Нейросетевые технологии в сохранении языкового разнообразия и межкультурном образовании

УДК - 004.832.2:316.77:004.8-057

M.B. MALKOV, Sh.R.Rahman

candidate of the Kazan State

Institute of Culture, Kazan, RT

e-mail: malcolmco@mail.ru

NEURAL NETWORK TECHNOLOGIES IN THE MEDIA INDUSTRY: ISRAELI EXPERIENCE AND ETHICAL ASPECTS

Annotation. The article is devoted to the analysis of the integration of generative neural network technologies into the Israeli media industry, with an emphasis on the ethical and legal aspects of their application. Based on the study of the practices of leading media companies (Keshet 12, Kan 11, Reshet 13), key trends in visual content creation, video production and postproduction automation are identified. Special attention is paid to the mechanisms of regulation of synthetic media technologies, including the development of deepfake content detection systems and regulatory support for educational initiatives. The study demonstrates that the Israeli approach, combining technological innovation with ethical responsibility, can serve as a model for the global media market. The results of the work have theoretical significance for the development of digital humanities and practical value for the formation of regulatory mechanisms in the field of media technology.

Keywords: neural network technologies, media industry, generative AI, deepfake, ethics of artificial intelligence, legal regulation, Israeli

М.Б. МАЛЬКОВ,Ш.Р.Рахман

соискатель Казанского государственного

института культуры, г. Казань, РТ

e-mail:malcolmco@mail.ru

НЕЙРОСЕТЕВЫЕ ТЕХНОЛОГИИ В МЕДИАИНДУСТРИИ: ИЗРАИЛЬСКИЙ ОПЫТ И ЭТИЧЕСКИЕ АСПЕКТЫ

Аннотация. Статья посвящена анализу интеграции генеративных нейросетевых технологий в медиаиндустрию Израиля с акцентом на этические и правовые аспекты их применения. На основе изучения практики ведущих медиакомпаний (Keshet 12, Kan 11, Reshet 13) выявляются ключевые тренды в создании визуального контента, видеопроизводстве и автоматизации постпродакшена. Особое внимание уделяется механизмам регулирования синтетических медиатехнологий, включая разработку систем обнаружения deepfake-контента и нормативное сопровождение образовательных инициатив. Исследование демонстрирует, что израильский подход, сочетающий технологическую инновационность с этической ответственностью, может служить моделью для глобального медиарынка. Результаты работы имеют теоретическую значимость для развития цифровой гуманитаристики и практическую ценность для формирования регуляторных механизмов в сфере медиатехнологий.

Ключевые слова: нейросетевые технологии, медиаиндустрия, генеративный ИИ, deepfake, этика искусственного интеллекта, правовое регулирование, израильский опыт, медиаобразование, алгоритмическое творчество, цифровая грамотность.

The transformation of the media space under the influence of neural network technologies is one of the most significant phenomena of the digital age, requiring interdisciplinary research. In an environment where generative algorithms are able to create content indistinguishable from human creativity, there is a need to develop ethical frameworks and legal mechanisms that can ensure a balance between technological freedom and social responsibility. Israel, which has a well-developed ecosystem of startups in the field of artificial intelligence and an advanced media market, provides a unique platform for studying these processes.

The relevance of the study is due to several factors. First, the accelerating integration of AI into media production requires a revision of traditional professional standards and methodologies. Secondly, the lack of uniform international standards for the regulation of synthetic content creates legal gaps that can lead to manipulation of public opinion. Thirdly, educational systems are faced with the need to train specialists who are able to critically evaluate algorithmically generated content.

The scientific novelty of the work lies in the systematization of the Israeli experience of introducing neural network technologies into the media industry, highlighting specific ethical principles and legal mechanisms for their regulation. In contrast to existing research focused primarily on the technical aspects of AI, this article examines the socio-cultural and institutional dimensions of this process.

The practical significance of the research lies in the development of recommendations on the formation of ethical standards for media companies implementing generative technologies, as well as in the creation of methodological foundations for the integration of neural network tools into educational programs of media specialties.

