FAQ#

A plain-language tour of the most common questions about this site, in rough order from “what is this?” through “how can I help?” to some “harder questions.” Nothing here asks you to believe anything — the whole project is offered to be tested, not believed. Hence #AuditTheMath. All answers are open by default — use Collapse all to scan the questions, Expand all to reopen them, or jump to any question from the list just below.

Getting oriented#

What is Balospe.com, in one sentence?

It is the home of LLoL’s Good News Pack and Matheo Study Series — a long, careful attempt to answer one question: is there a gentle, kind, reasonable way for humanity to avoid destroying itself, and if so, can it be written down clearly enough that anyone can check it?

I only have two minutes. Where do I start?

The Nano Flying Scroll Exhibit — the whole message on five business-card-sized pages, free to download, print, and share. For the full collection afterwards, see the Good News Pack MMv3 gateway. However, bear in mind that while the ~144 words of the first biz card offer a bird’s eye overview, they are necessarily so compressed, that different audiences will benefit from much more details on how to connect the dots to why it matter for them. The rest of the website is about working towards that goal and that is where LLoL will fail unless a great many others will help to support his efforts to find the best plain words to explain what he really means.

I don’t have much time — what’s a brief overview of all this?

There is a top-level overview of the Balospe.com cause in one place — it samples from across the ideas on Balospe.com into one condensed read, at the speed you have time for: 30 seconds, 5 minutes, or a 72-minute Requiem. (For the message in miniature instead, the five-card Nano Flying Scroll Exhibit is the two-minute option above — and may suit you better.)

Why is some of the writing so dense, with so many acronyms?

Because the ideas are compressed from 6+ years of work, and because precise terms let claims be checked rather than just argued about. We are adding plain-English on-ramps everywhere. Where you meet a term like BABL, ZION, or h_star, hover it for a tooltip or follow it to the glossary. If a page reads like a wall, that is a bug in the explaining, not a test of you.

The problem#

What is the actual danger you are worried about?

Chiefly, accidental catastrophe from systems too complex for anyone to steer well — with nuclear winter as the sharpest example (roughly 12,500 warheads still exist, and an exchange could begin by mistake in under an hour). The same coordination failure shows up in pandemics, AI risk, and climate. The claim is not that doom is certain; it is that doing nothing is the default road to locking in doom.

Isn’t this just doom-mongering, or another end-of-the-world cult?

A fair worry, and the site tries hard to earn the opposite. Three differences: (1) it leads with a constructive plan, not just alarm; (2) it refuses fake hope — every quantitative claim is published to be checked (hence #AuditTheMath); and (3) it asks for scrutiny, not obedience or blind trust. A cult wants you to donate your trust, eventually blindly. This undertaking asks you to audit its math and show publicly and transparently where it is wrong. Take out the radical and improving transparency out of LLoL’s vision for ResearchCity and it will die predictably. Hence, LLoL’s call to support #AuditTheMath: this task is so much bigger than any single person can do.

Why “Don’t Panic”?

Because panic is itself a failure mode. The cards open with Don’t Panic on purpose: the argument is that calm, coordinated, gentle kind reasonable action can measurably change outcomes — so the useful response is not fear but participation. LLoL’s Don’t Panic Guide contains re-imagined lyrics of some well-known songs (and some new songs) to help stay the course. The songs have been helping LLoL to keep focus in all this, so they are offered freely to anyone who may benefit from them. They also document aspects of LLoL’s spiritual inspiration for this work.

What are the two kinds of fear?

Aristotle noticed that most of us are moved more by fear than by reason, and so we need some structure outside ourselves to do right when fear takes the wheel. That leaves two very different ways to be afraid, and confusing them is costly.

The first is FEAR as Forget Everything And Run: panic, scatter, freeze, and hand the wheel to whoever sounds most certain. It is the oldest tool of control — a frightened, forgetful crowd is easy to steer — and “don’t think about it, it’ll be fine” is the lullaby that keeps it asleep.

The second is FEAR as Face Everything And Rise: the same danger, the opposite move. Not denial, not panic, but looking straight at it and standing up. The difference is not courage you are born with; it is whether there is something solid to rise into. That is what an epiocracy, and a ResearchCity built to run on it, is for: a gentle external structure to face the hardest things together instead of running from them alone. So the honest response to the danger is neither doom nor denial — it is to face it, and to #AuditTheMath that tells us whether we can actually lower the odds.

