulent change period

Intelligent era international science and technology competition is increasingly bet365 Play online games fierce,Scientific and technological level and innovation ability are the core elements of national power。Big model technology as an epoch -making innovation as an artificial intelligence field,Its development and diffusion have a profound impact on the international system that has entered a new turbulent change period。

On the one hand,Large model technology is likely to strengthen science and technology political polarization confrontation。The high data volume and technical reserve characteristics required for the large model architecture determine the trend of its existence of being monopolized by a science and technology country。Under the fierce technical competition now,Technical powers usually choose to protect existing technology to avoid opponent countries,The technical weak country that is unable to participate in the large model architecture can only be forced to choose the side station。In recent years, some developed countries have been expressed by examples of restricting chips such as technology and industrial development through limiting chips.,The competitiveness of international technology and politics has further intensified,The possibility of international order falls into the bipolar confrontation。

On the other hand,Large model technology may promote the formation of values ​​alliance。countries that master the big model technology are basically developed countries,This technology hegemony has exacerbated the technology dependence of developing countries to developed countries。At the same time,Due to the significance of the major model technology,Make developed countries further increased the spread of its own value superiority,strengthened the hidden dominance of Western values,It makes it easier for developed bet365 best casino games countries to use large model tools to promote the formation of value alliances。

 (Author Unit: School of Government Management, East China University of Political Science and Law)

Editor in charge: Chen Jing
QR code icon 2.jpg
Key recommendation
The latest article
Graphics
bet365 live casino games

Friendship link:

Website filing number: Jinggong.com Anxie 11010502030146 Ministry of Industry and Information Technology:

All rights reserved by China Social Sciences Magazine shall not be reprinted and used without permission

General Editor Email: zzszbj@126.com This website contact information: 010-85886809 Address: Building 1, Building, No. 15, Guanghua Road, Chaoyang District, Beijing: 100026