Details
Original language | English |
---|---|
Title of host publication | AIxEDU 2023 |
Subtitle of host publication | High-performance Artificial Intelligence Systems in Education |
Number of pages | 8 |
Publication status | Published - 2 Jan 2024 |
Event | 1st International Workshop on High-Performance Artificial Intelligence Systems in Education (AIxEDU 2023) - Rome, Italy Duration: 6 Nov 2023 → 6 Nov 2023 |
Publication series
Name | CEUR Workshop Proceedings |
---|---|
Publisher | CEUR Workshop Proceedings |
Volume | 3605 |
ISSN (Print) | 1613-0073 |
Abstract
Social media poses numerous dangers, such as the spread of toxic content, hate speech, false information, and moral outrage. In particular, the propagation of moral outrage on social media can result in harmful outcomes like promoting conspiracy theories, violent protests, and political polarization. Teenagers frequently use social media and are especially susceptible to these threats, being exposed to several risks that can impact their lives. Recent studies have confirmed that a combination of human and AI efforts can be highly effective in several domains. In this study, we examine human-AI collaboration’s potential and efficacy in detecting morality in online posts. To this end, we conducted a pilot study with teenagers to determine their ability to recognize moral content in online posts. Consequently, we exploited Prompt Engineering to program ChatGPT to recognize morality and comprehend its potential as an intelligent tutor in supporting students in such a context.
Keywords
- ChatGPT, Human-AI collaboration, moral classification, social media platforms
ASJC Scopus subject areas
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
AIxEDU 2023: High-performance Artificial Intelligence Systems in Education. 2024. (CEUR Workshop Proceedings; Vol. 3605).
Research output: Chapter in book/report/conference proceeding › Conference contribution › Research › peer review
}
TY - GEN
T1 - Using ChatGPT to Enhance Students’ Behavior in Social Media via the Moral Foundation Theory
AU - Schicchi, Daniele
AU - Upadhyaya, Apoorva
AU - Fisichella, Marco
AU - Taibi, Davide
PY - 2024/1/2
Y1 - 2024/1/2
N2 - Social media poses numerous dangers, such as the spread of toxic content, hate speech, false information, and moral outrage. In particular, the propagation of moral outrage on social media can result in harmful outcomes like promoting conspiracy theories, violent protests, and political polarization. Teenagers frequently use social media and are especially susceptible to these threats, being exposed to several risks that can impact their lives. Recent studies have confirmed that a combination of human and AI efforts can be highly effective in several domains. In this study, we examine human-AI collaboration’s potential and efficacy in detecting morality in online posts. To this end, we conducted a pilot study with teenagers to determine their ability to recognize moral content in online posts. Consequently, we exploited Prompt Engineering to program ChatGPT to recognize morality and comprehend its potential as an intelligent tutor in supporting students in such a context.
AB - Social media poses numerous dangers, such as the spread of toxic content, hate speech, false information, and moral outrage. In particular, the propagation of moral outrage on social media can result in harmful outcomes like promoting conspiracy theories, violent protests, and political polarization. Teenagers frequently use social media and are especially susceptible to these threats, being exposed to several risks that can impact their lives. Recent studies have confirmed that a combination of human and AI efforts can be highly effective in several domains. In this study, we examine human-AI collaboration’s potential and efficacy in detecting morality in online posts. To this end, we conducted a pilot study with teenagers to determine their ability to recognize moral content in online posts. Consequently, we exploited Prompt Engineering to program ChatGPT to recognize morality and comprehend its potential as an intelligent tutor in supporting students in such a context.
KW - ChatGPT
KW - Human-AI collaboration
KW - moral classification
KW - social media platforms
UR - http://www.scopus.com/inward/record.url?scp=85183204894&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85183204894
T3 - CEUR Workshop Proceedings
BT - AIxEDU 2023
T2 - 1st International Workshop on High-Performance Artificial Intelligence Systems in Education (AIxEDU 2023)
Y2 - 6 November 2023 through 6 November 2023
ER -