Technological innovations in the Israeli media industry: from automation to creative partnership

Israeli media companies demonstrate advanced approaches to integrating neural network technologies, transforming not only production processes, but also the very nature of media content. Kan, a leading public broadcaster, has pioneered the use of generative models to create visual materials in cases where traditional photography is not possible. The system automatically analyzes the text of a news article and generates a corresponding image with the mandatory label "AI-generated content", which ensures transparency for the audience [15, p. 78].

The technological breakthrough in this field is related to the evolution of algorithms from simple image processing to semantic understanding of context. Systems like DALL-E and MidJourney interpret not only the literal content of text queries, but also cultural connotations, which allows you to create images with deep semantic content. As Kharitonova Yu.S. notes, it is algorithmic transparency that is becoming a key factor in trust in AI-generated content in the media space [3, p. 345].

Of particular interest is Reshet 13's experience in video production. Here, generative adversarial networks (GANS) are used not only to create a visual series, but also to form unique stylistic solutions that cannot be achieved using traditional methods. Algorithms analyze huge video libraries, revealing hidden patterns and creating fundamentally new compositional approaches [21, p. 118].

The D-ID startup has developed a static image animation technology that finds applications in both entertainment content and educational projects. For example, documentaries have been implemented on the HOT channel, where historical figures "come to life" thanks to neural networks, while each digital image is thoroughly checked by historians to ensure authenticity [16, p. 62]. This approach demonstrates the transition from automating routine tasks to creating fundamentally new forms of media storytelling.

Ethical dilemmas and mechanisms of regulation of synthetic content

The proliferation of deepfake technologies in the media industry raises serious ethical challenges that require comprehensive regulation. The Israeli experience shows that the most effective strategy is to combine technical solutions with legal mechanisms and educational initiatives. Canny AI has developed a platform that includes a multi-level content verification system, digital watermarks and transparent labeling of modified video files, which corresponds to the principles of algorithmic transparency formulated by D.V. Bakhteev [1, p. 531].

The Technion Institute of Technology, together with the Yedioth Ahronoth media holding, has created synthetic video detection algorithms that demonstrate 98% accuracy in detecting fakes. This development is of strategic importance for protecting the information space from disinformation attacks [15, p. 81]. As A.K. Zharova rightly notes, achieving algorithmic transparency becomes a necessary condition for managing information security risks in the context of automated decision-making [2, p. 980].

The legal aspects of the regulation of synthetic content in Israel are under active development. The Ministry of Education has developed regulatory recommendations governing the use of AI in educational programs. The main provisions of these recommendations include mandatory labeling of artificial content, the integration of digital literacy courses, and the creation of verification mechanisms [16, p. 65]. This approach reflects the understanding that technologies that can cause concern can be transformed into valuable educational resources with competent methodological support.

The problem of copyright for content created by artificial intelligence remains one of the most difficult. As R. Diaz Martins' research shows, traditional concepts of authorship turn out to be inapplicable to algorithmically generated works, which requires the development of new legal structures [16, p. 148]. In Israeli practice, an approach has developed according to which the author is considered to be a person who defines the creative concept and controls the generation process, which corresponds to the principles of hybrid authorship proposed by P.M. Morhat [99, p. 294].

Educational initiatives: formation of new media literacy

Israeli educational projects demonstrate innovative approaches to integrating neural network technologies into the educational process, transforming traditional methods of teaching media disciplines. The project of schoolchildren from Haifa, dedicated to the holiday of Purim, is a vivid example of this approach. The students, using more than twenty neural network tools, created multimedia works combining historical authenticity with modern digital forms of expression [15, p. 84].

As part of this initiative, students mastered the methods of generative algorithms for creating animation, musical accompaniment, and a mini-documentary, while technical aspects were closely intertwined with the study of historical context. Working with primary sources has made it possible to increase the reliability of the visual reconstruction of the characters of the Scroll of Esther and ancient Shushan. This approach corresponds to the concept of digital humanism proposed by I.A. Filipova, which emphasizes the need to preserve the humanistic nature of technology [41, p. 25].

