Evaluating the Usability of an HIV Prevention Artificial Intelligence Chatbot in Malaysia: National Observational Study
Bibliographic record
Abstract
BACKGROUND: Malaysia, an upper middle-income country in the Asia-Pacific region, has an HIV epidemic that has transitioned from needle sharing to sexual transmission, mainly in MSM. Men who have sex with men (MSM) are the most vulnerable population for HIV in Malaysia. In 2022, our team developed a web-based artificial intelligence (AI) chatbot and tested its feasibility and acceptability among MSM in Malaysia to promote HIV testing. To enhance the usability of the AI chatbot, we made it accessible to the public through the website called MYHIV365 and tested it in an observational study. OBJECTIVE: This study aimed to test the usability of an AI chatbot in promoting HIV testing among MSM living in Malaysia. METHODS: An observational study was conducted from August 2023 to March 2024 among 334 MSM. Participants were recruited through community outreach and social-networking apps using flyers. The interactions between participants and the AI chatbot were documented and retrieved from the chatbot developer's platform. Data were analyzed following a predefined metrics using R software (Posit Software, PBC, Boston, USA). RESULTS: The AI chatbot interacted with 334 participants, assisting them in receiving free HIV self-testing kits, offering information on HIV, PrEP, and mental health, and providing details of 220 MSM-friendly clinics, including their addresses, phone numbers, and operating hours. After the study, 393 human-chatbot interactions were documented on the chatbot developer's platform. Most participants (304/334, 91%) interacted with the AI chatbot once, 30 (9%) engaged 2 or more times at different intervals. Participants' interaction time with the chatbot varied, ranging from 1 to 31 minutes. The AI chatbot properly addressed most participants' questions (362/393, 92.1%) about HIV and PrEP. However, in 31 interactions, participants posed additional questions to the chatbot that were not programmed into the chatbot algorithm, resulting in unanswered interactions. CONCLUSIONS: The web-based AI chatbot demonstrated high usability in delivering HIV self-testing kits and providing clinical information on HIV testing, PrEP, and mental health services. To enhance its usability in community and clinical settings, the chatbot must offer personalized health information and precise interaction, powered by a sophisticated machine learning algorithm. Additionally, establishing an effective connection between the AI chatbot and healthcare systems to eliminate stigma and discrimination towards MSM is crucial for the future implementation of AI chatbots.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.006 | 0.016 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".