*This is the eleventh post in a symposium on Orly Lobel’s The Equality Machine: Harnessing Digital Technology for a Brighter, More Inclusive Future, selected by The Economist as a best book of 2022. All posts from this symposium can be found here. Further reviews can be found at Science, The Economist, and Kirkus.
Equality is appealing, can be SMART, and as Orly Lobel convincingly argues in her tour de force, the Equality Machine, is potentially newly within reach due to AI. But also harnessing the power of inequality, not just equality, is what will be needed to move forward.
I am a middle-aged woman, a mother to two kids, a wife, a teacher, and a part-time civil servant. I also am the founder of the Paper Prisons initiative, a research project that has put me in direct contact with people with criminal records who have in many cases been written off, whether by AI or any one of tens of thousands of collateral consequence prohibitions that confront people who live with records, however old and however minor.
I am just as susceptible to my technology as the next person, and while, yes, in theory I could be using the AI on my phone to exercise more, eat healthier, and sleep better, the reality is that the AI usually gets me to do things I probably shouldn’t be doing as much: watching delightfully entertaining videos, buying things, and social media. The stuff that is really useful – texting and communicating with my loved ones – doesn’t require AI.
And yet, what enables me to put the phone down and focus on what’s important is a sense of purpose that comes, not from equality, but a sense of inequality, a sense that things aren’t great, especially when it comes to the topics that have been on my mind as of late: the teen mental health crisis, persistent racial wealth gaps, the millions of Americans trapped in “paper prisons” – the gap between eligibility and delivery of relief from the criminal justice system promised by the law, and a sense that a lack of diversity in our innovation ecosystem threaten America’s competitiveness, which led me to launch the Innovator Diversity Pilots conference and Initiative.
The idea of a more brighter, more inclusive future, through technology, as beautifully and thoughtfully revealed as possible in the Equality Machine, is truly exciting but also, is incomplete. What is missing is the context in which the pursuit of equality must always be understood, continually nurtured, and never taken for granted. The five ideas below, by no means exhaustive, challenge the notion that equality is “enough,” and in some cases, even helpful.
- What is equitable is not always what is equal.
As Brandon Bouier has compellingly written about the White-Black income gap, “incomes cannot equalize by giving everyone the same percentage increases, since Black incomes are significantly lower to begin with.” If we want to achieve equality of outcomes, redistribution, and not just equality, in hiring, education, and other aspects of human capital formation are likely to be necessary. In that sense, AI tools like the ones that Lobel describes in her book, for making processes more inclusive can be thought of as ways of keeping things from getting worse or at least much worse, but should not be thought of as sufficient to make things better.
- To learn what works, we need to treat people unequally (by randomizing the intervention).
Inequality also has the benefit of being ruthlessly practical. As such, what works, as gauged by lived experiences, matters much more than algorithmic parity or other notions of equality. But to discover the impact of various policies and practices, as Lobel acknowledges, will require experimentation, curiosity, and a dedication to testing and iteration. The gold standard for doing so is through randomization – treating one group differently than another, and seeing what the impact of the “treatment” on one group is as compared to the other.
Ironically, then, to learn what works to advance equality, with respect to our new AI overlords, we need to get comfortable with treating people unequally, by randomizing the intervention.
While companies have long harnessed the value of randomization – in realms ranging from drug development to e-commerce to gaming, the administrative state has been slower to adapt rigorous pilots. But to serve the public and scale what works, administrative agencies, like AI systems, can learn from experience as their peers have shown, and from rigorous piloting, like agencies like the USPTO has.
- Inequality – by spurring outrage and action – is what propels progress
Another advantage of inequality is its ability to spur corrective action and the redistribution of power that is needed to sustain long-term change. As Lobel describes, AI systems are only as good as the data that are fed to them, and the people who are minding the decisions that are subsequently made. But unless the people in power are committed to driving the delivery of tangible progress and improvements, things are likely to stagnate. Change is hard, and the destabilizing nature of inequality is both a curse and a blessing – creating precarity but also the opportunity for change.
- The promise and reality of more can foster innovation
Another surprising benefit stumbled upon when researching the Inequalities of Innovation is that under certain circumstances, inequality can be good for innovation. As written in that paper, in “Common Sense of Progress,” Austrian economist Friedrich Hayek explores a few ways. First, rich people and firms have rents in excess of others that can be used to underwrite innovations that will eventually benefit all. Using prizes, direct funds, and other means, philanthropists have catalyzed breakthroughs in a variety of disease areas, including HIV/AIDS and cystic fibrosis. Research firm Bell Labs has led the world in “corporate” Nobel Prizes, and pioneered the laser, transistor, and solar cell, among other key inventions. Second, the desire for novelties (e.g., space exploration) supports discoveries from which subsequent innovators and the public can learn. Today’s tech billionaires, many of them self-made, suggest the third, and perhaps most universal, mechanism by which inequality spurs innovation: the promise of riches incentivizes talented people to innovate. Incentives thus serve as a double-edged sword: they motivate innovation, but by giving larger rewards for more innovative ideas, they also imply unequal economic outcomes and contribute to inequality
As such economic inequality, which is often perceived as the problem, seems to be an important part of the solution with respect to innovation. The desire to get ahead motivates productive effort and lures talent, creating productive clusters. The resulting innovation, at times underwritten or driven initially by those who are ahead, benefits the rich but, the hope is, also everyone else eventually. According to Robert Merges, the inequality that intellectual property contributes to is a “justifiable form of inequality,” because it provides significant benefits to the “least advantaged,” as identified by John Rawls. Innovation not only is spurred by inequality, but even without explicitly intending to, it can also make even the worst-off wealthier through abundance. In the spirit of Winston Churchill’s statement, “[t]he inherent vice of capitalism is the unequal sharing of blessings. The inherent virtue of Socialism is the equal sharing of miseries,” innovation may be a blessing whose unequal distribution just may be worth it.
- Liberation – not equality – is the goal.
If the goal is not just equality, than what? Many would argue that liberation – and in
particular, economic libration, should be the priority. Accordingly, if extreme inequality is the problem, perfect equality of income, access, or opportunity is not necessarily the best solution if it means everyone will be worse off. The demand for justice is not necessarily that all incomes or outcomes be equal, or even that every single opportunity be open to all, but that all have the right to not only survive but thrive. Using this as a yardstick for evaluation, alongside equality, may provide a faster, more surer path to the rich future that Lobel paints, and now it is up to us policymakers, companies, and citizens, to realize.