AI Should be a Public Good

Governments can spur research and innovation in AI by establishing a secure and trusted data commons.

By Xische Editorial, November 14, 2019

Source: Fotoscool/Shutterstock

Why isn’t artificial intelligence (AI) a public good in the United States? That’s the question posed in a critical new piece by technologists Ben Gansky, Michael Martin, and Ganesh Sitaraman. Their argument is straightforward and long overdue. The development and use of AI technology is so important for society that it can’t be left to big tech giants alone. Moreover, the US government (and others like it) are missing a great opportunity to use AI for good across the public sector. About time, right?

Just look at the news in recent days. This week the Wall Street Journal reported that Google has been secretly collecting the medical records of millions of Americans through a venture code-named “Project Nightingale”. Working with Ascension Health, one of America’s largest healthcare providers with more than 50 million customers, Google gained access to millions of health records without any sort of anonymization. Remarkably, patients and doctors were unaware that the technology giant was involved with any sensitive information. 

Even more profound, neither Ascension Health nor Google seems to have violated any laws by secretly sharing these health records. Of course, there are clear ethical implications but the episode highlights how the US government (and others) need to update existing legislation to protect the data of its citizens. 

This regulatory blindspot highlights the central role that governments should play in our digital lives. If medical records were stored securely with the help of a government-administered data pool, such an overreach might not have occurred. It's time data is managed differently. The technology is out there, so what is stopping more governments from using it? 

Right now, according to the authors mentioned above, Google and other technology giants “have an extraordinary amount of data about how we behave, largely because they engage in widespread surveillance of much of our behaviour. Because AI depends on data, these companies have a huge market advantage over start-ups and entrepreneurs”. The solution is a public data pool enabling registered users access to a sort of digital commons. Verification protocols would be strict to keep out any hacker elements. 

There is another key line regarding the development of knowledge economies and access to data. The authors argue that greater government involvement with data will actually increase innovation in knowledge-driven societies. Far from stifling private sector growth, public data pools will help drive technology economies. “Proponents of big tech celebrate private-sector research and are right to do so,” the authors note. “But big tech companies, like all companies, have an incentive to fund research that will support their bottom line, and the profit motive doesn’t always mean a focus on the most important problems.”

Given the decentralised nature of technological innovation, there are many other countries where governments are taking proactive steps to increase the role of government in public data pools. Many of the steps outlined in the Times piece have already been taken by smaller countries like Estonia and the UAE. Given their nimble regulatory environments and commitment to future technologies, these countries have managed to integrate government and data. In Estonia, public medical records are already stored on a secure blockchain and the results have been encouraging. The UAE has established a Ministry of AI to create new ways for the Government to use the technology for good. It’s time for more governments to ensure that the benefits of AI are spread across society.

For the UAE, in particular, the idea that government participation in technology through public data pools will encourage more innovation is especially important. As a small country with large investment in building its own knowledge economy, any role the government can take in expanding research and innovation is critical for the health of the sector. 

The leadership fully understands this reality and has spent the last decade investing in everything from blockchain strategies to AI programmes and research centres. Abu Dhabi recently inaugurated the Mohammed bin Zayed Univeristy of Artificial Intelligence, the first of its kind in the world. Fresh investment into military research in advanced technologies has also been recently announced. One next step could be the creation of a public data pool in the country similar to the one outlined by the authors of this important Times piece. 

By taking the initiative, the UAE can test which solutions work best and then export its knowledge to other places. By doing so, it would develop another homegrown technology industry that is desperately needed from the United States to Zambia.