Collaborative Research: SaTC: CORE: Medium: Using Intelligent Conversational Agents to Empower Adolescents to be Resilient Against Cybergrooming

网络诱拐已经成为一个严重的社会问题，罪犯通过互联网利用与青少年的信任和情感纽带进行性剥削或性虐待。本计划旨在透过发展科技强化教育计划，教导青少年如何有效保护自己免受网路性侵害，以解决这个问题。这项研究的新颖贡献在于设计和开发了一种对话代理（即聊天机器人），青少年可以与之互动，练习（即“边做边学”）如何做出回应，从而使他们不那么容易受到网络美容技术的影响。对话代理提供互动指导，使青少年能够理解增加或减少他们对性剥削的脆弱性的反应，并解释原因。该团队正在收集父母和青少年的对话数据示例，以训练对话代理，并使用自然语言处理来开发聊天机器人。研究小组与青少年和专家合作，构思和设计功能，定制聊天机器人的界面，并建立一个教育程序。该小组与预防网络虐待和性交易方面的专家合作，编写培训材料，并确保对话代理对青少年是安全的。该项目的研究成果将帮助公众了解人工智能如何促进网络安全。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Cybergrooming has emerged as a serious social concern, where perpetrators exploit the trust and emotional bond with youth through the internet for sexual exploitation or abuse. This project aims to address this issue by developing a technology-enhanced educational program to teach youth how to effectively protect themselves from online sexual predation. The novel contribution of this research is in the design and development of a conversational agent (i.e., chatbot) that teens interact with to practice (i.e., “learning-by-doing”) how to respond in ways that make them less vulnerable to cybergrooming advances. The conversational agent provides interactive coaching so that teens can understand responses that increase or decrease their vulnerability to sexual exploitation and explain why. The team is collecting examples of conversational data from parents and teens to train the conversational agent and uses natural language processing, to develop the chatbot. The research team works with teens and experts to conceptualize and design features to customize the interface of the chatbot and builds an educational program. The team partners with experts in cyber abuse and sex trafficking prevention to develop training materials and ensure that the conversational agent is safe for teens. This project research outcome will help the public understand how artificial intelligence can promote online safety.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.