The company also highlighted the importance of diversity in AI development, introducing the Health Equity Assessment of Machine Learning (HEAL) framework to ensure AI models perform equitably across different demographics. Google's partnership with Apollo Hospitals aims to bring AI-powered screenings for TB, lung and breast cancer to under-resourced communities in India. Other updates include the use of Google Lens for skin condition identification, more visual mobile search results for health conditions, and the extension of crisis hotlines in search to more countries. YouTube's AI-powered dubbing tool was also mentioned, which helps creators reach a wider audience by breaking down language barriers.
Key takeaways:
- Google is developing personal AI experiences in Fitbit Labs, where premium users can test and provide feedback on experimental AI features. These features aim to provide personalized insights based on users' health and wellness data.
- Google is expanding its Med LM family of models to include multimodality, starting with Med LM for chest x-ray. This will enable findings, classification, semantic search, and more, with the aim of improving the efficiency of radiologist workflows.
- Google is using a framework called Heal (Health Equity Assessment of Machine Learning) to ensure that AI models are developed in a way that reflects the diversity of people and experiences around the world, and to correct biases in healthcare.
- Google Lens has been updated to allow users to take a picture of a skin concern and find visually similar matches from the web to inform their search. This feature is available in more than 150 countries.