The proposed solution#

What is ResearchCity?

A proposed neutral, well-supported place (scaled up in 7-8 well-defined Stages) where the world’s best people work full-time to serve Reality by solving the hardest shared problems. Such a coordination infrastructure does not currently exist at the scale the current existential threats require. LLoL offers this as his vision in a proposal to be reviewed. It is not an existing institution and the plans LLoL included (see SD8 and the ResearchCity summary) are not plans that can be implemented 1:1. They offer one of maybe dozens of ways for how to implement ResearchCity. The reason the 7-8 Stages for scaling up are essential is because each of these is designed to test specific aspects of the overall infrastructure and to rewrite all the plans based on the findings. Cutting the process short would be predictably crippling.

What is the ~$8 ask and cap of “~$8 per year per person per stadion” about?

~$8 is a deliberately chosen ask as a recommended voluntary buy in. It is a request to buy into one of ResearchCity’s 1600 research “stadia”. For one year. By one person. The cost is a little over two cents a day. New year new stadion, or maybe not. The allocation of funds is tentative; it is meant to be respected and recorded by ResearchCity, but it is not binding in the sense that ResearchCity stadia can reallocate funds received as they see fit in order to keep the overall integration work balanced. Not all stadia are necessarily glamorous enough to attract the funds they need to best serve all of ResearchCity and thereby all of the world. Hence reallocation. The ~$8 are the official limit for everyone /year/stadion, meant to eventually be adjusted up or down with the median income in the world’s poorest nation, precisely so that anyone on Earth can afford to buy into “their” stadion to be fully represented in ResearchCity — and moneyed special interests are kept at arm’s length to ensure they cannot buy influence that exploits others.

~$8 is not a claim that the initial ask of ~$8/person/year is enough for everything. It cannot run all of LLoL’s ResearchCity as envisioned. The real budgets are to be worked out stage by stage. The ~$8 are the entry point, posing the question to the world, whether (i) enough people are actually interested, and (ii) moneyed special interests and who-knows-which nefarious organizations will actually allow this to happen. It’s a proof of principle that such a system working for the common good can in practice be funded by a world as heterogeneous as ours. In that sense it is a historical experiment in bridge-building through an effort that is funded by a great many people barely noticing the cost rather than by a few who could then expect a say (beyond speaking for truth gentle kind reasonably). Theory suggests strongly that a multitude of small contributions can accomplish what several equivalently large ones cannot. Hence, general principles of complex systems design are the driving reasons behind the ~$8/year/person/stadion cap and the annual ~$8 recommendation.

More to give? You can back more stadia, not by paying more per stadion. ResearchCity is envisioned as a network of 1,600 research stadia (its structure), and you may back as many as you like, up to all 1,600. So the regular annual cap for any billionaire is about $8 x 1,600 ~ $12,800, and no one — however wealthy — may exceed it in a normal year. The cap is the feature: it is designed to stop anyone from buying too much undue influence.

Start-up exceptions (temporary, launch only). Two real-world frictions get a one-time allowance, so an initial moment of change is not wasted:

  • Catch-up: LLoL’s work is roughly 7 years late (the marathon began in 2020), so during the start-up phase you may also buy in for those earlier years — up to x7.

  • Sponsoring the excluded: only about 1 in 8 people worldwide even has a credit card, so you may also cover the buy-in of up to 7 others who currently cannot — x8 (yourself plus seven).

  • How you give is up to you: some may wish to only buy in with $8 initially but may then decide to opt in with more. That’s OK. However, please, keep track of the limits for the time being until ResearchCity can set up an infrastructure to automate that.

Together these set the absolute start-up ceiling for maximal buy-in at about 448x an All-Stadia Backer’s level ($8 x 1,600 x 7 x 8 ~ $716,800) — a temporary maximum (56x the normal cap) that returns to ~$12,800/year afterwards. The full arithmetic is on the Buy-In page.

