Some time ago, The Economist, a British news and public affairs journal, instituted a new feature: By Invitation. These are one-off columns on timely topics by a variety of authors. A recent one caught my eye. It’s entitled AI is the new Oracle of Delphi. That’s bad news. The article is by Carissa Véliz, an associate professor of philosophy and ethics at the University of Oxford.
Her article is brief, brilliantly reasoned and written, and worth reading in full. But in case you find it behind a paywall, or you’re short for time, here are a few excerpts. She starts this way:
Prediction is as old as intelligent life. In pre-industrial times, part of what made humans strong despite physical disadvantage was an ability to foresee the behaviour of other animals, which made it easier to hunt them. In the modern world prediction continues to confer an array of competitive advantages: if you run a company, for instance, everything from choosing what businesses to enter or exit to finding the right location for operations depends in part on forecasting.
The ai age has brought a boom in prediction…
She then makes a series of successive points:
Predictive algorithms are everywhere, opening and closing doors for us: deciding whether we get insurance, or a loan, or an apartment…
…The use of predictive technology to make decisions about people raises ethical questions that humanity has devoted worryingly little time to considering…
…Predictions are, at best, educated guesses. Often, they are riddled with prejudice…
…At a deeper level, there is arrogance in social predictions. When we predict people’s future as if we were forecasting the weather, we are treating them as inert objects, not as agents who have a say in their future and can defy their odds…
The outputs of predictive ai might sound like a description of the world, but they are “normative”: they implicitly tell us what to do…
…there might be things we shouldn’t predict, even if we could…
…It is disquieting that now that we are using prediction more than ever, we have no rules for it…
And she closes with this:
Beware of prophets and predictions. Only when we accept that we don’t know what the future holds, and act accordingly, can we be sure to live in a free society.
A lot to ponder. For example, a doctor’s diagnosis and prognosis about your health, when shared with you, risks becoming self-fulfilling. When/if shared with your prospective insurer, banker, employer, the impacts for you are even more profound.
But if Véliz had a meteorologist-rich audience, she might have taken the time to write that a weather forecast, though superficially less problematic, embodies similar ethical and moral difficulties. That’s because that forecast, like the medical one, is rich with implications for the future of people and communities in the forecast space-time window. Lives and property, safety and business continuity are at risk. Even as those in harm’s way are grappling with that threat, many outside that forecast window—including but not limited to tourists and travelers, commodity traders, potential investors, and competitors—will simultaneously be using that information for their self-benefit. From their position of safety, they will be making other travel plans, shorting or going long on market futures, seeking alternative suppliers of goods and services, and so on.
Ask Google AI (as I did within minutes of starting this post) if weather forecasts are a public good, and it’s likely you’ll get an initial response similar to the one I got:
Weather forecasts are widely considered a public good because they are non-rivalrous (one person using the forecast doesn’t prevent others from using it) and non-excludable (once produced, it is difficult to prevent anyone from accessing it). Taxpayer-funded agencies like the National Weather Service (NWS) provide this data to protect public safety and property.
But today’s weather forecasts are products of a complex, changing web of public-private partnerships. Numerical weather forecasts used to be the province of a few large national and international public enterprises. Today, AI-based forecasts are proliferating. Artificial intelligence is being harnessed to look past the physical atmospheric conditions to delineate the implications for the agribusiness, energy supply and demand, transportation, and other sensitive sectors. Very little of this information is publicly available.
A September 5, 2023 LOTRW post (also based on material from The Economist, as well as AGU’s Eos) looked at the implications for a core value of meteorology—international data sharing. It concluded:
Both articles make it clear that international data sharing is essential to achieving the needed weather forecasts and climate outlooks. But they stop short of addressing what’s needed if rich and poor nations alike are to fully share in the benefits of such information. In particular, mere sharing of weather forecasts per se stops short of what is needed. Harnessing the value of those forecasts depends upon additional information decision support for weather-sensitive economic sectors (agriculture, energy, transportation, water resource-management, etc.), for the financial community (e.g., investors and insurers). Such decision support is a given in wealthy nations; in the rest of the world, not so much. Thus, as matters now stand, it’s likely the wealthy benefit far more from weather information coming from poorer, more-remote corners of the world than do the peoples native to those regions. The current disparity is [sic] puts cooperation on shaky grounds at best. Only if the benefits of data sharing are themselves shared fairly can collaboration be sustained. More worrisome, if left unaddressed, the current gap in benefits to haves- and have-nots will widen as artificial intelligence comes into play. Nations of the world would do well to address all this sooner rather than later (presumably, under auspices of the WMO).
Only two years later, this language seems quaint and complacent in comparison to the immediacy and scope of the problem.
Forecasting is meteorology’s stock in trade. As forecasts grow more accurate, and as they grow more useful to decision-makers, we need to give their ethical and moral dimensions more attention. In this effort, we probably need help from other disciplines.
Philosophers: your attention is urgently requested!
(It’s not often that references to philosophers and urgency share the same sentence.)
