Group 1: Session report on the IPILM-conference

Introduction

The presentation explored the impact of AI on local culture, focusing on the aspect of language. It provided an overview of Natural Language Processing (NLP) and its applications in language technology, including machine translation, speech synthesis and speech recognition.

In the presentation, the underrepresentation of minority languages and the dependence on English-language texts were highlighted. The focus of the presentation was placed on the role of Artificial Intelligence in language preservation and documentation, particularly for endangered languages like Cherokee. In addition to technological advances, challenges such as ethical considerations and cultural biases were addressed.

The case study on the Cherokee language showed practical approaches to language revitalization and the importance of balancing technological innovation with cultural authenticity.

Discussion

The discussion addressed critical questions about making AI tools accessible to underrepresented communities, beyond researchers and developers. The group emphasized the need for user-friendly platforms that cater to different levels of technological expertise. Efforts should also be made to make these tools available in different languages, particularly for communities without advanced digital infrastructure.

Next, the participants explored how technology aids in documenting and revitalizing lesser-known languages. The group emphasized that AI, especially natural language processing (NLP), was important for documenting and revitalizing lesser-known languages by analyzing linguistic data to create digital archives, educational resources, and language learning apps.

A specific question focused on the role of the Cherokee community in developing AI tools for their language. In our opinion, the Cherokee community should have led the development of AI tools to preserve the cultural and linguistic integrity of their language, ensuring the decolonization of research.

Another important question delved into whether AI efforts in language preservation also represented different dialects and regional variations, to avoid promoting a homogenized culture. AI efforts in language preservation should recognize and represent regional dialects and variations to avoid creating a homogenized culture. This can be achieved by training AI models on diverse linguistic data that includes various dialects, regional accents, and cultural nuances.

Further information

If you are interested in learning more about Cherokee, you can visit the Cherokee nation website and watch the following video:

Aktivieren Sie JavaScript um das Video zu sehen.
https://www.youtube.com/watch?v=PUYoRT2EA5Q&t=26s

Further Readings

Zhang, Kexun; Choi, Yee; Song, Zhenqiao; He, Taiqi; Wang, William; Li, Lei. (2024): Hire a Linguist!: Learning Endangered Languages in LLMs with In-Context Linguistic Descriptions, 15654-15669.

This study explores how the innovative LINGOLLM project enables large language models (LLMs) to work with endangered languages by incorporating linguistic descriptions such as dictionaries and grammar books. This approach enhances AI’s translation and processing capabilities for rare and endangered languages.

---

Zaki, Muhammad Zayyanu; Ahmed, Umar (2024): Bridging linguistic divides: The impact of ai-powered translation systems on communication equity and inclusion. In: Journal of translation and language studies (5.2), 20-30.

This article explores how AI-powered translation systems influence communication equity and inclusion, addressing challenges such as linguistic bias and accessibility for marginalized communities.

---

Zhang, Shiyue; Frey, Ben; Bansal, Mohit (2022): How can NLP Help Revitalize Endangered Languages? A Case Study and Roadmap for the Cherokee Language. In: Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics, 1529- 1541.

The study explores how Natural Language Processing (NLP) can aid in revitalizing endangered languages, with a focus on Cherokee. It examines the language’s current state, resources, and challenges, and suggests strategies for collaboration with indigenous communities.

---

Our presentation

Below, you’ll find the screencast of the presentation on “AI Impact on Culture: Language”:

Aktivieren Sie JavaScript um das Video zu sehen.
https://www.youtube.com/watch?v=CyQKql5qjy0

Introduction

The conference session of group 2 focused on AI impact on local culture: Cultural perception of AI regulations and how AI influences cultural values and ethics focusing on regulatory and governance challenges. It focuses on the challenges of regulating AI and adapting governance frameworks to align with local cultural perceptions. An overview of the governance challenges posed by the rapid evolution of AI, as well as its ethical and regulatory implications, was presented. Moreover, it emphasizes that AI systems must be developed and deployed with principles like transparency, accountability, and fairness to address these challenges effectively.

Insights from a survey conducted across Germany, Poland, Bosnia, and India which illustrated the complex relationship between AI’s global development and its local cultural impacts. This underscores the importance of integrating cultural values into regulatory frameworks.

Discussion: Question and answers

The discussion addressed several questions about approached towards AI regulations as well as questions on the survey that was conducted for the presentation. One question was based on the group’s opinion on a universal AI regulation approach and its challenges. The group emphasized that creating a universal framework is very complex and even impossible. The reason is that each culture has their own criteria to consider when developing such regulations. In addition, an example as South Americans might take an example of the EU AI Act but have to adapt this framework to their own unique cultural landscape.

Another question was about the conducted survey and any surprising results. The group mentioned that in regard to their own country, many answers from the survey were not surprising but rather expected. This for example included answers from German participants and their focus on data privacy in their answers.

Our screencast

Below, you can find the screencast that group 2 created for this topic.

Aktivieren Sie JavaScript um das Video zu sehen.
https://youtu.be/zMOS4TD1tsI?si=MnoH3UEcBRJbT8zb

Further information about AI impact on local culture: Cultural perception of AI

For more information, group 2 gathered a few resources for you:

This paper covers more regions and their cultural perception of AI and its current development on AI regulations

Ge, X., Xu, C., Misaki, D., Markus, H. R., & Tsai, J. L. (2024). How culture shapes what people want from AI. Association for Computing Machinery. Walter, Y. (2024). Managing the race to the moon: Global policy and governance in artificial intelligence regulation—A contemporary overview and an analysis of socioeconomic consequences. Journal of Global Policy and Technology, 5(2), 45-67.