Where the number came from (for those who care). The ~$8 is scaled up from the two small coins of the widow in the Gospel (Mark 12 / Luke 21), who gave her last two cents while wealthy people gave large sums — and of whom Jesus said she had given more than all of them, because she gave out of her poverty. In fact she literally gave more than all of them, because she inspired LLoL’s idea of using this crowd-funding strategy for ResearchCity. The 1,600 stadia are an echo of Revelation 14:20, reinterpreted as LLoL’s prayer — that instead of filling 1,600 stadia (like that one in ancient Laodicea) with the blood of ~8 billion people (roughly the volume — the arithmetic comes startlingly close), Reality may relent instead and grant the grace that this world needs to gentle kind reasonable scale up instead 1,600 research stadia to avert “Armageddon”-disasters. Take or leave the inspiration, like you may take or leave Kekulé’s snake that inspired his chemistry. The ~$8 ask, the cap, and the small-contributions principle stand on their own. (The volume coincidence behind the 1,600 — where the modern forecast and the ancient prophecy meet — is laid out on the 1600 Stadia page.)

What is the Jubilee System?

A multi-scale “reset” framework for keeping an innovation economy healthy over the long term — its smallest cycle is the weekly 6:1 Shabbat rhythm of work-and-rest, and its largest is an agreed upon and well-prepared 50-year reset for updating in order to ensure that everyone gets to benefit from the innovation economy developed by those who get the chance to do pioneer new work. Practically, proper 50-year Jubilees form a more perfect union, improve justice, grow tranquility, common sense, well-being, and liberty for all, over the long term. The Matheo Study Series works out the mathematics behind it. The rhythm is defined by nested Shabbats and more; the Jubilee System is the larger multi-scale framework.

How you can help#

How do I buy in? (Become a Select-Stadia Backer)

Become a Select-Stadia Backer: back one or more research stadia of your choice. It starts at just $1 for a single stadion; about $8 is the recommended buy-in (through the live GoFundMe campaign). You can scale up to as many of the 1,600 stadia as you like. See how to buy in.

Can I do more?

Yes — become an All-Stadia Backer and back all 1,600 stadia at once. It starts at $1 for each stadion (for exactly $1,600) and is capped: the usual ceiling for each stadion is $8, so the overall cap is at max ~$12,800. During the startup period the absolute cap for various reasons is increased by x7x8 = 56-fold to compensate for diverse roadblocks (max ~$716,800 to compensate for up to 7 years of start-up catch-up, plus sponsoring up to 7 others who are otherwise locked out). Long term the limit will return to about ~$12,800 per year as the start-up difficulties get resolved. The strict cap limits exists so that no one, however wealthy, can buy or inadvertently shop for influence. See Become an All-Stadia Backer.

What can I actually do?

Pick whatever fits: share the Nano Flying Scroll Exhibit (print cards, make a sticker, post a card online with #AuditTheMath); review the math or other expert insights and show where it is wrong; or back the work to help others review the math and to integrate the results of all such reviewing in an efficient way, such that no insight is lost. Such work is not free, as those who do it also need to eat and live, hence the need for buy in. Yet, the single most valuable thing is honest, critical feedback. If ResearchCity turns out to be a false hope and LLoL has been merely making it up, stringing himself and everyone along, deluding himself that he has found something important, whereas in Reality, it was merely his own brain pattern-matching what he shouldn’t have, if that is the case, then the rest of the world better know it fast, because the last thing that LLoL wants to do is to spread false hope: it makes the heart sick and in treacherous times like today that is particularly deadly.

Can I reuse the cards, posters, and text? Do I need permission?

No permission needed. The material is released under the Jonah License (with CC0 / public-domain intent for the cards): download, print, fold, translate, put it on a bumper sticker or the side of a building. That openness is the point. Please stick to the local legal regulations to avoid making a mess. Zoning exists for a reason, but it should not keep you from Investigating and Organizing to help others Navigate the myriad challenges represented in the Good News Pack.

How do I give feedback or get in touch?

Most pages carry a FeedbackFlow (FF) link at the bottom; contact details are on the exhibit’s contact card (Nano Flying Scroll Exhibit). Critical, specific feedback is especially welcome — it is how the work improves.

However, please be patient. This whole website was set up by 1 (one) person. LLoL isn’t a committee or a huge organization and replying to email is not his strong suit. If not sufficient funds are generated by those interested in supporting such work, LLoL will not be able to sustain keeping this door open for much longer. Every marathon must come to an end eventually. Hence, therefore, for now, please self-organize. Fig. 16 in the Don’t Panic Guide has practical tips for how to do that among friends. It’s LLoL’s Continuity Of Operations Plan for whatever transitions ResearchCity’s introduction may trigger.

About LLoL, and the harder questions#

Who is LLoL, and why should I listen?

