2. Algorithmic Risks in IPE in Colleges and Universities from the Perspective of Information Cocoons
Algorithmic recommendation refers to the process of delivering personalized content to users by analyzing data such as their interests, behaviors, and preferences through algorithms. According to Cass Sunstein, an "information cocoon" is "a communications universe that we have constructed for ourselves, where we listen only to what we have chosen and what pleases us"
| [3] | Sunstein K R. Infotopia: How Many Minds Produce Knowledge [M]. Translated by Bi Jingyue. Beijing: Law Press, 2008: 8. |
[3]
. Algorithmic recommendation is a product of the intelligent media era. With the rapid development of algorithmic recommendation technologies, the "information cocoon" effect has become increasingly pronounced. When students are exposed over the long term to singular, one-sided, or preference-aligned viewpoints, their opportunities to access diverse information and perspectives are restricted. This limits their comprehensive understanding of mainstream ideologies, pluralistic values, and complex social realities, thereby forming an "IPD cocoon."
2.1. The Risk of Ideological and Value Deviation
The design logic of algorithms, the data used for their training, and the ultimate objectives of commercial platforms render them particularly susceptible to serving as covert vehicles for ideological infiltration. Leveraging advantages such as ample funding, institutional flexibility, and technological precedence, certain Western capitalist countries exploit the algorithmic "black box" to engage in cultural export and value presupposition. A significant amount of persuasive and inflammatory content, which glorifies Western values, is disseminated in cyberspace. This may subtly deconstruct or dilute the core socialist values and collectivist spirit that IPE strives to convey. More critically, many users, unknowingly and subconsciously, internalize the worldviews and value orientations of the bourgeoisie. Over time, this can foster a sense of alienation from the correct values championed by China's mainstream ideology, leading to a crisis of identification with the discourse of IPE.
Furthermore, to enhance user engagement and maximize traffic, algorithmic recommendations prioritize datafied user preferences. They tend to promote entertaining or even vulgar topics that are more likely to attract attention and clicks. This leads to an increasing weighting of homogeneous information within users' "recommendation pools." By constructing closed information systems based on user profiles, algorithmic recommendation technologies create an environment for college students where they are predominantly exposed to content aligning with their interests. Information that does not align with these interests is automatically reduced or filtered out. This insidiously exacerbates information homogeneity, gradually shaping fixed mindsets and limiting students' opportunities to encounter diverse perspectives. Consequently, the content of IPE struggles to penetrate these value-based barriers and enter the "echo chamber." This results in the solidification and encapsulation of students' values. Content that is profound, intellectually stimulating, and aligned with mainstream values—often characterized by a more formal and serious mode of expression—appears incongruous within a cyberspace saturated with humorous and exaggerated discourse. Such content suffers from low visibility and may be "neglected" by algorithms. As a result, the influence and guiding power of information dissemination are continually diminished, leading to an insufficient supply of IPE content and hindering users' access to genuinely valuable information.
2.2. The Risk of Deconstructing the Discourse Power of IPE
Whoever masters generative artificial intelligence technology holds the right to speak
| [4] | Ou Tingyu. On Generative Artificial Intelligence Empowering Discourse Practice of Ideological and Political Education in Universities [J]. Journal of Ningxia University (Social Sciences Edition), Jul. 2025, Vol. 47, No. 4: 118. |
[4]
. Teachers are no longer the sole "agenda-setters," as algorithms have emerged as new centers of power. Students are exposed to diverse and often fragmented ideological trends through short videos, challenging the authority of teachers' discourse. Algorithms tend to favor content that is short, fast-paced, highly emotional, and intensely conflict-driven. This creates a significant tension with the logical rigor and seriousness emphasized in IPE courses, potentially leading to the "fragmentation" and dilution of systematic theories. Algorithmic recommendations intensify the "echo chamber effect," confining students within their "information cocoons." This makes it difficult for mainstream external discourse to penetrate their internal circles, rendering the traditional "I speak, you listen" model ineffective and reinforcing the echo chamber phenomenon.
2.3. The Risk of Alienating Educational Subjectivity and Interpersonal Relationships
Teachers' leading role and the "embodied interaction" between teachers and students risk being diluted. Over-reliance on various intelligent educational platforms that use algorithms for learning analytics, content recommendation, and outcome assessment may reduce the educational process to mere data computation. The teacher's pedagogical role could be diminished to that of a content "reviewer" or "distributor." Meanwhile, the information students receive becomes highly dependent on algorithmic preferences, undermining teachers' comprehensive judgment and pedagogical leadership, which should be based on experience, emotion, and wisdom. Students are reduced to a series of behavioral data points (clicks, dwell time, quiz answers), where their complex ideological dynamics, emotional confusion, and value struggles may be simplified and misinterpreted. This hinders their development as active agents capable of critical thinking and autonomous value construction. Furthermore, the increase in human-computer interaction may partially replace the direct, sincere, and emotionally resonant intellectual exchanges between teachers and students. The crucial elements of IPE—"teaching by precept and example," "emotional resonance," and "personality influence"—are filtered out in virtual interfaces, compromising the depth and impact of the educational outcome.