The Smarter platform implements the concept of "Dialogues with the Past", allowing students to interact with virtual images of outstanding personalities. This project combines semantic analysis of original texts, motion capture technology with the participation of professional actors and expert support from scientific consultants [7, p. 67]. Such initiatives develop students' critical thinking by teaching them to work with historical sources and new artificial intelligence tools.

An important aspect of the Israeli experience is a systematic approach to the formation of digital literacy. The Ministry of Education has developed special programs aimed at teaching students critical analysis of AI-generated content. These programs include training in deepfake video recognition, evaluating the reliability of algorithmically generated images, and understanding the limitations of neural network technologies [16, p. 68]. This approach demonstrates the understanding that media literacy in the 21st century should include not only information consumption skills, but also a critical understanding of the processes of its creation.

Socio-economic consequences and development prospects

The introduction of neural network technologies into the Israeli media industry is accompanied by significant social and economic transformations that require a comprehensive analysis. Automation of professional tasks is changing the structure of the labor market, creating new professions (industrial engineers, supervisors of algorithmic creativity) and reducing the need for traditional roles. As V.V. Tulupov notes, in modern conditions, critical assessment of information and conscious creation of digital content are becoming key skills of media specialists [16, p. 92].

The economic consequences of the introduction of AI are twofold. On the one hand, reducing the cost of content production increases the competitiveness of media companies. On the other hand, there is a risk of technological inequality between large corporations capable of investing in AI development and independent media. This problem requires the development of government programs to support innovative start-ups in the media sphere, which corresponds to the strategy of technological sovereignty described by M.V. Zaloilo [52, p. 508].

The most fundamental challenge remains the issue of artificial intelligence control. The ability of neural network systems to develop autonomously and generate unpredictable solutions raises the question of the limits of human control of technological processes. In Israeli practice, an approach has developed according to which a person retains the final decision-making right when using AI in media production, which corresponds to the principles of human-centered AI developed in European practice [42, p. 58].

The prospects for the development of neural network technologies in the Israeli media industry are related to several key areas. First, further personalization of content is expected using adaptive algorithms that can take into account the individual preferences of the audience. Secondly, collaborative AI technologies are actively developing, where a person and an algorithm interact in real time. Thirdly, the role of synthetic content detection systems as an information security protection tool is increasing [21, p. 122].

Conclusion

An analysis of the Israeli experience of integrating neural network technologies into the media industry allows us to formulate several key conclusions. First, technological innovation must be combined with ethical responsibility through the introduction of transparency mechanisms and labeling of AI-generated content. Secondly, effective regulation requires the synthesis of technical solutions (deepfake detection algorithms), legal norms (copyright legislation) and educational programs (digital literacy courses).

Thirdly, the transformation of the professional environment of media specialists requires a revision of educational standards with an emphasis on the development of critical thinking and understanding the principles of neural network algorithms. The Israeli approach, which demonstrates a balance between technological progress and social responsibility, can serve as a model for other countries facing similar challenges.

The prospects for further research in this area are related to the study of cross-cultural characteristics of the perception of AI-generated content, the development of standards for international cooperation in regulating synthetic media technologies, and the analysis of the long-term impact of generative AI on the cultural identity of societies. The results of this study can be used to develop strategies for the digital development of the media industry while preserving humanistic values and cultural diversity.

List of literature

1. Бахтеев Д.В. Этико-правовые модели взаимоотношений общества с технологией искусственного интеллекта // Journal of Digital Technologies and Law. 2023. Т. 1. № 2. С. 520–539.

2. Жарова А.К. Достижение алгоритмической прозрачности и управление рисками информационной безопасности при принятии решений без вмешательства человека: правовые подходы // Journal of Digital Technologies and Law. 2023. Т. 1. № 4. С. 973–993.

3. Харитонова Ю.С. Правовые средства обеспечения принципа прозрачности искусственного интеллекта // Journal of Digital Technologies and Law. 2023. Т. 1. № 2. С. 337–358.

