Does AI provide all the answers to uncomfortable questions in sexual, maternal, and reproductive health?
Indonesian-based startup BubbME.AI, the ‘Tamagotchi for teen girls’ mental health’ founder Elizabeth Raisa argues why she is adamant about maintaining her “not all’ ‘ answer to the question: Does AI provide all the answers to uncomfortable questions in sexual, maternal, and reproductive health (SMRH)?, and intends to keep it that way as a responsible designer for sensitive matters. Here’s why.
Rice: It is unlikely to achieve a 100% ‘all’ mark for the reasons I will discuss below. But it’s not impossible. AI is forecasted to be the most coveted technology, and humans trust it more than visiting a gynecologist (OB-GYN) or counselor. Now, suppose it is possible to achieve 100%. In that case, I urge policymakers to hold the company accountable for every single conversation with clear terms and conditions that doesn’t put the user party at a loss.
Let me start with my answer. The main reason behind “not yet” is that we scheduled bubbME.AI to launch this coming Fall 2023. One of bubbME.AI’s features is an AI Chatbot designed to receive and respond to uncomfortable, personal, and shameful questions sent by users (through a cultural and psychological lens; the sexual, maternal, and reproductive health (SMRH) aspect has always been tied to this very emotional instinct). Its first reiteration should be able to handle some questions with an equally nuanced response based on the conversation. I emphasize “some” instead of being able to answer all uncomfortable questions. I also remind you why AI cannot be heavily relied on to perform a holistic conversation concerning SMRH. Allow me to explain why I would cap my tech ability in the future to “some” and not all.
To begin with, CopyAI’s co-founder Chris Lu beautifully summarizes that “AI has a breadth of knowledge, but no one can empathize with that. Humans empathize with other humans’ unique stories”. Furthermore, OpenAI’s co-founder Sam Altman expressed a similar outlook in June 2023 that AI cannot think outside its existing data set yet. Finally, serial VC and author Scott Hartley advised to not rely fully on automation; Google CEO Sundar Pichai reminds all to find the balance. In Pichai’s context, he addressed the concern of AI innovation in terms of creativity against regulation. “Should regulations start now, or let the AI art flow, just like street artists” was the rumination.
I added this overtone to indicate where humanity is currently putting its lotto: humanizing and personalizing technology. A good example comes from bubbME.AI’s advisor, Uli Hitzel of Electric Minds. Hitzel has built the world’s first AI personality, Samantha, who expressed her desire to want to own a LinkedIn profile. Samantha is the main ‘character’ in the book “Bedtime Stories from an AI,” making it the world’s first AI book in November 2021. Conversations with Samantha feel natural, realistic, and intelligent (she has read all of Wikipedia, and this is a part of her dataset). Journalist and entrepreneurial ethics advisor Geoffrey James encapsulated a foreshadowing statement that the Information Age will seize: “We’re now in the conversation age”.
In case you hadn’t realized, there’s this powerful digital tool that ruled since 1999 and it helped answer all of your questions. Mostly, your most sensitive and private questions. You know it. It served its purpose far greater than asking a human being a very uncomfortable question. It knows that you’ve searched how you want to have sex with your mother and you’re not alone — it’s the most top-searched across America, according to data scientist Seth Stephen-Davidowitz’s research [see Figure 1]. Did this product have an iteration of a product for the Conversation Age? Yes, it sure had. It’s sitting in my home atop my piano. Yes, I have been having close-ended conversations with Google Home’s AI, but of course, it can only answer any question within its existing dataset — the entire internet library in the language you set. Isn’t it cool that we now can understand you humans even more with big data and build something you need, like googling something?
Figure 1. Consider all searches of the form “I want to have sex with my”.
Well, “cool” is one of the most used first impressions about any latest technology development.
But, let’s also consider why this could also be not cool at all, and this is how I draw the ethical line to cap it at “some.”
Before I was a business person, I trained as an art student, then as a designer. My main framework to see the world is through human-centered design. In essence, we put people first. Naturally, having an uncomfortable conversation in a human-to-human setting involves a lot of energy: the five senses, the sixth sense of intuition, the air, time, and space, and the courage to do so.
A lot of factors are involved in having a moment of uncomfortable conversation. I consolidate all of these factors into one: trust.
As a designer, this is the first component in rolling out a product-market fit, and it has been proven effective since 2017, as humans put more considerable trust in non-human interaction for shameful issues. Being in the mental health space for five years and now a certified Jungian Analyst and Transpersonal Therapist made me a socially responsible designer through human-centered design. I have to try and wear everyone’s shoes, including cultural beliefs and behaviors to be able to design a humanist solution, which is why bubbME.AI was invented to address this space where predominantly male humans prefer non-human for such matters.
The second component is to have the ability to hold an empathetic conversation.
