How can we focus more on what really matters?
At times I feel very optimistic about the developments in artificial intelligence (AI). Yesterday I wrote about a new company changing the way we design our cities. I do believe we can design better cities with AI, however simultaneously I cannot embrace this wholeheartedly without exploring the other side of the equation: growth means decline, to some extent.
Amongst computer engineers at a developer event surrounded by free beers and free pizza — this perspective may seem rather hypocritical. Well, that is because of course it is. It is right there in between excitement and scepticism. I have friends that would be angry for me at pointing it out, and I could stare back at them sheepishly from the other side of a slice of pizza. Which AI company talks about inequality? What company would dare to even start handling suck a wicked problem? Let’s take another cheesy bite.
A Wicked problem is difficult or impossible to solve because of incomplete, contradictory, and changing requirements that are often difficult to recognise.
Technology is both a blessing and a curse, so we have to understand both the negative and positive.
Culture shapes technology and technology shapes culture, we have a responsibility to react both negatively and positively towards technology.
Technologists cannot save the world, it is the collaboration between technologists and social scientist together that can change the world for the slightly better.
A number that gets mentions a lot is extreme poverty, and that the percentage of those who are extremely poor is decreasing. This does seem to be the case, and it is something to celebrate.
At the same time there are other more worrying trends that we should be aware of and that requires us to pause and reconsider where we currently are as a global society. This is something that also should concern anyone embarking on the development of technologies related to the field of AI.
A goal could be a severe income decrease for the top 1–5%, however that seems an easy conclusion, yet quite challenging in practice. Earning less money is not a tantalising topic to most who have capital. AI companies going through pre-seed, seed funding or series A-C may not have this readily in mind? They are trying to survive as a company.
According to the Global Rich List, a website that brings awareness to worldwide income disparities, an income of $32,400 a year will allow you to make the cut. $32,400 amounts to roughly 300,000kr in Norway, at which point you would be in the global top 1% percentile.
The top percentile in Norway itself follows the global trend and has in the last few decades according to numbers from 2017 more than tripled their value. This does not include the offshore tax that is expected to some degree be present and unaccounted for within this top 1% percentile. So Norway is becoming more unequal too.
Could you use artificial intelligence to trace tax paradise money with a combination of legal or financial technology? Perhaps, but would this trigger a backlash of investment in legal tech to deter this chase or further complicate legal proceedings? Certainly not impossible. There is a trend of state capital decreasing and private capital increasing in some of the largest economies of the world and this could be a worrying trend.
In this sense the state does not have the means financially to outspend private capital to ensure accountability. So if those working for a state wanted to pursue this it is questionable if they could over time. There is seemingly additionally a lack of technical expertise within technology
On the other hand there is a increased degree of social spending.
Gini coefficient (/ˈdʒiːni/ JEE-nee), sometimes called Gini index, or Gini ratio, is a measure of statistical dispersion intended to represent the income or wealth distribution of a nation’s residents, and is the most commonly used measurement of inequality.
Government spending per capita has not increased much around the world except in Norway which has already been high. Healthcare provided nearly free of charge for all citizens and free education offered on university-level is partly a utopia.
The Folly of a Growth Mindset
Let me talk about two aspects of this somewhat misguided notion.
- The growth mindset so popularised is in a certain sense a ‘mirage’. Jay Lynch explains this well and sums up different research efforts that have been undertaken in this regard. The praising students for ‘working hard’ rather than for ‘being smart,’ produces to large of a difference and is too simple. Replication studies have not gained the same results etc.
- Exponential growth. In the larger picture as mentioned in a previous article about sustainability and cities we have known for a very long time that there are definite limits to planetary growth, since the 70s in fact. This knowledge has not become immediately widely available, however it is now to a certain sense an established understanding of the world.
If we compare these two simple facts to the current environment of entrepreneurship there seems to be a disconnect. The dialogues that are happening there in the large magazines, here and in the US — particularly in the field of young AI companies is different.
Even in impact investment circles in Norway there is an immense focus on ‘green’, ‘social’, ‘mind’ or otherwise growth of consciousness. I do not want to go to much into diverging viewpoints of prosperity to ruin or innovation. This could be read in the book by Charles Mann the Wizard and the Prophet.
The Prophets, he explains, follow William Vogt, a founding environmentalist who believed that in using more than our planet has to give, our prosperity will lead us to ruin. Cut back! was his mantra. Otherwise everyone will lose! The Wizards are the heirs of Norman Borlaug, whose research, in effect, wrangled the world in service to our species to produce modern high-yield crops that then saved millions from starvation. Innovate! was Borlaug’s cry. Only in that way can everyone win!
-Summary from Penguin Random House about The Wizard and The Prophet
It is not that easy, of course this is not the case for every company and it is a broad generalisation. Yet the focus on attracting funding, exponential scaling, autonomous operations and such does at times give of an eerie feeling.