2.4. The Risk of Technological Ethics and Data Security
Complex deep learning models often function as "black boxes," making their decision-making logic difficult to interpret. First, algorithm designers frequently do not disclose their algorithmic principles and mechanisms, leaving most users unable to understand the logic and process of data collection, filtering, and processing. This lack of transparency leads to misconceptions about algorithmic operations and makes it difficult for users to assess whether the data sources in recommendation systems are legitimate or whether the computational results are reasonable. In such contexts, users may question the interpretative power and guiding influence of IPE discourse, thereby fostering a crisis of trust in algorithmically enhanced IPE content. Second, the datafication of the learning process may evolve into "panoptic surveillance" of students, tracking attention patterns, social interactions, and even emotional responses. Defining reasonable boundaries for data collection, protecting student privacy, and preventing data misuse constitute significant ethical challenges. Third, if sensitive data concerning the ideological dynamics and political orientations of teachers and students is stored overseas or on commercial platforms, it presents national security risks related to data leakage, tampering, and potential manipulation for ideological purposes.
2.5. The Risk of Formalizing the Educational Process and Outcomes
Using facial recognition and textual analysis for "affective computing" to assess learning engagement may lead students to "perform" in ways that cater to algorithms, producing superficial conformity rather than genuine understanding. Employing such methods to evaluate students' ideological dynamics or course effectiveness may deviate from the fundamental goal of IPE, which is to foster deep internalization of values. Algorithms excel at processing structured, quantifiable problems; however, the shaping of values and the establishment of ideals and beliefs are complex, long-term, and iterative internal processes. Attempting to "solve" ideological problems with a one-click technological solution may result in a superficial and formalized educational process.
3. The Tension and Internal Logic Between Algorithms and IPE
American scholar Neil Postman pointed out, "Every technology is both a burden and a blessing; not either-or, but this-and-that"
| [5] | Zhang Yu, Gao Fujin. New Forms and Risk Governance of Digital Narration of Fine Traditional Chinese Culture [J]. Social Sciences in Xinjiang, 2025(4): 121-129. |
[5]
. The tension between algorithms and IPE represents a profound contest between instrumental rationality and value rationality, and between commercial logic and educational logic. The contemporary mission of IPE is not to reject technology, but to harness it based on a deep understanding of algorithmic logic, ensuring that value rationality guides instrumental rationality, and ultimately achieving an organic integration of "algorithms" and "IPE."
3.1. The Roots of the Tension
The operational logics of the two are fundamentally different. In terms of core drivers, algorithms follow the logic of capital, aiming for user retention and click-through rates in pursuit of commercial efficiency and profit. In contrast, IPE follows the logic of value, fundamentally committed to fostering virtue and nurturing people in pursuit of ideological guidance and value shaping. In terms of information distribution, algorithms deliver personalized content recommendations, reinforcing users’ existing preferences through an "information cocoon" and inadvertently entrenching social echo chambers. IPE, however, emphasizes guidance by mainstream values, helping educatees construct a comprehensive worldview and step out of their personal comfort zones. In their view of "human beings," algorithms see users as datafied consumer subjects—collections of preferences and behaviors. IPE sees students as holistic individuals requiring comprehensive development, encompassing their thoughts, emotions, and beliefs. In terms of value stance, algorithms claim to be "value-neutral," yet the data and models themselves implicitly contain the biases of their designers. IPE, by contrast, possesses distinct ideological attributes and must undertake the mission of upholding mainstream values.
3.2. An In-depth Analysis of the Internal Logic
From the perspective of epistemological logic, algorithms, by reducing "information encounters," create "cognitive narrowing." This contradicts IPE's pursuit of "comprehensive understanding," potentially exposing students only to homogeneous information, fostering simplistic and one-sided interpretations of complex social issues, and weakening their capacity for dialectical thinking. From the perspective of subjectivity logic, algorithms stimulate dopamine secretion through immediate feedback, immersing students in sensory pleasure and leading to an "entertainment-oriented" tendency. IPE, however, requires students to engage in deep thinking, overcome inertia, and pursue lofty values. Algorithmic logic allows "comfort" to triumph over the "arduous" process of self-cultivation, contradicting the goal of comprehensive human development. From the perspective of power logic, algorithms, by shaping the information environment, indirectly influence and even alter people's ideas and behaviors, forming an intangible "technological power." Such power is decentralized, concealed and underregulated. It may compete with IPE for dominance over ideology, and even undermine the solemnity and authority of the latter. From the perspective of value logic, algorithms amplify emotions and cater to instincts, potentially giving rise to irrational voices and exacerbating societal "value fragmentation." In contrast, IPE aims to consolidate value consensus, promote mainstream values, and facilitate social integration.