Laurence Loewe of Laodicea — an evolutionary systems biology researcher (Dr. rer. nat.; formerly Edinburgh and UW-Madison) who has spent 6+ years laboring to produce this work on his wid-e research marathon to avert existential disasters. See about LLoL. But “should I listen?” is the wrong test: the project asks the world and you to check the arguments, which stand or fall independently of who makes them.

Does he really claim he could have stopped COVID?

No — and the distinction matters. What he owns is one auditable claim (Claim A): that by early 2020 he had relevant modelling expertise and a working, actionable pandemic model — and that he did not publish it in time (and that he could have published in 2019 the core model of what became his Matheo-b19 study, if he had cared to stay true to a certain long-standing research interest of his). For that earlier publication to have actually changed the pandemic’s course, a whole chain of further conditions would each have had to hold (Claims B–D; more details in the Supporting Information of the Matheo-b19 study). He can neither prove nor disprove that the chain held: he cannot show he would have made the difference — but no one can show he would not have, because the possibility was genuinely open if his Claim A holds. What he can show is what he actually built; whether timely publication would have mattered is for others to judge. It was that unresolved possibility — that it might have actually mattered — that set LLoL on his journey into mathematical theology. But the story behind this is long and detailed and not many reviewers are willing to entertain the necessary details for long enough to understand the big picture that convinced LLoL. Hence, until the books are opened on Judgement Day, LLoL has come to suspect (until someone presents better evidence) that there is a very real chance that Yah worked hard to give him the chance to stop that pandemic, but that he blew it for reasons that are, in retrospect, beyond ridiculously selfish and silly. LLoL has become used to everyone having their own narrative of who to blame for the pandemic disaster that happened. Most people LLoL met are so set in their thinking that they are not willing to entertain a different possibility; and if they would be willing, then they would lack the expertise and experiences in the diverse areas that are essential for understanding all critically important observations that LLoL made. LLoL’s suspected causality chain isn’t what everyone has to believe. It’s what LLoL happens to believe, based on many long and detailed analyses, done several times, because, frankly, LLoL wouldn’t believe it himself, if he hadn’t lived his life and done the analyses himself. So, welcome in the club of the doubters.

Do I have to believe in God to take the Matheo Study Series seriously?

No. The series writes claims about God and the world in ordinary logic so that anyone — believer or not — can check whether they hold. Its striking finding is that several traditions, written formally, converge on a shared structure and disagree in locatable places. It turns out that the Pan-En-Theology (“God is in all things and so much more than all things”) described in Matheo-b11 is remarkably well supported across diverse faith traditions. If indeed true, then every aspect of the visible world is somehow also an aspect of God, albeit without reducing God to the visible world (that would be what pantheism claims, without the “en”, ancient Greek for “in”). Such a definition of God reduces the question about “believing in God” or “seeing God” to the much simpler question of “believing what we can see and understand about our world”, because that is a part of God too. Hence, any scientist studying nature is thereby by definition also studying an aspect of God. Hence most people asking that question most probably already believe in God in one form or another. Does that make them experts in theology? No. Do experts in theology always believe the right things about God? No. Look at how the Prophets of Israel have been critiquing the theological experts of their time. LLoL has long intuitively held that the “New Testament” does not have an equivalently long sequence of prophetic books at its end, because God doesn’t like to sound like a broken record. Anyone interested can go look up the critiques offered in the Hebrew Bible. They apply remarkably well to any later scenarios for reasons deep in the structure of how innovation engines work. Unfortunately, for LLoL, they apply equally well to LLoL’s own life. If that conclusion surprises you — on any side — you are in good company: it forced LLoL to re-evaluate a great deal too. So do not take it on anyone’s say-so; the formal model is laid out to be checked, and broken if it is wrong (PET — Matheo-b11 intro; full formalism in b11-form).

See the Matheo Study Series.

What do h_star, h_dark, and h_zero mean?

They name a seat, not a rank: h_star is whoever holds the most decisive influence at a given moment and makes the right calls; h_dark is that same seat, failed; h_zero is the deliberate choice to carry the risk of failure with the hope to succeed for everyone, in the open, inviting scrutiny. Full definitions and papers are in the glossary.

What is a “MADI” and how is it related to the Mahdi or a messiah?

