Coronavirus, AI and Digital Infrastructure

Original article can be found here (source): Artificial Intelligence on Medium

Coronavirus, AI and Digital Infrastructure

Artificial Intelligence and Shifting Capacities in a Crisis

Societies around the world increasingly depend on well-functioning and consistent digital infrastructure. What is the relationship between digital infrastructure, AI and the Coronavirus? Planning capacity in terms of the use of digital services is incredibly important — services can be disrupted when you last need it. Digital services are not in all cases set up to handle traffic in a crisis. Services that one could have predicted necessary beforehand do not necessarily have emergency measures in place.

Although speaking from a Norwegian perspective it is remarkable how much actually works here, one must also consider where services are being disrupted. In particular regards to education and social welfare — even the main information point for the coronavirus in Norway went down.

Compilation of websites and services crashing in Norway made by author

How is the Digital Infrastructure Coping?

The open-ended question is stated above, yet it is not easy to know the answer. The most visible aspect is the front-end of websites that can go ‘down’ due to traffic or for other reasons. However there are likely services that have such an increased amount of people approaching it or using it that it could be creaking at the seams or going from turning a profit to running into a great deal of expenses.

Netflix as an example might expect a certain amount of capacity per user every single month with space they pay for on a server — managing this capacity or expense depends on owning the servers being operated or the cost of available space. Currently that space could be rather low, so as long as people are buying they could keep selling.

Then again one could ask: is there any more strain than usual on the digital infrastructure? As has been shown on maps the decrease of transportation of people has drastically decreased the emissions across different countries — can the same be said for the digital or the opposite? Due to the amount of people that are working from home or in a quarantine using digital devices is there a cause for increased emissions from these other sources? What is being done to address the capacity by the actors who are hired to do so: that is Amazon, Google, Microsoft or other large cloud providers.

Infrastructure is an innovation priority in some cities often relating to the physical, built environment. Digital infrastructure is not even thought of as part of several planning departments — not even factored in as part of the built environment.

Mapping the City for Citizens

On the citizen side is there a digital infrastructure to help people in a city navigate away from the areas that are most affected? Before going into this there are of course heavy privacy and social issues relating to this — seeing as it could be stigmatised to live in certain areas or this can be attempted managed by various owners in one area. It is however interesting to consider this aspect. Especially since this is often a private initiative, which can be a concern due to the danger of health data going unwanted places — instead of a state or multilateral organisation taking a clear initiative to do so.

Over the last few weeks Dr. Kraft has been mobilizing innovators around the world to brainstorm and build a new “WAZE for COVID-19” app. Waze is a GPS navigation software app owned by Google. Daniel Kraft however showed during his presentation at Singularity University that there are several of these types of solutions being developed around the world. One example showing hotspots of activity, an app called TrackVirus. COVID-19 personal contact log and as well as using Google Map features to increase understanding where the worst areas are in terms of disease. Using AI and machine learning together with thermal screening. Testing however is very important, smart testing and early-testing.

Does the Hype Outstrip Reality?

On the other hand an article by Will Douglas Heaven in MIT Technology Review called AI could help with the next pandemic — but not with this one takes a different perspective arguing that the: “The hype outstrips reality […] too much confidence in AI’s capabilities could lead to ill-informed decisions that funnel public money to unproven AI companies at the expense of proven interventions such as drug programs.” Pragmatically the article goes against its title showing in fact a range of different applications AI can be used for arguing also that there are interesting developments in terms of the developments in the field of AI within image recognition in CT scans.

“A handful of preprint papers have been posted online in the last few weeks suggesting that machine learning can diagnose Covid-19 from CT scans of lung tissue if trained to spot telltale signs of the disease in the images. Alexander Selvikvåg Lundervold at the Western Norway University of Applied Sciences in Bergen, Norway, who is an expert on machine learning and medical imaging, says we should expect AI to be able to detect signs of Covid-19 in patients eventually.”

Yes, perhaps the hype outstrips reality both when it comes to artificial intelligence and digital infrastructure — yet that makes it even more important to break through the hype and learn what real value can be provided by exploring these aspect in the context of the current crisis.