Yute Chatz Chat Bot Suspended - How to Harmonize AI Innovation with Cultural Values

The recent controversy surrounding the National Family Planning Board’s (NFPB) Yute Expression’s Yute Chatz Chat bot in Jamaica, which was suspended by the Ministry of Health following outcry from various advocacy groups, highlights several critical points about the intersection of artificial intelligence, social values, and public health initiatives.

Context and Concerns

Purpose of The Yute Chatz Chat Bot:

The chat bot was most likely designed to provide a platform for youth to discuss health-related issues, including sensitive topics such as sexual health, mental health, and other personal matters. These types of chatbots can offer anonymity and accessibility, making them valuable tools for reaching out to young people who might not otherwise willingly seek help or information.

Outcry from Advocacy Groups:

The backlash indicates a significant cultural and social friction. The content or the way it was delivered was perceived to be promoting behaviours contrary to Jamaican values and morals. Concerns about promoting deviant behaviour stemmed from discussions on sexual health, sex and sexuality which were taking place with children on the app. According to the Jamaica Observer the groups said “ they were horrified about the information given to youngsters on such sensitive topics, especially without parental consent or supervision”.

Key Issues and Considerations

Cultural Sensitivity and Appropriateness:

The app was funded by international organizations which may or may not have a pulse on the local values. When developing AI-based health initiatives, especially in culturally diverse regions, it’s crucial to tailor the content to be culturally sensitive and culturally appropriate. This involves engaging with local groups and communities, understanding their values, and ensuring that the information provided aligns with cultural norms while still promoting health and well-being. This will lead to a greater chance of the initiative being accepted.

Possible biases in the training data:

Most likely the Large Language Models (LLMs) used for the chatbot were trained on datasets comprised of culturally inappropriate material and these were used without correction. Data from the Health and Family Life Policy of the Government of Jamaica should have been used to align the bot with cultural norms in Jamaica, reduce the chances of culturally inappropriate material being output, and thereby reduce the public backlash and mistrust that has now surfaced.

Transparency and Communication:

Going forward, the National Family Planning Board and developers of The Yute Chatz Chatbot should have clear communication strategies. They need to explain the purpose, benefits, and safeguards of the bot to the public, addressing any misconceptions and demonstrating how it supports youth health without promoting unwanted behaviours.

Involvement of Stakeholders:

Including representatives from these advocacy groups in the development and review process could help mitigate backlash. Their input can guide the creation of content that is both helpful and acceptable to the Jamaican public.

Ethical Considerations:

Balancing the need to provide accurate, comprehensive health information with respect for citizen values does not have to be challenging. Ethical considerations should guide the development and implementation of such technologies, ensuring they do not inadvertently harm the community they aim to serve.

Moving Forward

Community Engagement:

Ongoing dialogue with all stakeholders is essential. This includes not only with the advocacy groups, but with religious organizations, healthcare professionals, educators, privacy professionals and AI practitioners in the realm of ethical AI. Engaging in open, honest discussions can help build trust and create a more acceptable framework for the bot’s content and functionality.

Review and Adjustment:

A thorough review of the bot’s content and its delivery mechanisms should be conducted. Adjustments based on feedback from citizenry and experts can help align the bot’s objectives with national values as per the Health and Family Life Policy of the Government of Jamaica.

Education and Awareness:

Educating citizens about the benefits and limitations of AI in healthcare can foster a more informed discussion. This includes clarifying what the bot can and cannot do, how it protects user privacy, and its role in complementing traditional healthcare services.

Monitoring and Evaluation:

Implementing a robust monitoring and evaluation framework can help track the bot’s impact and reception over time. This data can inform ongoing adjustments and improvements, ensuring the tool remains relevant and effective.

By addressing these considerations, it is possible to develop AI-driven health initiatives that respect cultural values while providing valuable support to the users they aim to serve.

Jeehan Miller is a Director, IT Consultant and Digital Skills Trainer at Global Business Vault Limited. Contact us to access our training course "Artificial Intelligence for Productivity".