Apologies in advance for the length, but the quoted extract is highly illuminating!

Recently, I came across the work of a graduate student, whose identity is charitable to keep hidden. The text was written rather well and included a critical review of the experimental studies conducted on one of my research topics: therefore, my name and the name of the colleague who worked with me on that line of research, Marco Marini, often appeared in the text, and consequently also in the final bibliography. The consultation of the references, however, had in store a few surprises. Among other entries, the following were recorded:

–Marini, M. (2013). When it’s better to choose the one you love: The effect of attractiveness biases in consumer choices. Judgment and Decision Making, 8(5), 476-485.

–Marini, M. (2019). How to get people to take risks? A choice-based measure of risk preference. PloS One, 14(1), e0209983. doi: https://doi.org/https://doi. org/10.1371/journal.pone.0209983

–Marini, M. (2019). Luring to a suboptimal option: The effect of payoff reduc- tion in a risky choice framing. Judgment and Decision making, 14(2), 198- 207.

–Marini, M. (2020). The asymmetrically dominated compromise effect in a dynamic setting. Journal of Economic Psychology, 76, 102-257.

–Paglieri, F. (2009). The attractiveness of decoys in economic contexts: An experimental investigation. Judgment and Decision Making, 4(4), 335-342.

Formally, this bibliography extract is flawless: the entries are correctly formatted according to the standards of the American Psychological Association (APA), the relevant information is all present, the articles are consistent with the topic of the student’s assignment, and the titles of the various contributions are, objectively, quite intriguing. The only problem is that… none of these publications exist!

The incident was neither a brave, subversive act of provocation (to demonstrate that university instructors no longer read carefully the written assignments of their students), nor a symptom of terminal stupidity in the student (only a very dumb cheater would try to falsify the references of the very same people tasked with evaluating their work): instead, it was the outcome of a naïve and inappropriate use of generative AI. The student, after writing the assignment themselves and inserting the appropriate references in the text, using the author-date APA standard, had incautiously asked ChatGPT to prepare the reference list, giving it their own text as part of the prompt. Unfortunately, the software compiled a bibliographic list in full compliance with APA standards, but without any attention to the truthfulness of the information included therein.

Here, however, we are not interested in the student’s misadventures, but rather in how ChatGPT produced its output, which was certainly not random: there is method to this madness. Firstly, the journals in which the fake contributions would have appeared are plausible, both thematically, and because Marini and I have already published in those venues in the past, or in very similar ones. Secondly, the volume numbers that are mentioned refer to issues that have indeed been released, and usually the numbering and year of publication match; in one case, the entire reference (PloS One, 14(1), e0209983. doi: https://doi.org/10.1371/journal.pone.02099 83) refers to an existing article, except that it is a study on a completely different topic, i.e. gender barriers in research at the South Pole (Nash, M., Nielsen, H., Shaw, J., King, M., Lea, M.-A., & Bax, N (2019), “Antarctica just has this hero factor…”: Gendered barriers to Australian Antarctic research and remote fieldwork).

The inconsistencies that emerge upon closer inspection are also revealing: the 2020 article attributed to Marini is listed as appearing between page 102 and page 257, except that there never was a single 155-page long contribution published in that particular journal, and probably not even in others, at least in the field of economic psychology; delving deeper, one discovers that the Journal of Economic Psychology, from 2020 onwards, no longer reports the page numbers of individual articles, but only their identification number, which is composed of a 6-digit code starting with 102, and the code 102257 (that ChatGPT creatively transformed into page numbers, 102–257) corresponds to the editorial of the issue following the one cited in the invented bibliographic reference.

At other times, the system falls prey to ambiguities of meaning: the decoy effect, which was the main focus of the student’s paper, is also referred to as the attraction effect in the literature, and the word “attraction” evokes the semantic field of affects, which instead has nothing to do with the technical phenomenon in question (i.e., a shift of preferences towards an option that is manifestly superior to another inserted ad hoc, called decoy). It is because of this semantic ambiguity that ChatGPT came up with a title like “When it’s better to choose the one you love: The effect of attractiveness biases in consumer choices” – a wonderful title, by the way, which I will certainly use, as soon as the opportunity presents itself.

In short, this false output is not due to anomalies or errors in the functioning of the software, but on the contrary it illustrates perfectly what ChatGPT is built to do (and does very well): generate linguistic strings (in this case, bibliographic entries) that have the maximum probability of satisfying the user’s request, based on similar instances present in the (huge) database to which the program had access during training. What ChatGPT does not do, and cannot do due to the way it functions (at least for the time being), is consulting the real world or an internal representation of it: the system does not work by checking the state of the world and describing it, but rather by constructing responses that are maximally consistent with the vast mass of linguistic data at its disposal, whose adherence to reality is by no means guaranteed.

https://link.springer.com/article/10.1007/s13347-024-00743-x (footnote deleted for readability)

More than a year ago a joint statement was issued by the Center for AI Safety. It was the one sentence quoted above. Famously, it was signed by more than 350 AI experts and public figures.