3.3. The Possibility of Moving from "Tension" to "Integration"
Despite the existing tensions, algorithms and IPE are not absolutely antithetical. IPE can leverage algorithms to achieve precise and personalized dissemination, while algorithms themselves require value calibration to avoid "technological backlash." The key lies in how to realize an integration that positions "algorithms as the tool and education as the foundation." On one hand, mainstream values should be embedded into the top-level design of algorithms, transforming them from mere commercial tools into vehicles for disseminating positive energy. On the other hand, students' "algorithm literacy" must be enhanced, cultivating their critical thinking skills so that they can actively harness algorithms rather than being shaped by them.
4. Construction of Governance Paths for Algorithmic Risks in IPE in Colleges and Universities
In 2024, General Secretary Xi Jinping emphasized at the National Education Conference: "The great modern education country we aim to build is one with Chinese characteristics and socialism, which should boast strong ideological and political guidance, talent competitiveness, scientific and technological support, livelihood security, social coordination, and international influence."
| [6] | Xi Jinping Emphasizes at the National Education Conference: Focus Closely on the Fundamental Task of Fostering Virtue Through Education and Make Solid Progress Toward the Strategic Goal of Building a Powerful Country in Education [N]. People’s Daily, 11 Sep. 2024. |
[6]
The core of governing algorithmic risks in IPE in colleges and universities lies in taming technological logic, reaffirming the educational orientation, and constructing a new AI-enabled IPE ecosystem characterized by people-oriented, technology for good, clear rights and responsibilities, and safe and controllable.
4.1. Clarify Values & Educational Goals, Uphold Education-led, Technology-enabled Principle
We must prioritize values, take mainstream value orientation as the core,”We should guide algorithms with mainstream values"
| [7] | Xi Jinping. The Governance of China III [M]. Beijing: Foreign Languages Press, 2020: 318. |
[7]
, always adhere to the ideological attribute and educational goals of IPE as the foundation, and promote the integration of IPE into the inherent logic and operation mechanism of intelligent technology. In all links, human judgment and the value of educational laws must take precedence over algorithmic efficiency. In key decision-making links (e.g., ideological dynamics assessment, content recommendation), we adhere to human-in-the-loop (HITL): algorithms only provide references, and final decisions are made by teachers.
First, integrate IPE into the process of information tagging to increase the proportion of mainstream values in the "content pool" of intelligent technology. Take mainstream value orientation as the basis for information tagging; through semantic analysis, sentiment judgment, and keyword clustering of massive online information, screen high-quality information that conforms to mainstream value orientation.
Second, fully reflect and meet the diverse personalized needs of educational objects in data collection and analysis. Accurately grasp the ideological and behavioral dynamics of educational objects through historical records, user tags, search records, and other information. Big data information collection systems, network intelligent analysis centers, and public opinion intelligent supervision projects track educational objects’ posting, liking, sharing, and commenting behaviors in real time, and analyze and predict their personalized information needs in advance.
Third, apply orthodox educational content to scenario-adaptive push. Through topic addition, scope restriction, targeted promotion, and other methods, precisely deliver educational content to knowledge Q&A platforms, audio-visual communities, short-video social platforms, and bullet-screen interactive live-streaming platforms where educational objects gather, realizing full coverage of educational content in the learning, living, and entertainment processes of educational objects.
4.2. Enhance Educators’ "Algorithm Literacy" to Maintain Pedagogical Leadership and Professional Judgment
Marx pointed out, "Ideas can only be realized by human beings who wield practical power"
| [8] | Central Compilation and Translation Bureau of the Works of Marx, Engels, Lenin and Stalin. Collected Works of Marx and Engels: Volume I [M]. Beijing: People’s Publishing House, 2009: 320. |
[8]
. Provide systematic training on "digital literacy" and "algorithm literacy" for ideological and political teachers. This enables them to critically understand, evaluate, and utilize AI tools, actively adapt to technological transformations and meet the demands of the times. By leveraging the advantages of algorithms, they can optimize educational content and methodologies, enhance the appeal and effectiveness of education, and thereby promote the innovative development of IPE.
