Building and Beta-Testing Be Well Buddy Chatbot, a Secure, Credible and Trustworthy AI Chatbot That Will Not Misinform, Hallucinate or Stigmatize Substance Use Disorder: Development and Usability Study
Bibliographic record
Abstract
Background: Artificially intelligent (AI) chatbots that deploy natural language processing and machine learning are becoming more common in health care to facilitate patient education and outreach; however, generative chatbots such as ChatGPT face challenges, as they can misinform and hallucinate. Health care systems are increasingly interested in using these tools for patient education, access to care, and self-management, but need reassurances that AI systems can be secure and credible. Objective: This study aimed to build a secure system that people can use to send SMS with questions about substance use, and which can be used to screen for substance use disorder (SUD). The system will rely on data transfer via third party vendors and will thus require reliable and trustworthy encryption of protected health information . Methods: We describe the process and specifications for building an AI chatbot that users can access to gain information on and screen for SUD from Be Well Texas, a clinical provider affiliated with the University of Texas Health Sciences Center at San Antonio. Results: The AI chatbot system uses natural language processing and machine learning to classify expert-curated content related to SUD. It illustrates how we can comply with best practices in HIPPA (Health Insurance Portability and Accountability Act) compliance in data encryption for data transfer and data at rest, while still offering a state-of-the-art system that uses dynamic, user-driven conversation to dialogue about SUD, screen for SUD and access SUD treatment services. Conclusions: Recent calls for attention to user-friendly design concerning user rights that honor digital rights and regulations for digital substance use offerings suggest that this study is timely and appropriate while still advancing the field of AI.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".