Now, of course, we cannot dismiss the actual and potential harms of artificial intelligence.

But, just as clearly, these 350 people must be among the last people on Earth you’d turn to for pandemic and nuclear war scenarios of sufficient granularity against which to appraise their AI crisis scenarios.

The more you aim for just-enough reliability, the more specific and narrow are the criteria of “just how good is just-good-enough” (“you must respond within x minutes of a call. . .”). Goal displacement is risked where, e.g., meeting government regulatory compliance equates to ensuring the infrastructure’s continuous and safe provision of its critical service. In so doing, safe and continuous service reliability eventually falls to the side when system operators feel compelled to be fast enough with just enough, knowing this is never enough all the time.

If we want anything more by way of highly reliable service provision, then that is left to us, not so much as consumers or citizens, but as amateurs who are now be their own reliability managers. So goes the economists’ wet dream of just-enough reliability. Gone are the days when anyone felt comfortable with discussions that include, “Elementary economics demonstrates that. . .”

Start with what passes today as a cliche: “Policymakers also need to worry about those other factors—societal, political, economic, historical, cultural, geographical, governmental, psychological, technological, ethical, religious etc—that are so undeniably part of policy analysis and management.”

That pincushion of “etcetera’s” Indicates things are just critical enough to get a nod, but not that critical to actually name. So too Samuel Taylor Coleridge argued “matter” was treated like a pincushion whose surface was hidden by all the sensations, thoughts and properties stabbed into it.

You ask today’s version of, “What are the important factors?,” and you get a pincushion of responses affixed with all manner of “etc’s” “Hail, Muse! Et Cetera,” the poet, Byron, put it in the third canto of Don Juan.

And yet, writes Wittgenstein: “Again and again, my ‘etc’ has a limit.”

World is suddener than we fancy it.

World is crazier and more of it than we think, 
Incorrigibly plural.
                            Louis MacNeice, "Snow"

There are so many different programs, projects, activities and initiatives connected to “income inequality” that the immediate challenge is to compare and contrast them before drawing generalizations about anything like an [Inequality] bracketed off from really-existing variability.

The comparison is not so much at the level of that country’s family support program contrasted to this country’s family support program, when it comes to a capitalized benchmark called [Inequality].

The comparison is more across many family support programs, much along the lines that no single heart is the same as another but these different and other different hearts set the stage for recognizing patterns across really-existing ones. That pattern recognition is of inequality, with a small-i.

Jesus Christ having a lot to say, but wise enough not to write it down

Everyone having the right not to be killed by people they don’t know

“A thing is a hole in the thing it is not” (Carl Andre, artist)

Decorum demanding that Medea kill her children offstage and that Macbeth do the same for King Duncan

Each person on earth being allocated a randomly unique number: “This one is yours. It’s irreplaceable.”

Understanding that carbon pricing and cap-and-trade are easy to talk about because they’re hard to implement. How else to buy time to avoid all the other approaches that are quicker by being context specific?

Seeing that in Trump, Boris Johnson, Putin, and Xi Jinping, we are weaponizing a late-version of collapse with its very own celebrity brands

---

Principal sources available on request.

I

Reliability professionals are central to translating statements of systemwide policies, laws and regulations into reliable real-time operations within and across the system infrastructures. This means reliability professionals are neither macro-designers located in the infrastructure’s headquarters nor micro-operators at individual facilities. Instead, they operate in between the macro- and the micro-levels, working in a very important middle domain within the infrastructure as a whole.

In this domain of expertise, infrastructure reliability in systemwide operations is achieved only if macro-designs are modified into different scenarios that take into account local conditions affecting infrastructure operations and where the real-time better practices that have evolved across a diversity of really-existing cases of operations are applied so as to ensure achievement of the original reliability mandates of policies, laws and regulation.

II

For example, the land board’s longstanding policy may be that livestock watering boreholes should be spaced 8 kilometers (5 miles) apart in order to reduce the effects of overgrazing. Indeed, land board members and staff may still insist it is their policy, even when your map of actual livestock water boreholes shows conclusively that boreholes are not spaced 8km apart on the ground. Does your map of allocated boreholes mean the 8km rule is not really land board policy?

No, it doesn’t

It is better to say that any such policy has to be modified in practice because variability in site conditions, aquifers, range composition and livestock characteristics differ so much (e.g., the hardveld is not the same as the sandveld). Furthermore, actually-existing practices for siting and spacing livestock borehole evolving across all the land boards and all their sitings, and this more up-to-date knowledge helps them in the placement of new livestock water boreholes (e.g. more knowledge and mapping now exist about the underground aquifers).

