Silicon Valley

Google is set to take control of its subsidiary DeepMind’s health division, raising concerns that the company will be able to access personal, identifiable health data.

Some background: Google acquired London-based artificial-intelligence startup DeepMind in 2014 for $500 million. DeepMind is best known for its AlphaGo algorithm, which beat the world champion of strategy game Go in 2016, but it has expanded its AI capabilities into new areas, including health care. It’s been put to impressive use, notably helping to speed up diagnoses.

Streams of data: DeepMind aims for its mobile app Streams to become an AI assistant for nurses and doctors, using real-time patient data. It’s being used in several NHS hospitals. It will now be controlled by parent company Google, DeepMind’s founders announced in a blog post.

Broken promises? DeepMind has repeatedly promised never to share data with its parent company. It seems to be weakening that commitment. DeepMind's customer the Royal Free Hospital was found to have breached UK data protection laws in July 2017 over the handling of NHS patients’ data by the Streams app.

In July an independent ethics board said it needed to do more to establish its independence from its parent company. The decision to hug Google more tightly will surely deepen fears it isn’t truly committed to patient privacy.

DeepMind's response: A spokesperson said, "Our contractual agreements with existing partners, and their restrictive rules on patient data, are still in force and unchanged. Patient data remains under our partners’ strict control, and all decisions about its use will continue to lie with them. The move to Google does not affect this."

This story first appeared in our daily newsletter The Download. Sign up here to get the news first!