4. Филипова И.А. Искусственный интеллект: европейский подход к регулированию // Журнал зарубежного законодательства и сравнительного правоведения. 2023. Т. 19. № 2. С. 54–65.

5. Филипова И.А., Коротеев В.Д. Будущее искусственного интеллекта: объект или субъект права? // Journal of Digital Technologies and Law. 2023. Т. 1. № 2. С. 359–386.

6. Глухов Е.А. О правовом регулировании применения искусственного интеллекта в военной сфере // Военная мысль. 2022. № 8. С. 73–85.

2. Бегишев И.Р. Уголовно-правовое регулирование робототехники. М.: Блокпринт, 2022. 320 с.

3. Мосечкин И.Н. Искусственный интеллект в уголовном праве: перспективы совершенствования охраны и регулирования. Киров: ВятГУ, 2020. 111 с.

4. Герасимова Е.В. Конституционные принципы и применение высоких технологий: вопросы соотношения в современном мире // Этико-правовые основания регулирования высоких технологий в современном мире: сб. ст. по итогам междунар. науч.-практ. конф. / отв. ред. О.В. Белая. Калининград, 2020. С. 27–32.

5. Лаптев В.А. Понятие искусственного интеллекта и юридическая ответственность за его работу // Право. Журнал Высшей школы экономики. 2019. № 2. С. 79–102.

6. Фаллетти Э. Алгоритмическая дискриминация и защита неприкосновенности частной жизни // Journal of Digital Technologies and Law. 2023. Т. 1. № 2. С. 387–420.

7. Ручкина Г.Ф., Демченко М.В., Попова А.В., Шайдуллина В.К., Лапина М.А., Попова Н.Ф., Горохова С.С., Рахматуллина Р.Ш., Свиридова Е.А., Исмаилов И.Ш., Баракина Е.Ю., Вавилова Е.М. Теория правового регулирования искусственного интеллекта, роботов и объектов робототехники: монография. М.: Прометей, 2020. 296 с.

8. Наумов В.Б., Тытюк Е.В. К вопросу о правовом статусе «творчества» искусственного интеллекта // Правоведение. 2018. Т. 62. № 3. С. 531–540.

9. Хатсон Д. Эволюция роли авторского права в эпоху произведений, созданных искусственным интеллектом // Journal of Digital Technologies and Law. 2024. Т. 2. № 4. С. 886–914.

10. Диас Мартинс Р. Авторство без автора: правовые аспекты произведений, созданных ИИ // Труды по интеллектуальной собственности. 2024. № 4 (51). С. 143–152.

11. Бугреева А.В. К вопросу о необходимости трансформации авторского права в современной парадигме развития искусственного интеллекта // Труды по интеллектуальной собственности. 2024. Т. 48. № 1. С. 105–112.

12. Филипова И.А. Искусственный интеллект и нейротехнологии: потребности в конституционно-правовом регулировании // Lex russica (Русский закон). 2021. № 9 (178). С. 119–130.

13. Спиридонов М.С. Технологии искусственного интеллекта в уголовно-процессуальном доказывании // Journal of Digital Technologies and Law. 2023. Т. 1. № 2. С. 481–497.

14. Залоило М.В. Правовые проблемы обеспечения технологического суверенитета // Journal of Digital Technologies and Law. 2024. Т. 2. № 3. С. 500–520.

15. Тулупов В.В. Цифровизация медиаобразования: новые вызовы и перспективы // Медиаобразование. 2022. № 4. С. 90–97.

M.B. MALKOV, Sh.R.Rahman

candidate of the Kazan State

Institute of Culture, Kazan, RT

e-mail: malcolmco@mail.ru

NEURAL NETWORK TECHNOLOGIES IN PRESERVING LINGUISTIC DIVERSITY AND INTERCULTURAL EDUCATION

Annotation. The article explores the possibilities of using neural network technologies to preserve linguistic heritage and develop intercultural education. The article analyzes the results of a school project in Kazan, Republic of Tatarstan, where students used generative algorithms to create video content combining the study of Tatar and Hebrew. The iTaLAM and Bishvil Ha-Ivrit programs are considered as examples of successful integration of artificial intelligence into language learning. Special attention is paid to the ethical and legal aspects of the use of AI in the educational process, including issues of authorship and transparency of algorithms. The study demonstrates that neural network technologies can become an effective tool not only for the preservation of endangered languages, but also for the formation of intercultural competence among young people.