In other words, to say this map of livestock watering boreholes shows that the spacing policy was NOT in fact implemented misses the fundamental point that the policy was indeed implemented by land board members and staff in ways that cannot be attributed to their being expedient or corrupt, full stop. Even if the latter were true in some cases, no policy can be reliable if it is one-size-fits-all.

III

The chief implication of the preceding example is that the locus and focus of “implementation” shifts from micro-site—”drilling his borehole right here and right now”—to the middle domain where reliability professionals convert macro-policies into local contingency scenarios—”siting the borehole this side differs for us from siting the borehole that side”—and where better practices that have emerged out of all siting and spacing activities since the policy was adopted are used to modify new placements under the overall 8km policy.

This means that the micro-operators at any individual site—the drilling rig and operator, the borehole owner(s) and their specific herds and herders—are not the only unit and level of analysis for the actual implementation, here and now of the 8km policy. Implementation of borehole siting and spacing also takes place when teams or groups of reliability professionals adapt borehole siting and spacing in light of both locally contingent conditions and newer systemwide practices developing across different conditions relevant for up-to-date, reliable borehole placement.

IV

This also means that active micro-operators and reliability professionals–or at least their roles–need to be distinguished from each other. One or two drilling rig operators may be preferred by livestock owners because of their skills in getting results. But these drilling rig operators are reliability professionals when they also work with land board members/staff in the latter’s effort to identify more reliable scenarios for actual sitings as well as more up-to-date systemwide siting/spacing practices. Here they are in the role of reliability professionals because they have a bigger picture of borehole siting and spacing than when they work as a single driller at a single site with a specific livestock owner.

Or take another example. When the paravet is great one-on-one, developing unique relationships with each of his or her clients, then s/he is a micro-operator. When that same paravet acts according to his or her official job definition–“A para-veterinary worker is a veterinary science expert who, as part of a veterinary aid system, performs procedures autonomously or semi-autonomously”–then that system and team component points to his or her being a reliability professional. (Note these networks can be informal and not just formal ones.)

Last but not least, a case study rich in examples of networks of reliability professionals, involving pastoralists and others, is to be found in: Alex Tasker & Ian Scoones (2022). “High Reliability Knowledge Networks: Responding to Animal Diseases in a Pastoral Area of Northern Kenya,” The Journal of Development Studies 58(5): 968-988.

Bells were increasingly used not only to summon people to church, but also to provide another prompt for a belief act to those laity who had not attended: the major bells were to be rung during the Mass at the moment of consecration of the Host, and from the late twelfth century onwards we find texts calling upon lay people to kneel and adore where ever they were at that moment…

John Arnold (2023). Believing in belief: Gibbon, Latour and the social history of religion. Past & Present, 260(1): 236–268. (https://doi.org/10.1093/pastj/gtac012)

I

I suggest that early warnings promulgated as part of official drought management systems are designed to be bells in the above sense: People are to demonstrate their belief in the warnings when issued. They are to take action then and there because of them.

But, as Arnold also reminds us, demonstration of obedience always entails the possibility of failure. Heeding the warning might not work.

Indeed, some early warning systems are designed to fail because they are meant also for non-believers. The latter include, most notably for our purposes, those who subscribe to other types of warnings (e.g., https://pastres.org/2023/05/12/local-early-warning-systems-predicting-the-future-when-things-are-so-uncertain/).

This matters because the stakes are high when it comes to drought for both believers and non-believers. How so?

II

It is important to understand the conditions under which the designers themselves don’t believe in their own bell-ringing systems. In their article, “Drought Management Norms: Is the Middle East and North Africa Region Managing Risks or Crises?,” Jedd et al (2021) examine the efficacy official systems in the MENA region. They conclude:

Drought monitoring data were often treated as proprietary information by the producing agencies; interagency sharing, let alone wider publication, was rare. Government officials described the following reasons for this approach. First, it could create pressure on decision-makers to take action (politicizes the issue). Second, intervention measures are costly, and so, taking measures creates strong and competing demands for financial resources from agencies and/or ministers (increase political transaction costs). Therefore, given existing policies and institutions in the countries, it is unclear to what extent drought decision-making processes would be improved or expedited with increased transparency of monitoring information. . . .

This creates a difficult puzzle: In order to mitigate future drought losses, a clear depiction of current conditions must be made publicly available. However, publishing these data may require that agencies take on the burden of allocating relief if the release of this very information coincides with a future drought crisis.

https://journals.sagepub.com/doi/10.1177/1070496520960204

III

So then the obvious policy and management question is: When it comes to the efficacy of early warnings for drought, who do you want to start with: believers or non-believers?