4.2.1. Break Discourse "Barriers" and Facilitate Discourse Transformation in IPE
On the one hand, set discourse agendas based on the needs of educatees. Leverage algorithm recommendation technologies to collect data on college students’ learning behaviors and interest preferences, construct user profiles, and predict the IPE needs of different student groups. On this basis, precisely set discourse agendas for IPE, covering major national policies, campus cultural activities, and social hot topics. This effectively ensures that educational content aligns with students’ needs and interests, improving the targeted and practical outcomes of IPE.
On the other hand, enhance the interactivity and participation of discourse agendas. Ideological and political educators in colleges and universities should shift away from traditional, repetitive, and abstract discourse modes. By virtue of the technological advantages brought by algorithms, they should present abstract theories and grand political discourses in visualized, interactive, and everyday-oriented forms, making them concrete, livable, and popular. This increases college students’ engagement and interest, and enhances the affinity and contemporary relevance of IPE discourses.
4.2.2. Grasp Educatees’ Ideological Dynamics via Algorithms and Innovate Educational Communication Methods
Through intelligent algorithm recommendation systems, conduct in-depth analysis of students’ emotional tendencies, ideological dynamics, and cognitive patterns. Customize personalized educational content for them to provide targeted ideological clarification and value guidance, promoting the transformation of IPE communication toward a style that is more closely integrated with daily life.
Meanwhile, leverage the advantages of algorithms to adopt diverse communication methods such as short videos and live streaming. Conduct "one-on-one" information push on official accounts and online platforms to ensure that each student can access learning resources best suited to their individual needs. This further improves the attractiveness of educational content and students’ sense of identity.
4.3. Build Positive Multi-discourse Content Matrix to Optimize IPE Discourse Dimension
Marx pointed out: “As long as a theory convinces people, it can seize the masses; and as long as a theory is thorough, it can convince people.”
| [9] | Marx K, Engels F. Selected Works of Marx and Engels: Volume I [M]. Beijing: People’s Publishing House, 2012. |
[9]
For educational content to convince people, it is necessary to give play to the theoretical interpretation functions of political discourse, academic discourse and life discourse.
First, elaborate political discourse with a firm stance. The expressive characteristic of "discourse is power" embodied by ideological discourse power reveals that discourse itself constitutes ideological practice
| [10] | Xu Baoqiang, Yuan Wei. Politics of Language and Translation [M]. Beijing: Central Compilation & Translation Press, 2000. |
[10]
. Educators should clearly and unequivocally expound the content of the mainstream ideology, be good at speaking out, daring to take a clear stand and brave in refuting fallacies on major issues of right and wrong. They should actively integrate the new discourse emerging from the theoretical and practical development of socialism with Chinese characteristics in the new era into the content system of intelligent IPE, so as to enhance the credibility and persuasiveness of educational content with concrete, systematic and profound political discourse.
Second, explain academic discourse with valid reasoning and sufficient evidence. Marx pointed out, "A theory is capable of winning over the masses once it convinces people; and it can convince people only if it is thorough"
| [11] | Marx, Engels. Complete Works of Marx and Engels: Volume 1 [M]. Beijing: People’s Publishing House, 1956. |
[11]
. Marxist theory is a scientific theory validated by practice. Educators should take the initiative to follow the linguistic logic and discourse norms accustomed to educational objects, and use expressions with rigorous logic and substantial arguments to explain clearly, thoroughly and comprehensively the issues that interest or puzzle educational objects.
Third, express life discourse vividly and lively. Educators should focus on integrating theoretical narration with the real-life world, replace rigid and tedious theoretical preaching with vivid life discourse, and convey educational content in forms popular with educational objects, so as to strengthen the affinity and appeal of educational content.
4.4. Improve the Multi-level Legal and Standard System, and Consolidate the Legal Framework for Value Guidance
General Secretary Xi Jinping pointed out at the 12th collective study session of the Political Bureau of the 19th Central Committee of the Communist Party of China: "No form of media, whether online or offline, on large screens or small screens, is beyond the reach of the law or public opinion."
Therefore, in the face of a series of problems arising from the empowerment of IPE by algorithm technology, relevant authorities should fulfill their gatekeeping responsibilities, strengthen supervision, and regulate the operation of algorithms through legal systems.