A MADI is a Mutually Assured Destruction Inhibitor — someone who works to effectively defuse paths to mutual annihilation. It is a function, not a title. LLoL is openly seeking to be a MADI to avert accidental nuclear winter. How such a MADI function may possibly relate to other prophetic roles is genuinely hard to determine — a question LLoL defers to dedicated eschatological work in ResearchCity (see below) rather than pronouncing on it himself.

Read on only if you care about details of eschatological complexities.

The Mahdi is a prophetic role in Islamic eschatology, expected to bring peace to the world by fighting the Dajjal, the greatest deceiver. In that sense the Mahdi is a messiah, but not the Messiah. According to Islam and Christianity that is Isa = Jesus = YhowShua (see Yas).

Figuring out reliably who instantiates such prophetic figures will likely require no less than at least one dedicated research stadion for eschatological studies in ResearchCity (like the STa4-REV stadion I envision). Any of these prophetic roles could be anyone, including you not yet realizing it’s you. But this should not make a difference to what we do if we stay gentle kind reasonable because the latter is the key to everything in any case.

Any responsible student of eschatology realizes eventually how hard it can be to reliably identify actual persons with prophetic roles predicted by ancient scriptures. Some even think it’s impossible or futile and give up altogether (as LLoL did long ago). However, it’s complicated because humans are wired to be “pattern-matchers”, whether they want to or not. And some of the patterns in ancient prophetic writings are derived from such long-standing observations and deep insight into how the world works, that they are only dismissed at one’s own peril. Therefore, these eschatological questions will not disappear because some think they are inconvenient, incomprehensible, or irrelevant. The future of any group lies in its eschatology and visions of the apocalypse keep shaping official policy decisions. In fact, it doesn’t take much surveying to realize that a significant fraction of many people’s work, worry, and wealth continues to be invested into resolving relevant eschatological questions and fears. The miners of old had a canary in the coal mine to placate their fears. For as long as the canary was alive, they knew that they were safe from invisible toxic gases that they knew could kill them. Once the canary was dead they knew they had to evacuate the mine. What LLoL proposes to become is such a “Canary in Earth’s Mine”, allowing anyone to easily check if ResearchCity is still on track to solve the world’s hardest existential challenges. These include averting accidental nuclear winter as well as developing a mathematical theology for gentle kind reasonably reconciling the various eschatologies that are otherwise set to blow up the world by mistaking those for enemies who in fact are allies. The story of godly King Josiah is a lesson in point. He died in a needless battle by “friendly fire”, because he failed to see that the real enemy was in the wrong assumptions he had been making about some other king. That battle near Megiddo arguably gave Armageddon its name. LLoL eventually summarized it as this: Every Armageddon-disaster is triggered by a Datageddon failure to connect the dots that matter most. One may as well say: every failure is a failure to understand. Such considerations have brought LLoL to strongly doubt that it will be possible for any individual or small group of scholars to reliably resolve any of these endtime riddles unless the world’s leading experts, supported by everyone else, work together on these questions in a ResearchCity stadion like STa4-REV. LLoL envisions it to create the most complete memomics alignment of all ideas about the end-times that can be constructed (a technical challenge and design-test for Evolvix). This enormous alignment (leveraging the tools of meta-genomics) may hold important clues about which patterns were seen by whom, how they have been playing out repeatedly in history, and what we can learn from them to avert the mistakes of the past.

Hence, the complexity of eschatological identifications precludes simple answers, because many sages agree that anyone could make the ultimate difference in the fate of the world. For example, what if the following fascinating account expresses some deep truth by claiming that, in effect, everyone is some form of a messiah, called to improve the world in unique ways? See at Ancient Jewish Wisdom: The True Identity of Moshiach.

LLoL can neither confirm nor deny many of the claims that are readily made in eschatology, not because he doesn’t have some suspicions of his own, but simply because his mathematical theology hasn’t caught up yet. That is why he is calling for a ResearchCity to #AuditTheMath for studying these questions. To ensure that LLoL doesn’t derail the good ideas in that effort, LLoL applies openly for the job of MADI, asking to be watched, because he realized how easily any mis-analysis in “us-vs-them” discussions can descend into nefarious suspicions made up by nothing out of nowhere. LLoL has been called by nothing to stand against such nonsense.

How is AI used on this site, and is that disclosed?

Yes, openly. Some text and summaries were drafted with AI assistance (Claude Opus 4.6-4.8 at max effort), always under LLoL’s direction so far. All scientific claims and final decisions are LLoL.