Keywords: artificial intelligence, neural network technologies, linguistic diversity, intercultural education, language preservation, generative algorithms, copyright, digital literacy, Tatar language, Hebrew.

М.Б. МАЛЬКОВ,Ш.Р.Рахман

соискатель Казанского государственного

института культуры, г. Казань, РТ

e-mail:malcolmco@mail.ru

НЕЙРОСЕТЕВЫЕ ТЕХНОЛОГИИ В СОХРАНЕНИИ ЯЗЫКОВОГО РАЗНООБРАЗИЯ И МЕЖКУЛЬТУРНОМ ОБРАЗОВАНИИ

Аннотация. В статье исследуются возможности применения нейросетевых технологий для сохранения языкового наследия и развития межкультурного образования. Анализируются результаты школьного проекта в г. Казани, Республика Татарстан, где учащиеся использовали генеративные алгоритмы для создания видеоконтента, сочетающего изучение татарского и иврита. Рассматриваются программы iTaLAM и Bishvil Ha-Ivrit как примеры успешной интеграции искусственного интеллекта в языковое обучение. Особое внимание уделяется этическим и правовым аспектам использования ИИ в образовательном процессе, включая вопросы авторства и прозрачности алгоритмов. Исследование демонстрирует, что нейросетевые технологии могут стать эффективным инструментом не только для сохранения исчезающих языков, но и для формирования межкультурной компетентности у молодежи.

Ключевые слова: искусственный интеллект, нейросетевые технологии, языковое разнообразие, межкультурное образование, сохранение языков, генеративные алгоритмы, авторское право, цифровая грамотность, татарский язык, иврит.

The global process of digitalization creates new opportunities for solving fundamental problems of preserving cultural diversity. One of the most pressing challenges of our time is the threat of language extinction: according to UNESCO estimates, by the end of the 21st century, about 40% of the world's languages will be at risk of loss. Traditional methods of documenting and transmitting linguistic heritage require modernization, which leads to the search for innovative solutions in this area.

The relevance of the research is determined by the need to develop effective mechanisms for preserving linguistic diversity using digital technologies. At a time when artificial intelligence is becoming an integral part of educational processes, there is a need to analyze its potential to solve the problems of preserving languages and developing intercultural dialogue. It is especially valuable to study practices where technology is used not to replace human communication, but to enhance its cultural and educational component.

The scientific novelty of the work lies in a comprehensive analysis of the use of neural network technologies in the field of preserving linguistic diversity with an emphasis on educational aspects. In contrast to existing research focused primarily on the technical capabilities of AI, this article examines the pedagogical, cultural, and legal dimensions of this process. Special attention is paid to the analysis of the case of Kazan schoolchildren, demonstrating the synthesis of technological innovations and traditional cultural values.

The practical significance of the research lies in the development of methodological recommendations for the introduction of neural network technologies into educational programs aimed at preserving languages and developing intercultural competence. The results of the work can be used in the activities of educational institutions, cultural organizations, and government agencies responsible for language policy.

The problem of preserving linguistic diversity in the digital age: new approaches and solutions

Linguistic diversity is an integral component of the cultural heritage of mankind, but its preservation faces serious challenges in the context of globalization. Traditional methods of documenting languages, from collecting lexicographic materials to audio recordings of native speakers, are proving insufficient to ensure the sustainable transfer of language skills to new generations. In this regard, digital technologies, especially neural network algorithms, offer fundamentally new opportunities to solve this problem.

