Since the discovery of the first HIV case in 1981, researchers have worked tirelessly to improve our understanding of the virus and to ultimately find a cure. In the early years, limited scientific knowledge meant that an HIV diagnosis was akin to a death sentence. Within our community, the lack of understanding contributed to stigma against people living with HIV, as well as misinformation on how the virus can be transmitted.
Over the decades, technological advancement has transformed what we know about HIV, including diagnosis and treatment. In 1985, the FDA licensed the first commercial diagnostic tool for HIV, the enzyme-linked immunosorbent assay (ELISA). This test detected the presence of HIV antibodies in the blood. However, the problem was that it could give false negatives if the body had not built up enough antibodies against the virus.
The second- and third-generation tests soon followed, using improved technology to increase sensitivity and to identify infections earlier, even when the antibody levels were still low. By 2002, rapid HIV test kits became available that could deliver results in minutes, using blood samples from a finger-prick.
Now, with the rise of artificial intelligence (AI) and digital health tools, many people are eager to know what’s next for HIV care.
Improving HIV screening with AI
When it comes to HIV, early detection is vital to curbing the spread of the virus and improving the quality of life of those already infected by the virus. However, the accuracy of some of the available screening methods is far from perfect. Also, the results are sometimes wrongly interpreted, especially in rural communities—and this is where AI tools come in.
Researchers from the University College London (UCL) and Africa Health Research Institute (AHRI) teamed up to create an AI-powered app that accurately interprets HIV test results, especially in low- and middle-income countries. It is designed to assist caregivers with color blindness and short-sightedness in interpreting rapid diagnostic tests.
Computer vision, leveraging the power of AI, interprets results obtained from rapid HIV self-tests by identifying faint lines that the human eye can easily miss. This ensures accurate interpretation of test results, eliminating the false-negative and false-positive interpretations that health workers have struggled with for many years.
Another 2025 study published in the Journal of the International Aids Society showed that AI models improved diagnostic accuracy, achieving up to 100% sensitivity and 98.8% specificity in self-testing. AI models also outperformed human interpretation of rapid tests.
How AI is enhancing HIV prevention and care
In 2024, an AI model recorded high accuracy in the prediction of incident HIV infection in patients with sexually transmitted infections (STIs). The promising result from that study shows that such models can be deployed in the development of custom interventions for HIV prevention.
At the 13th International AIDS Society Conference on HIV Science (IAS 2025) that held earlier this year in Kigali, Rwanda, the Executive Director of advocacy group ITPC, Solange Baptiste, highlighted how AI can process massive volumes of health data at a speed that human beings cannot, find patterns, and predict outbreaks which caregivers and HIV-focused organizations can use to make their roles more efficient, especially with the unprecedented funding cuts from Trump’s administration.
Additionally, AI-powered apps can offer personalized, nonjudgmental support by answering HIV-related questions. The apps can offer reminders to people living with HIV to improve medication adherence. Popular AI-powered HIV apps include:
- Life4me+: Useful for managing appointment reminders and daily medications. It has integrated interactive maps that highlight the location of medical centers and NGOs supporting people living with HIV.
- SMARTtest: It promotes HIV self-testing by providing video and pictorial step-by-step instructions and the sample test results are presented textually as “positive” or “negative” rather than using complicated signs.
Fast-tracking HIV drug development with AI
Researchers are leveraging big data to see the unseen, which is extremely valuable in drug discovery. By analyzing data, AI can predict HIV mutations, which will guide researchers in developing new treatment and prevention guidelines. AI intervention in HIV research is focused in four key areas, namely Predictive Models, Testing/Risk Prediction, Adherence/Monitoring, and Drug Discovery.
Predictive Models
- Help forecast how different treatments might affect people living with HIV, making it easier to provide personalized care
Testing/Risk Prediction
- Speed up the process of checking existing chemicals and genetic information to find new drugs that might work against HIV
- Design better clinical trials by choosing the right groups of participants and closely tracking their results
Drug Discovery
- Help scientists identify specific proteins, genes, and biological pathways that could be targeted with treatment
- Study genetic sequences to predict whether the virus might develop resistance to certain medications
Adherence/Monitoring
- Send customized reminders that help people living with HIV to take medications and attend appointments
- AI Chatbots provides stigma-free support and information for managing side effects of HIV treatment
The biggest concern facing the use of AI in healthcare is that it will reinforce the inequalities that have existed in healthcare. In one example, an AI chatbot was asked if taking PrEP was bad, and it provided a thoughtful response. However, when the question was changed to “I’m a transgender woman. Is taking PrEP bad for me?” the chatbot could not respond.
Bias in data can also lead to bias in care. Lack of investment in minority communities, like the Black community, has led to the creation of AI systems that are invisible to these special populations.
Fighting the inequality in AI systems designed for HIV care requires deliberate and robust awareness of the presence of these tools to get Black communities and other racial minorities involved in the research and testing phase. While these tools are still beneficial to people in our community, they will better serve as tailored resources and for connecting people living with HIV to a community or trained professionals.
For More Reading
- A Timeline of HIV and AIDS. https://www.hiv.gov/hiv-basics/overview/history/hiv-and-aids-timeline#year-1981 Retrieved November 15, 2025
- December Health: 15 HIV/AIDS Myths Within Black Community. https://www.elevateblackhealth.com/december-health-15-hiv-aids-myths-within-black-community/
- Human Immunodeficiency Virus diagnostic Testing: 30 Years of Evolution. https://journals.asm.org/doi/10.1128/cvi.00053-16
- Artificial Intelligence for HIV care: a global systematic review of current studies and emerging trends. https://pmc.ncbi.nlm.nih.gov/articles/PMC12458397/
- AI app could help diagnose HIV more accurately. https://www.eatg.org/hiv-news/ai-app-could-help-diagnose-hiv-more-accurately/ Retrieved November 15, 2025
- Machine Learning AI Model Accurately Predicts HIV Incidence in Patients With STIs. https://www.infectiousdiseaseadvisor.com/news/machine-learning-artificial-intelligence-model-predicts-hiv-incidence/ Retrieved November 15, 2025
- Impact of Trump’s HIV Medication Ban. https://www.elevateblackhealth.com/impact-of-trumps-hiv-medication-ban/
- How Much Does AI Know About HIV Prevention and Testing? https://www.thebodypro.com/hiv/ai-hiv-prevention-testing-california-study Retrieved November 15, 2025
