Say hi to Sarah, the friendly new assistant from the World Health Organisation. She’s different from any other common chatbot in that she accesses your camera and microphone to interact. Her facial movements and gestures are very lifelike. Having had a couple of conversations, she is remarkably easy to talk to (does that make me sound sad?). Let’s dive in…

News

New AI weather forecasting model makes big strides: Zhiji, the latest iteration of the groundbreaking Pangu weather model from Huawei brings five day accurate forecasting down to a 3km range, making it very local. Read all about it here.

AI is coming for your trash: not only is AI coming for you rjobs, its coming for your trash too! A recent survey showed that a staggering 30-40% of US food products goes to waste, mainly ending up in landfill sites. AI is actively analysing trash to come up with strategies for restaurants and households to significantly minimise waste. See here for more.

WHO launches S.A.R.A.H to give health advice: in a first for the World Health Organisation, they have launched an interactive chatbot called Sarah who has a chat with you - not a text based conversation, but a chat chat! Try it out here and remember to give access to microphone and camera…

US Air Force chief to fly AI powered fighter drone: Frank Kelley, US Air Force Secretary, will be in the cockpit of a fully autonomous modified F-16 fighter jet to see first hand the power of the AI tech behind the weapon. See here for more.

AI makes retinal scans 100 times faster than normal method: another week, another health breakthrough, this time in eye health. Giving the task of generating retinal scans to AI produces results 100 times faster and with 3.5 times the contrast of traditional methods. Read about it here.

Tool

One of the cool things that AI can help with is research, into any topic you can think of. I was intrigued in learning about the SARAH model from the news section above, but thought that there might be more to it than the front end AI for personal health. I recently discovered the GPT AskYourPDF Research Assistant and put it to the task with the following simple question: “Find articles on the WHO AI model S.A.R.A.H and summarise the possibilities of this tyep of model in less than 400 words” What follows below is its output, along with sources for reference…

Prompt

Summary of WHO AI Model S.A.R.A.H

The World Health Organization's AI model named S.A.R.A.H (Synthetic Algorithmic Response and Assistance for Health) represents a pivotal advancement in employing artificial intelligence for global health initiatives. S.A.R.A.H is designed to aid in predicting and managing public health crises by analyzing vast amounts of health data to model disease outbreaks, recommend preventative measures, and optimize resource allocation.

The potential of models like S.A.R.A.H lies in their ability to integrate diverse data sources, including epidemiological data, genetic information, and environmental factors, to provide a comprehensive view of health dynamics. This integration allows for real-time surveillance and predictive modeling, enhancing the responsiveness of health systems to emerging threats. For instance, during the COVID-19 pandemic, similar AI models were used to track disease transmission and evaluate the effectiveness of non-pharmaceutical interventions worldwide.

Moreover, S.A.R.A.H can support decision-making processes by offering simulations and projections based on various public health strategies, thereby assisting policymakers in crafting evidence-based responses. The model also contributes to advancing personalized medicine by predicting individual health risks and suggesting tailored interventions.

Despite these benefits, the deployment of such AI models raises ethical and governance challenges. Issues such as data privacy, algorithmic bias, and the transparency of AI decisions need to be addressed to ensure the equitable and effective use of AI in health care.

In summary, while AI models like S.A.R.A.H have the potential to revolutionize public health management through enhanced predictive capabilities and decision support, careful consideration of the associated ethical implications is crucial for their successful implementation.

References

- World Health Organization. (2024). Ethics and governance of artificial intelligence for health: guidance on large multi-modal models. Retrieved from [WHO](https://iris.who.int/handle/10665/375579)

- Suryanarayanan, P., Tsou, C.-H., Poddar, A., Mahajan, D., Dandala, B., Madan, P., Agrawal, A., Wachira, C., Samuel, O., Bar-Shira, O., Kipchirchir, C., Okwako, S., Ogallo, W., Otieno, F., Nyota, T., Matu, F., Barros, V., Shats, D., Kagan, O., Remy, S., Bent, O., Guhan, P., Mahatma, S., Walcott-Bryant, A., Pathak, D., & Rosen-Zvi, M. (2021). AI-assisted tracking of worldwide non-pharmaceutical interventions for COVID-19 - Scientific Data. [Nature](https://www.nature.com/articles/s41597-021-00878-y)