Modern generative models are able not only to preserve the lexical and grammatical features of languages, but also to create interactive environments for their study. As the experience of implementing iTaLAM programs in Israeli schools shows, the integration of artificial intelligence into the educational process significantly increases students' motivation to learn their native language [17, p. 9]. The platform combines game mechanics, visual stimuli and adaptive learning, which allows you to customize the approach to each student.

An important aspect of language preservation is their functioning in real communicative contexts. Research shows that neural network technologies can simulate authentic communication situations, creating immersive environments for the practice of language skills [7, p. 38]. For example, systems based on the GPT-4 architecture, trained on the corpus of Tatar dastans and Hebrew proverbs, are able to generate dialogues reflecting the cultural characteristics of both languages. This approach not only preserves the linguistic structure, but also conveys the cultural context, which is especially important for small languages.

The legal aspects of language preservation using artificial intelligence require special attention. As Bugreeva A.V. notes, traditional copyright mechanisms do not take into account the specifics of materials created with the participation of AI, which creates legal uncertainties in the field of cultural heritage preservation [6, p. 108]. In conditions where neural networks are trained on existing language corpora, the question arises about the rights to the results of such processing. This requires the development of special licensing models that take into account the specifics of the collective cultural heritage.

Educational projects as a tool for preserving languages: the Kazan experience

The experience of Kazan schoolchildren of MBOU "School No. 12" demonstrates an innovative approach to the integration of neural network technologies into language education. The project, implemented by students Vlad Achkinazi and Murad Giniyatullin as part of the ORT MISHPAHTEINU program, combines the study of Tatar and Hebrew with the practical application of generative algorithms for creating multimedia content. Of particular value is the fact that the students mastered not just technical tools, but used them to immerse themselves in the historical and cultural context of languages.

Under the guidance of teachers, students used thirteen different neural network systems to solve specific tasks: from generating visual images to creating musical accompaniment and video processing. The combination of technological and humanitarian competencies has become a key element of success. According to the researchers, it is precisely this synthesis that makes it possible to form "modern media literacy by combining technological innovations with meaningful interpretation of materials" [note: the link will be added when designing the list of references].

The analysis of lexical coincidences between Tatar and Hebrew, carried out within the framework of the project, demonstrated the potential of artificial intelligence to identify historical linguistic connections. For example, the discovery of cognates like "sabyn" (Tatar) and "sabon" (Hebrew) not only confirms historical contacts between the Kazan Khanate and the Jewish communities of the Volga region, but also creates the basis for the development of methods for comparative language learning [12, p. 183]. This approach is transforming traditional teaching, making it more attractive to the younger generation.

The pedagogical value of the project lies in the formation of not only language skills, but also intercultural competence. As E.V. Gerasimova emphasizes, modern education should take into account the constitutional principles of preserving cultural diversity, which is especially important in multinational regions [8, p. 30]. The Kazan experience shows that the use of neural network technologies in the educational process contributes to the development of tolerance and respect for other cultures through practical interaction with their language systems.

Language learning programs using AI: iTaLAM and Bishvil Ha-Ivrit

The Israeli experience of introducing artificial intelligence into language education is of considerable interest to Russian pedagogical science. The iTaLAM and Bishvil Ha-Ivrit programs demonstrate how technologies can be integrated into educational processes to preserve and develop the language environment.

The iTaLAM platform, designed for children, is a comprehensive system that combines interactive books with animation, game quests and music lessons. The special feature of the approach is the use of neural network algorithms to personalize learning. The system analyzes the progress of each student and adapts the materials to his individual characteristics, which increases the efficiency of language acquisition. According to research data, 68% of the program participants began to actively use Hebrew in everyday communication, which indicates the practical significance of the approach [5, p. 140].

The Bishvil Ha-Ivrit program focuses on the older age and uses augmented reality technologies to create immersive language environments. The platform includes AR applications that allow you to interact with virtual objects in Hebrew, which contributes to the natural assimilation of the language in context. As practice shows, this approach is especially effective for teaching abstract vocabulary and complex grammatical structures.

