Responsible AI Map — from AI Readiness Index 2020

Thoughts About the AI Readiness Index From 2019 to 2020

A short article exploring a few surface changes

To begin with I think it is important to be critical of rankings. Still, they are interesting and is not unlikely be a discussion point for people in business and politicians. This article does not go in-depth, rather it is a rapid look at the changes in the AI Readiness Index from 2019 to 2020.

The AI Readiness Index is created by Oxford Insights and the International Development Research Centre. It began in 2017.

“In 2017, Oxford Insights created the world’s first Government AI Readiness Index, to answer the question: how well placed are national governments to take advantage of the benefits of AI in their operations and delivery of public services?” [Oxford Insights]

According to the makers it was not created to fuel the global AI race, rather to help policymakers everywhere: “…see where they are performing well, and in which areas they may wish to target their attention going forward.”

Last year the 2019 edition of the Government Artificial Intelligence (AI) Index, showed the extent to which governments around the world were prepared to take advantage of the benefits of AI.

The index covers 194 countries and territories.

It ranks them according to their preparedness to use AI in the delivery of public services. This is important to stress.

Each country is allocated a score comprised of 11 input metrics under 4 clusters: (1) governance, infrastructure and data, (2) skills and education, and (3) government and (4) public services.

The data for 2019 as an example is openly available:

They mix a variety of indicators from different reports.

People working within the field of AI or as politicians may be interested to see how their countries rank.

What is the ranking for 2020?

The following is a screenshot of the top ten for 2020.

Screenshot of top ten slots from the AI Readiness Index 2020

2019 AI Readiness Index

It is perhaps interesting to some extent to look at the overall ranking from 2019 as a frame for comparison.

Screenshot from the top 12 slots from the AI Readiness Index 2019.

PS: I included Norway as it is my home country, thus extending the list to 12.

Changes over time and representation of data?

As mention in the beginning this is not an in-depth article.

It is interesting to briefly look at how data is represented.

Here is the representation for the US (top one) for 2020:

As such, it is only representative of certain aspects, and with some interesting measurements.

It is interesting to see the coloured map as well:

Here is the map for 2020:

Compared to the map for 2019, albeit with some different coloration.

One clear aspect to be noticed is the changing scale.

That has gone from 0–10 in 2019 to 0–100 (15–85) in 2020.

Additionally in 2020 there is a map that receives focus.

In terms of responsible AI the Nordics seem to be ‘in the lead’ so to speak:

Responsible AI Map — from AI Readiness Index 2020

In this sense there seems to be more focus on this ethical aspect.

It may be interesting to dive deeper into this another day.

How is responsible AI measured?

This is #500daysofAI and you are reading article 487. I am writing one new article about or related to artificial intelligence every day for 500 days.

AI Policy and Ethics at www.nora.ai. Student at University of Copenhagen MSc in Social Data Science. All views are my own. twitter.com/AlexMoltzau