His Matheo-b19 study documents this collaboration in unusual detail, and a companion study (Matheo-b21) works through when AI co-authorship would be earned. Transparency about AI use is treated as part of the integrity of the work and LLoL has gone to great length in trying to document who did what. This effort has shown LLoL that current cutting edge AI tools like Claude are still lacking the reporting infrastructure required for reliably separating who did what. Hence, some of the early pages on this site very clearly spell out in great detail who did what. Later it became clear that if LLoL really wanted to get Balospe.com launched in a comprehensible way (and not merely as a “filedump” of his earlier Good News Pack MMv3) then he would have to find more efficient ways to engage with AI. That work led to the the Matheo Study Series MMv5 after which the rest of the front-pages were cleaned up in a way that LLoL directed to the best of his ability and AI Claude implemented. While LLoL attempted to check everything with great care, the practical singularity (see Matheo-b21) has made it impossible to do so. Hence, some pages may contain “silly errors by Claude” that LLoL failed to see, while other pages may contain “silly errors by LLoL” that he failed to have Claude check. The whole point of asking for buy in to raise funds to #AuditTheMath is that working through all the relevant details takes time even with AI and it is in the public interest that a site like Balospe.com be maintained and updated as best as technically possible. That mandate is impossible for any one person to fulfil. Therefore, LLoL is asking the rest of the world for help. LLoL got the baby so far. Now he needs help with raising it. As a result, LLoL classifies the overall quality of Balospe.com still as a MockupModel (MM) until further notice.

Is this affiliated with a government, church, or company? Is it after my money or data?

Balospe.com is an independent project by LLoL, Laurence Loewe of Laodicea, developed currently (to start simple) as his sole proprietorship DARL, the Datageddon Armageddon Research Lab Loewe. Past tool development and research was supported by the National Science Foundation (the NSF CAREER grant to Laurence Loewe that funded his Evolvix research), and the University of Wisconsin-Madison, which provided generous startup funds for his lab when he was hired as an Assistant Professor in Medical Genetics in 2010. Without their generous support and that of many others over the years, here and there, Loewe’s Evolvix research would not have been possible. Realizing that he wouldn’t be able to follow his most important long-term visions for Evolvix at his dream job (in WID at UW-Madison), Loewe decided to leave UW-Madison by the end of Dec 2019. He continued his research in his “American dream house”, where his independent marathon to avert existential disasters started. Still lacking any form of external support, but accumulating ever more ground-breaking results, he decided to keep going, because he thought that averting existential disasters like accidental nuclear winter was more important than his temporary discomfort. This website is offered “as is”, as the result of LLoL’s marathon. It’s far from perfect. But it’s independent in its search for finding gentle kind reasonable solutions for today’s world. Either the world will accept LLoL’s offer to scale up ResearchCity, or it will not. There is no paywall and no data harvesting here; the material is free to use.

Is this naive? Is LLoL just systemically deluded?

Maybe — and that possibility is taken seriously, not waved away. But notice what “systemic delusion” usually does in an argument: it is a label that lets someone avoid engaging (“no need to look — he’s deluded”). It can be aimed at anyone whose worldview you would rather not test. So the honest move is to make the question checkable instead of trading the accusation back and forth.

The one difference that actually separates a hypothesis from a delusion is corrigibility. A clinical delusion is a fixed belief held regardless of evidence, with absolute certainty, immune to disconfirmation. This project is built to be the opposite: every load-bearing claim — starting with the worse-than-1-in-40-per-year forecast — is published to be broken (hence #AuditTheMath), is stated with explicit uncertainty (“likely wrong in more ways than I have hairs on my head”), and is calibrated to documented Cold-War near-misses rather than to private conviction. Show where the math fails and it gets corrected or dropped — that responsiveness is exactly what a delusion structurally cannot do.

The mirror-image belief deserves naming too: “accidental nuclear winter is nothing to worry about” is the comfortable one that never has to test itself. One of these two beliefs can be proven wrong tomorrow by checking a number. That is the difference between a hypothesis and a delusion — so check the number.

How can I check any of this for myself?

That is the whole invitation: #AuditTheMath. The formal papers are public, the model code and data for the b19 study are archived, and the pages ask you to show where the math is wrong rather than to take anything on faith. Start with the Matheo Study Series — and for how to audit (what to check, where to send a refutation, and what mathematicians, lay readers, and teachers can each do), see the Audit the Math page, which carries its own FAQ for the questions specific to auditing.