YouTube video: The EU’s AI Act Explained

The EU has rolled out the world’s first AI regulation, classifying AI into four risk levels with tailored rules for each. This video explains the EU’S AI Act in a short and compact way.

Aktivieren Sie JavaScript um das Video zu sehen.
https://www.youtube.com/watch?v=s_rxOnCt3HQ

YouTube video: Understanding Understanding AI Regulations Across the World | Experts Share All

Experts on AI governance share which AI policies you need to be aware of and how governments are treating AI regulations across the world

Aktivieren Sie JavaScript um das Video zu sehen.
https://www.youtube.com/live/JCFao04h1ZM?si=2pZl0lbBZics7So-

Conference Session Report – Group 6

The presentation analyzed the growing significance of Artificial Intelligence (AI) in the field of education, particularly the use and acceptance by students, as well as the challenges from the perspective of educators. A case study was conducted to examine how students use ChatGPT for academic purposes such as research, text production, and proofreading. The study revealed increasing acceptance, but also concerns regarding ethical issues such as data privacy and plagiarism risks. The discussion highlighted that while AI can offer advantages for the learning process, it could also impair critical thinking skills. Additionally, the need for clear guidelines, ethical standards, and professional development for educators was emphasized to integrate AI meaningfully and sustainably into education.

Several key questions regarding the use of AI in education were raised during the presentation. One important question was how understanding AI concepts affects students’ ability to critically evaluate the use of AI tools. A solid knowledge of how AI works and its limitations enables students to use the technology purposefully and responsibly. This allows them to better distinguish between useful and problematic applications, avoiding the risk of AI replacing their own analytical abilities.

Furthermore, the question of how students can use AI without abusing it was addressed. An important answer to this is the teaching of ethical principles and the promotion of responsibility when using AI. This includes raising awareness of potential risks such as plagiarism, data breaches, and the danger of excessive reliance on technology. The question of whether AI can foster students’ critical thinking and problem-solving skills was also discussed. While AI offers potential for supporting complex tasks, it should not replace the need for students to continue thinking independently and developing creative solutions.

Finally, the importance of verifying the accuracy of information from AI tools was emphasized. Cross-referencing information with reliable sources and training in information literacy are crucial to avoid errors or inaccuracies and to ensure the quality of learning processes.

In conclusion, the presentation highlighted the complex challenges and opportunities of using AI in education. A solid understanding of AI is essential to allow students to work with this technology responsibly and creatively, while also addressing ethical and didactic challenges.

Further Reading:

• Becker, K. P., Parker, J. L., & Richter, D. (2024, June 18). Framework for the Future: Building AI Literacy in Higher Education. Moxie. https://moxielearn.ai/wp-content/uploads/2024/06/Ai-literacies-white-paper.docx.pdf
• Hale, Jerold & Householder, Brian & Greene, Kathryn. (2003). The theory of reasoned action. The Persuasion Handbook: Developments in Theory and Practice. 259-286. Accessed on November 25, 2024, on https://www.researchgate.net/publication/288101472_The_theory_of_reasoned_action
• Stanford University. Understanding AI Literacy. https://teachingcommons.stanford.edu/teaching-guides/artificial-intelligence-teaching-guide/understanding-ai-literacy#:~:text=Functional%20AI%20literacy&text=You%20might%20begin%20by%20accessing,AI%20development%20and%20training%20process.

Aktivieren Sie JavaScript um das Video zu sehen.
https://www.youtube.com/watch?v=dA7j391mCEQ

IMAGE IN HEADER

LinkedIn (April 24, 2024). Artificial Intelligence as a Teaching Aid: Transforming the Educational Process.

https://www.linkedin.com/pulse/artificial-intelligence-teaching-aid-transforming-educational-b-qmvpc

Discussion

The topic of politics and democracy continues to resonate with many young people. Culturally divergent understandings of these concepts were reflected in several contributions and in our small study. Broadly speaking, the questions and discussions can be grouped into three thematic groups:

• Risks such as fake news and distrust, particularly regarding political campaigns and elections.

• Potentials including greater participation in political processes and the strengthening of democratic systems.

• Questions about possible cultural differences in our study concerning attitudes towards democracy and AI.

Our study revealed potential cultural differences in the evaluation of AI. While there was general agreement on the need for AI regulation, differences emerged regarding the perception of transparency: 45% of Indian students noted a lack of transparency, compared to 72% on the German side. It was debated whether these results stemmed from cultural differences or the small sample size. The discussion then shifted to the risks of AI in democratic processes, with a focus on distrust toward political actors and institutions. AI has introduced numerous possibilities for voter manipulation and the production of fake news. Examples in this instance included the U.S. presidential campaigns and local elections in India. We emphasized that supranational or even international agreements are essential to achieve transparency in AI usage. This, along with other measures, is crucial for rebuilding trust in AI and political actors, ultimately benefiting democracy.

Literature from our session:

Interesting case studies

Our Video

Aktivieren Sie JavaScript um das Video zu sehen.
https://www.youtube.com/watch?v=kytXgZU-qrc&t=1s

Dear students, instructors and further interested,

on December 12h, 2024 from 2 PM until 6 PM CET (14:00 – 18:00 CET), the 6th Conference on “Intercultural Perspectives on Information Literacy and Metaliteracy” took place.

The conference used the GDPR compliant video conference system located at https://meet.gwdg.de/b/joa-fwe-eor-dys

The videos of all IPILM conferences can be found at our YouTube channel.

For our newest course on Intercultural Perspectives on Information Literacy and Metaliteracy in winter term 2024/25 we prepared some FAQ and Requirements for (successful) Participation in our transnational online courses. You can find them here, linked to the project information in the main menu.