OpenAI Vakfında Sağlık için Kamu Verileri

In April, we launched ⁠ AI for Alzheimer’s, the OpenAI Foundation’s first program in Life Sciences and Curing Diseases. Today, we are introducing our second science program: Public Data for Health.

Our focus in AI for Alzheimer’s is to go after one disease affecting many families where we believe AI can help scientists develop better treatments. Our focus in Public Data for Health is to enable progress across the life sciences, on many diseases, by funding the creation and preservation of high quality scientific datasets made broadly available to researchers.

Scientific data are observations about the world around us, the foundational input to research and discovery. In fields where breakthroughs are verifiable without the collection of new data, such as parts of mathematics, AI systems have recently started contributing ⁠ new ⁠ knowledge ⁠ . In biology, meanwhile, AI systems are increasingly able to analyze biological information at scale and recover hidden structure even from incomplete evidence, sometimes with astonishing efficiency.

However, we expect many remaining breakthroughs in preventing and treating disease to come from pairing the intelligence of new models with more observations of the world—in other words, more data.

Some datasets with enormous public value may never be created or shared because no individual institution has enough incentive or capacity to fund them.

That makes the support of public data for health a strong fit for the OpenAI Foundation.

We are starting by supporting more than $125 million in grants across an initial tranche of nonprofits and universities, spanning many layers of data: from molecules, to epidemiology, to regulatory knowledge.

OpenADMET ⁠ will create open datasets, benchmarks, and blinded competitions to test whether AI models can be trained to predict how small molecules are absorbed and distributed across the body, to make drug development more predictable and reduce the failure rate of new drugs.

90% of drug candidates fail in clinical trials ⁠ . That is often due to the difficulty of predicting how they will be absorbed and move around the body—their “ADMET” properties ⁠ . Unsolved prediction challenges like this can be a good fit for AI, when paired with high quality data to ground the accuracy of predictions from different models.