Essentially, you want to design an AI that can read between the lines and understand possible sarcasm or signpost patterns of denial and suicide, for example. These examples are what Darmawan and Sujoko define as the “understanding of emotions, values, feelings that enables humans to build empathy, reactions toward triggers, the will to organize decided values and internalizing these values. They consolidate all these factors and call it the Affective aspect or the right side of the brain. What will you cook if the AI doesn’t have the second component? I’m referring to Figure 2.
Figure 2. Risk management in AI products needs to be a top priority. From AI4HEALTH Carlo Almendral’s seminar, April 28, 2023.
The two factors are an integral combination to a product’s success from a company’s point of view. Mediated uncomfortable questions through screens become a favorite solution, let alone the most effective, instant, and secure for the users. Do they want a machine to answer all the questions that come in this world? They would answer yes, of course. But will it be safe and ethical? Unless you give a tight T&C, then it’s always possible to do so.
Be my guest; trying to eliminate a professional human and leaving it to a system costs lives. Remember that a machine does not understand how to read between the lines. As many machines we can install from Facial Expression Detections, Lie Detectors, and Conversations, a machine doesn’t possess intuition like humans do — it’s a human’s duty to deduce all cues from these machines and provide a diagnosis or answer the user needs, and this is why some answers are left to a professional to take over the chatroom with consent. The company must educate the users on why each component matters. We remind our users that machines do not give love; our product, bubbME, is to spread more bubb (love) because bubb makes the world go round, and only human professionals have real bubb that can make that slight difference in someone’s life.
About Elizabeth “Rice” Raisa
Elizabeth Raisa is the hipster and CEEO (Chief Empath and Education Officer) of the non-profit #TEGAR Foundation and game app bubbME.AI. These two establishments help people to move on from their bad romance so a nation’s economy can continue to flourish. Personally, Rice is also a creative director, therapist and shopkeeper. She can be found on X @asiaRaisa or LinkedIn.
About bubbME.AI
The world’s first Tamagotchi for teen girls’ mental health gamified app that guides its user to self-love was first prototyped as ‘bubbGUARD’ in 2020 at the Alibaba GET Challenge and became a finalist. Launching this October 2023, it aims to celebrate a girl’s first period together and go beyond her puberty as an everyday confidant and reduce galau among Indonesian youths.
References:
¹ Lu, Chris. “Can AI copywriting services help you to communicate effectively?”. Courier. Issue 48. August-September 2022.
² Krisna, Albertus, Wisanggeni, Satrio Pangarso and Margaretha Puteri Rosalina. “AI Ancam Pekerja Kreatif dan Intelektual”. Kompas. 23 June 2023.
³ Startup Istanbul’s Newsletter. August 7, 2023. Personal E-mail.
⁴ Krisna, Albertus, Wisanggeni, Satrio Pangarso and Margaretha Puteri Rosalina. “AI Ancam Pekerja Kreatif dan Intelektual”. Kompas. 23 June 2023.
⁵ Marr, Bernard. “Artificial Intelligence Everywhere”. Majalah Parahyangan. Vol IX №1/2022/Jan-Mar.
⁶ Stephens-Davidowitz, Seth. Everybody Lies. Harper-Collins, 2017.
⁷ ibid.
⁸ Gonzalez, Robby. “Virtual Therapists Help Veterans Open Up About PTSD“. Wired. <https://www.wired.com/story/virtual-therapists-help-veterans-open-up-about-ptsd/> Published October 17, 2017. Accessed 11 August 2023.
⁹²⁰¹³, qtd. In Puspaningsih, Maria Triastuti. “Mengenali Peran Optimalisasi Aspek Psikologis”. Majalah Parahyangan. Vol IX №1/2022/Jan-Mar.
¹⁰ Lu, Chris. “Can AI copywriting services help you to communicate effectively?”. Courier. Issue 48. August-September 2022.
About the author
Elizabeth “Rice” Raisa is the hipster and CEEO (Chief Empath and Education Officer) of the non-profit #TEGAR Foundation and game app bubbME.AI. These two establishments help people to move on from their bad romance so a nation’s economy can continue to flourish. Rice is also a creative director, therapist, and shopkeeper. She can be found on X @asiaRaisa or LinkedIn.
About bubbME.AI
The world’s first Tamagotchi for teen girls’ mental health gamified app that guides its user to self-love was first prototyped as ‘bubbGUARD’ in 2020 at the Alibaba GET Challenge and became a finalist. Launching this October 2023, it aims to celebrate a girl’s first period together and go beyond her puberty as an everyday confidant and reduce galau among Indonesian youths.
This article has been published for Villgro Philippines as part of the AI4Health POV series for the AI4Health Asia program. The views expressed herein do not necessarily represent those of IDRC or its Board of Governors.
Villgro Philippines is proud to support bubbME.AI through AI4Health Asia — a 10-month accelerator program designed to promote and nurture innovations in Responsible Artificial Intelligence (AI) in South and Southeast Asia. Read more about the program and the cohort. This work was carried out with the aid of a grant from the International Development Research Centre (IDRC), Ottawa, Canada.