We will improve laws and regulations on data governance, strictly implement the Data Security Law of the People's Republic of China and the Personal Information Protection Law of the People's Republic of China, and formulate more specific compliance management systems for algorithm application within the existing legal framework, including the disclosure of algorithm principles and the operation mechanisms of algorithmic decision-making, so as to provide a solid legal foundation for the healthy development of algorithm technology. We will formulate measures for the classified and graded management of campus education data, clarify norms for the entire process of collection, storage, use, and destruction of data related to IPE, and ensure data sovereignty and security. We will clearly define the rights, obligations and responsibilities of all parties, including algorithm recommendation service providers, users and regulators, and establish and improve a responsibility accountability system to ensure that all links of algorithm recommendation, such as design, development, implementation and supervision, comply with laws, regulations and ethical standards, and promote the transparency of the algorithm recommendation process. "In the processes of algorithm design, training data selection, model generation and optimization, and service provision, effective measures shall be taken to prevent discrimination based on ethnicity, religious belief, nationality, region, gender, age, occupation, health and other grounds"
| [13] | Interim Measures for the Management of Generative Artificial Intelligence Services [J]. Journal of Computer Products & Circulation, 2024(2): 162-164. |
[13]
. We will establish an ethical review and filing mechanism for core algorithms applied in the field of IPE (especially recommendation and evaluation algorithms), requiring operators to disclose basic principles, data sources and potential biases, and accept assessments from education authorities and supervisory institutions. We will formulate technical application standards, clarify the scenario boundaries, quality requirements and negative lists for technical application.
4.5. Innovate Controllable, Trustworthy, Explainable Algorithms to Build Positive Interactive Value System
Developing "Specialized Algorithms forIPE". Universities, research institutions and qualified domestic enterprises shall be encouraged to collaborate in developing independently controllable specialized algorithm models and platforms based on the Chinese context, mainstream values and the laws of IPE.
First, develop more explainable AI models to enhance transparency. Platforms should be adept at analyzing public opinion dynamics and seizing opportunities for educational guidance. At the early stage of the fermentation of various social hotspots, pain points and difficulties, they should timely guide the direction of topic discussions, take the initiative in agenda-setting, and break the dilemma of "information cocoons".
Second, the construction of intelligent IPE application platforms should give full play to the positive role of mainstream media, enabling them to occupy the commanding heights of communication in public opinion guidance, ideological leadership and cultural inheritance. Through vivid and down-to-earth authoritative interpretations of common topics, high-quality information in line with mainstream value orientations should be actively promoted. In 2023, the new media platforms of People's Daily actively applied intelligent technologies to improve the quality of coverage of the Two Sessions, launching such contents as the national image online publicity film PRC, the integrated media interview column People's Suggestions Direct Train for interaction between cadres and the masses, and the live-action animation video Rabbit Must Be No. 1. These initiatives not only effectively enhanced the public's understanding of mainstream values, but also motivated all sectors of society to actively participate in major theme publicity campaigns, achieving excellent educational and guiding effects.
Third, mobilize platform users to actively participate in platform construction. Encourage users to take part in the development and application of algorithm platforms, enable them to fully express their value demands and social concerns, contribute their usage experience and improvement suggestions, and promote positive communication and interaction between users and platform enterprises, so as to make algorithm platform construction better meet user needs and social requirements.
Fourth, accelerate the in-depth integration and development of media to build an intelligent ideological and political all-media communication platform. General Secretary Xi Jinping pointed out that we should promote the in-depth development of media integration and "accelerate the construction of an integrated all-media communication structure"
| [14] | Xi Jinping. The Governance of China: Volume III [M]. Beijing: Foreign Languages Press, 2020: 318. |
[14]
. At present, the working environment for IPE has become a realistic composite field featuring the deep integration of traditional and emerging media. To break the "information cocoons" in intelligent IPE, we must conform to the trend of in-depth media integration and accelerate the construction of an intelligent ideological and political all-media communication platform.
Deepen theoretical and applied research on the integrated development of intelligent technologies and IPE, and timely apply cutting-edge theoretical research results to educational practice.
4.6. Ecological Dimension: Construct a Collaborative Governance Pattern of Multiple Co-governance
A collaborative governance committee shall be established with the joint participation of education authorities, colleges and universities, technology enterprises, legal experts, ethicists, as well as representatives of teachers and students, to conduct consultation and judgment on major risks and disputes so as to form a joint governance force. Interdisciplinary research (encompassing pedagogy, computer science, ethics and political science) shall be encouraged, and critical academic research strengthened, to continuously track and evaluate the application effects and risks of artificial intelligence in IPE, providing theoretical support for governance. Students’ awareness of digital citizenship shall be cultivated, and contents such as digital ethics, media criticism and algorithmic thinking integrated into IPE, enabling students to become sober, independent and responsible citizens in the digital era with the ability to resist algorithmic manipulation.
Share This Article
Twitter
Linked In
Facebook
Copy
Download