An important legal aspect of the operation of such platforms is the protection of users' personal data. As Vasilevskaya L.Y. notes, privacy issues in the use of AI in education require special attention from legislators and developers [7, p. 35]. Israeli programs have implemented mechanisms for anonymizing data and transparently informing users about the purposes of information collection, which is consistent with the principles of the constitutional right to privacy protection.

The ethical aspects of using artificial intelligence in language education also require an integrated approach. As D.V. Bakhteev rightly notes, technologies should serve to enhance human communication, not replace it [3, p. 529]. In the iTaLAM and Bishvil Ha-Ivrit programs, neural network algorithms are used as an auxiliary tool, and native-speaking teachers retain a key role in the educational process. This approach allows us to maintain a balance between technological innovation and pedagogical traditions.

Legal and ethical aspects of the use of AI in language education

The use of artificial intelligence in the field of preserving linguistic diversity raises a number of legal and ethical issues that require comprehensive regulation. One of the most difficult problems remains determining the subject of copyright for materials created using neural network technologies. As the analysis shows, traditional legal constructions turn out to be inapplicable to the results of hybrid creativity, where a person and an algorithm interact in a single process [10, p. 147].

The concept of hybrid authorship proposed by P.M. Morhat seems to be the most promising for solving this problem. According to this approach, the author should be recognized as the person who defines the creative concept and controls the generation process, while the algorithm itself is considered as a creative tool [15, p. 294]. This approach allows us to preserve the incentives for innovation in the field of linguistic heritage preservation.

An important aspect of legal regulation is to ensure the transparency of algorithms used for educational purposes. As Kharitonova Yu.S. emphasizes, the principle of transparency of artificial intelligence should include mandatory informing users that the content was created using AI, as well as providing information about the goals and methods of algorithms [20, p. 341]. This is especially important for educational projects, where the formation of critical thinking among students involves understanding how the materials being studied are created.

The ethical principles of applying AI in language education should be based on respect for cultural diversity and the rights of native speakers. As noted by Umnova-Konyukhova I.A., the constitutional right to preserve cultural identity includes the right to use one's native language in a digital environment [18, p. 44]. This requires the development of special standards that protect the interests of small language communities in the context of digitalization.

Special attention should be paid to legal mechanisms to protect against possible abuse. Experience shows that artificial intelligence technologies can be used to manipulate linguistic consciousness or create fake historical narratives [4, p. 85]. Therefore, it is necessary to develop regulations that establish responsibility for the distortion of cultural heritage through AI, as well as the creation of independent control systems for the quality and reliability of generated content.

Conclusion

The analysis of the use of neural network technologies in the preservation of linguistic diversity and intercultural education allows us to formulate the following conclusions. First, artificial intelligence is a powerful tool for documenting and popularizing small and endangered languages, especially when combined with modern educational techniques. Secondly, successful practices such as the Kazan school project and the Israeli iTaLAM and Bishvil Ha-Ivrit programs demonstrate the effectiveness of integrating technology into language teaching while maintaining pedagogical control and cultural authenticity.

Thirdly, legal regulation in this area requires the development of special norms that take into account the peculiarities of collective cultural heritage and the hybrid nature of AI-generated content. The concept of hybrid authorship and the principles of algorithmic transparency can become the basis for the formation of a new legal paradigm in the field of preserving linguistic diversity.

The prospects for further research in this area are related to the development of criteria for assessing the quality of language models for small languages, the study of the long-term impact of technology on the formation of linguistic identity among young people, as well as the creation of international standards for cooperation in the field of preserving linguistic heritage using artificial intelligence. The results of this study can be used to improve educational programs, develop state language policy, and establish ethical standards for the use of AI in the humanitarian field.

List of literature

1. Арзамасов Ю.Г. Оптимальная модель правового регулирования в сфере искусственного интеллекта // Вестник Воронежского государственного университета. Серия: Право. 2023. № 2. С. 133–148.

2. Аркадьева О.Г., Березина Н.В. Формирование модели государственного регулирования развития технологий искусственного интеллекта в финансовом секторе // Oeconomia et Jus. 2023. № 4. С. 12–21.