I

Say you are residents of Oregon, a state in the US Pacific Northwest facing a magnitude 9.0 earthquake just off its coast. Aftershocks will likely be around magnitude 8.0 with a 60′ tsunami hitting the shore first thing.

Nothing has ever happened like that to Oregon. Some began thinking seriously about this earthquake and its aftermath only a decade or so ago. Thinking about the infrastructure interconnectivities within a regional focus is even more recent. People talk about the more recent spate of snowstorms, fires, flooding and heat dome effects as “eye-openers and wake-up calls” than as sources from which lessons are to be learned. According to the experts, emergency management is itself a relatively new profession and organizational priority there.

The good news, if you can call it that, is that key resources, like electricity generation and regional transmission, are on the eastern side of the state. But that too is at jeopardy if instead of a Cascadia subduction zone earthquake off the coast, we are talking about, say, a massive geomagnetic storm like the Carrington Event of 1859. That too can happen and take out a much wider swathe of electric and telecom assets.

II

What to do in response to these prospects of “earth-shattering” events?

One thing is: “get out of Dodge.” But then do you know what’s in store when you arrive somewhere you’ve never have been? That even the state’s infrastructure operators aren’t fleeing like mad indicates people’s preferences for known unknowns over unknown unknowns.

Known unknowns, after all, can be cast in the form of scenarios, and scenarios can be more or less detailed. Restoring water, electricity, telecoms and roads after the earthquake will be an immediate priority once saving lives is underway. We imagine the known unknown called the unimaginable all the time.

And the second we try to anticipate the unimaginable–that is, prepare for it–the preparedness scenarios beg to become granularized for now, not some other time. Operating in the blind during the 1755 Lisbon earthquake, which had a major impact on European history, is quite different than operating blind in the Cascadia one and its aftermaths.

Your scenarios are what separate you from unstudied/unstudiable conditions. “Humans can only really know that which they create,” as the older insight has it.

---

The Augustinian threefold present—as in: the past as present remembrance, the present as current consciousness, and the future as present expectation—is the cognitive linchpin of real-time management.

Yet, it’s an odd sort of a-historicism to deny utopian possibilities because we live in an endless present that forecloses on anything like the unexpected future.

---

Actually, the unexpected event is informative. In this case, uncertainty isn’t the absence of information.

It is one thing to say the present advances to the future it helps render for itself; it is quite another thing to say the future advances to the contingencies the present affords. That latter is like finding the best hamburger in town at the Vietnamese diner, at least for now.

---

Catastrophism, like irony, is a knowingness, not knowledge. As when people are more wont to quote Rilke about death being the part of life turned away from us rather than Wittgenstein to the effect that death is not an event in life.

Ironically, the post-apocalyptic novel—doomer lit generally—nails home the fact that we don’t need widespread fear and dread of COLLAPSE to provoke remedy and progress, because, well, we no longer believe in either. Less repeatedly said: But this means you are still here, reading the very words.

---

If, as they say, need connects everything, then rarely have we been as connected as we were when isolated from each other during the pandemic.

Janet Flanner, the journalist, reported in 1945 from war-struck Paris: “Everything here is a substitute for everything else.” Think: Cigarettes could be traded for food, food could be traded for clothes, clothes could be traded for furniture, and so on. It is in disaster where everything is connected to everything else; that’s why the only thing complete is “complete disaster” these days.

---

Not to worry, we’ll scale up later. Later on, presses the happy-talk of true believers, we’ll relax assumptions and add realism. Don’t bother the details; we know how to reduce overpopulation (just don’t have babies!) and save the environment (just don’t cut down the trees!). Just keep fossil fuel in the ground! So much of this suffocates in its own fat: This time it’s different; leave the complications for later.

You’d think that “radical” in “radical uncertainty” would require responses other than the same-old same-old. Yet in his book on the last financial crisis, Mervyn King, former head of the Bank of England, ends up recommending more of the same, i.e.: Radical uncertainty–King’s term–needs to be better reflected in economic and financial theories and practices. It seems that “radical” is dumbed down at the precise moment when needed most.

---

There is something worryingly “closed-time” about the Precautionary Principle. If we accept its definition—“the principle that the introduction of a new product or process whose ultimate effects are disputed or unknown should be resisted”—then the precaution is based on what is known/unknown by way of effects understood now. If so, there appears to be little or no possibility—no second chance—of any “afterwards” that demonstrates when the initial precaution was irreversibly in error.

And yet. . .the more you have to lose, the less you can take for granted. We are left somewhere between “Though to/hold on in any case means taking less and less/for granted…” and “to lose/again and again is to have more/and more to lose…” (Amy Clampitt from her “A Hermit Thrush” and Mark Strand from his “To Begin”). What to do? Elizabeth Bishop suggests in “One Art”: “Then practice losing farther, losing faster”.

---