I type this on my expensive Mac, again hypocrisy kicks in. Googling the etymology of hypocrisy it is from c. 1200, ipocrisie, “the sin of pretending to virtue or goodness,” and it does seem that way. The average Norwegian salary places us well in the global 1%. It is to some degree an equal country in an unequal world.
Resource usage required by technology
The field of AI operates through a network of established technologies of large amounts of cables, server etc. As such it is important to know that all of this requires resources to maintain its operations.
- Everyone uses 16 kilos of resources extracted from earth every day — metal, fossil energy, and minerals. If you live in the western world this number is much higher — up to 57 kilos of newly-mined minerals per day.
- Telephones, for example, are made from as many as 42 different minerals, including aluminum, beryllium, coal, copper, gold, iron, limestone, silica, silver, talc and wollastonite. A television requires 35 different minerals, and a computer more than 30.
- A newborn infant will need: 360 kilos of lead, 340 kilos of zinc, 680 kilos of copper, 1,630 kilos of aluminum, 14,800 kilos of iron, and 560,000 kilos of stone, sand, gravel and cement.
- Numbers are going up. Production of mined metal commodities is expected to increase by 250 percent by 2030. Most of this will originate from small and medium-sized mining operators, particularly in developing countries…
- Increased negative impacts. Small and medium-sized mining operators often lack the know-how and resources to apply sufficient health and environmental safeguards. The environmental impacts of mining is therefore likely to increase.
All facts The World Counts on the environmental impacts of mining.
The Optimisers and the Optimised
At the same recent event parallel processing with multiple stacks and great appreciation for Google Cloud’s capacities in easing this transition. We make machine learning techniques within the field of AI to make sense of a ‘messy world’. The first guide on Google’s website for AI education is the AI for Social Good Guide.
As mentioned in the introduction technology is not neutral, because people are not, and bias is inevitable. It is not like there is no mention of inequality in the AI community yet it is usually in regards to safety:
AI expert Yoshua Bengio suggests that equality and ensuring a shared benefit from AI could be pivotal in the development of safe artificial intelligence. Bengio, a professor at the University of Montreal, explains, “In a society where there’s a lot of violence, a lot of inequality, [then] the risk of misusing AI or having people use it irresponsibly in general is much greater. Making AI beneficial for all is very central to the safety question.” In fact, when speaking with many AI experts across academia and industry, the consensus was unanimous: the development of AI cannot benefit only the few.
–Future of Life Institute
There is a notion of risk that has been usual to think about in different fields, this is sadly nothing new and is described accurately by the human rights lawyer Cori Crider:
Now a similar logic pervades the modern marketplace, the sense that total certainty and zero risk — that is, zero risk for the class of people Lanier describes as “closest to the biggest computer” — is achievable and desirable. This is what is crucial for us all to understand: AI isn’t just about Google and Facebook targeting you with advertisements. It’s about risk.
–Cori Crider in Al Jazeera 13th of June 2019
The solution seems to be getting everyone online and connected. There is a disconnect between the ‘knowers and the known’. As Cori says a divide between the optimisers (data miners + data scientists), that optimise according to their values, and those optimised.
AI can predict the product a customer is going to buy before they buy it. We all encounter this when we order something on Amazon. It is undeniably handy when the website “recommends” items to add to a cart. Every business owner with a product to sell should be using AI in this way […] AI has the capability to read signals and sense each customer’s unique intent to purchase, upgrade or cancel.
–Straight Talk On Artificial Intelligence, Machine Learning and How It Can Help Your Business in Forbes Technology Council
In this sense you might be able to fire someone before they are leaving your company or organisation. You might need new skills when skills become automated, which can be the case for certain applications of machine learning with the field of artificial intelligence. Wealthier individuals can re-train for new jobs, perhaps not so much those who cannot afford new training or even education in the first place.
AI could be used to reduce inequalities, yet this is often in monitoring living conditions and understanding rising inequalities in cities. However justifications for monitoring, protecting and securing can be used for good purposes as much as the complete opposite. The idea of Surveillance Capitalism follows such thoughts further – surveillance as the business model of the Internet or freemium. Technology does play a role in income inequality.
These are only some starting thoughts and does not even start us on the topic of gender equality in technology. Not to mention gender inequality in series A or B funding.
It is important to stress that I do in no way want to claim any moral high-ground. Yet it is important to talk about both the great benefits we gain when using technology and what it requires in terms of resources or does not fulfil in terms of promises to a larger portion of the global population.
I simply think that this complex issue needs to be talked about particularly in the field of AI since it is not much present in most discussions I have seen so far.
This is day 12 of my project #500daysofAI. I hope you enjoyed the article!
What is #500daysofAI?
I am challenging myself to write and think about the topic of artificial intelligence for the next 500 days with the #500daysofAI.
This is inspired by the film 500 Days of Summer where the main character tries to figure out where a love affair went sour, and in doing so, rediscovers his true passions in life.
If you want to read more check out some of my existing articles:
11. The Rise and Rise of Spacemaker AI –published in Towards Data Science
Thank you for reading!