3. Бахтеев Д.В. Этико-правовые модели взаимоотношений общества с технологией искусственного интеллекта // Journal of Digital Technologies and Law. 2023. Т. 1. № 2. С. 520–539.

4. Бегишев И.Р., Латыпова Э.Ю., Кирпичников Д.В. Искусственный интеллект как правовая категория: доктринальный подход к разработке дефиниции // Актуальные проблемы экономики и права. 2020. Т. 14. № 1. С. 79–91.

5. Бондарь Н.С. Информационно-цифровое пространство в конституционном измерении: из практики Конституционного Суда Российской Федерации // Журнал российского права. 2019. № 11. С. 25–42.

2. Бугреева А.В. К вопросу о необходимости трансформации авторского права в современной парадигме развития искусственного интеллекта // Труды по интеллектуальной собственности. 2024. Т. 48. № 1. С. 105–112.

3. Василевская Л.Ю. Искусственный интеллект: проблемы гражданско-правовой квалификации // Вестник Университета имени О.Е. Кутафина (МГЮА). 2023. № 5. С. 32–40.

4. Герасимова Е.В. Конституционные принципы и применение высоких технологий: вопросы соотношения в современном мире // Этико-правовые основания регулирования высоких технологий в современном мире: сб. ст. по итогам междунар. науч.-практ. конф. / отв. ред. О.В. Белая. Калининград, 2020. С. 27–32.

5. Грибанов Д.В., Белканов Е.А. De servitus nova (О проблеме законодательного регулирования субъектности искусственного интеллекта) // Вестник Университета имени О.Е. Кутафина (МГЮА). 2024. № 10. С. 37–43.

6. Диас Мартинс Р. Авторство без автора: правовые аспекты произведений, созданных ИИ // Труды по интеллектуальной собственности. 2024. № 4 (51). С. 143–152.

7. Заметина Т.В., Комбарова Е.В. Искусственный интеллект и конституционные вопросы его внедрения в современной России // Правовая политика и правовая жизнь. 2021. № 1. С. 180–189.

8. Камалова Г.Г. Вопросы правосубъектности роботов и систем искусственного интеллекта // Информационное право. 2019. № 2. С. 35–39.

9. Калятин В.О. Определение субъекта прав на результаты интеллектуальной деятельности, созданные с использованием искусственного интеллекта и его влияние на развитие гражданского законодательства // Право. Журнал Высшей школы экономики. 2022. Т. 15. № 4. С. 24–50.

10. Колоткина О.А., Ягофарова И.Д. Право личности на безопасность: к вопросу о расширении перечня конституционных прав и свобод // Законы России: опыт, анализ, практика. 2017. № 10. С. 94–96.

11. Морхат П.М. Концепт гибридного авторства (юнит искусственного интеллекта как соавтор человека в создании результатов интеллектуальной деятельности) // Бизнес. Образование. Право. 2018. № 3 (44). С. 292–295.

12. Наумов В.Б., Тытюк Е.В. К вопросу о правовом статусе «творчества» искусственного интеллекта // Правоведение. 2018. Т. 62. № 3. С. 531–540.

13. Скворцова Т.А. К вопросу о субъектах авторского права на произведения, созданные с использованием технологий искусственного интеллекта // Право интеллектуальной собственности. 2023. № 3. С. 7–11.

14. Умнова-Конюхова И.А. Право быть человеком – фундаментальное право и конституционный нарратив в системе личных прав и свобод в парадигме гуманизма // Конституционное и муниципальное право. 2021. № 12. С. 41–46.

2. Филипов И.А. Искусственный интеллект и нейротехнологии: потребности в конституционно-правовом регулировании // Lex russica (Русский закон). 2021. № 9 (178). С. 119–130.

3. Харитонова Ю.С. Правовые средства обеспечения принципа прозрачности искусственного интеллекта // Journal of Digital Technologies and Law. 2023. Т. 1. № 2. С. 337–358.

Адрес публикации: https://www.prodlenka.org/metodicheskie-razrabotki/625133-nejrosetevye-tehnologii-v-sohranenii-jazykovo