Audit the Math#
Don’t believe a word of it. Check it.
The claims on this site are big enough that you should not take them on trust — not from LLoL, not from an AI, not from anyone. So the ask is the opposite of “believe me.” It is: break it where you can. That is what
#AuditTheMathmeans.
First — you do not have to be good at math
Worried that “audit the math” means you have to do the math? You don’t — and that is the whole point. Almost nobody can personally check 32 interlocking papers. I couldn’t either, for most of my life: I grew up math-anxious, and learned a little only by refusing to quit. (People who think I “know a lot of math” are mostly not mathematicians.) So if the word math makes your stomach drop, you are not the problem here — you are exactly who this is for.
There are really two jobs, and only one of them is likely yours:
If you can check mathematical models — wonderful, and rare. Everything below is for you: the load-bearing claims, the papers, and how to send a refutation. Read on. ⬇
If math makes you anxious — like most people, and like me, once — your job is not to audit equations. It is to answer one plain question:
The question for the rest of us (which is most of us)
Do you want to live in a world where the math behind life-and-death decisions is guarded like a Babylonian temple secret — readable only by a priesthood — or in a world where enough people work to make that math open, checkable, and shareable by everyone?
That second world is what your buy-in builds. Not every model can be made simple — but most could be made far more accessible than they are today. (The open model that runs accidental-nuclear-winter simulations is already shareable enough for a curious non-specialist to follow.) The few who can audit, will. Everyone else can fund the patient work of making the math serve everyone — instead of math that serves, first, the wealthiest (a wealth-defense reflex) or the most violent (the military-industrial complex).
The long-run aim, in five words: democratize modeling for gentle kind reasonable decision-making. Backing that is how a math-anxious person “audits the math” — by making sure it can be audited at all, by anyone.
Why it is worth your two cents: the cumulative chance of an accidental nuclear winter over time, built from real Cold-War near-misses. No math needed to read it — see it explained in plain language on the science overview.#
And the way out: the same model’s escape ladder — up and out of MAD, through testing true Jubilees, to a ResearchCity that stabilises the system instead of gambling it. See the science.#
The whole call on one business card. The full five-card Nano Flying Scroll Exhibit takes about three minutes — no equations required.#
The rest of this page is for the few who can, and want to, check the math themselves. If that is not you, you have already done the most important thing by reading this far — thank you.
This whole project is offered to be tested, not believed. The Matheo Study Series writes claims about God, risk, and the world in ordinary logic precisely so that anyone can check them — and so that, if the math is wrong, you can point to exactly where. Finding a fatal flaw here is not an embarrassment to be avoided. It is the single most useful thing you could do, and it is welcomed in public.
The challenge#
A handful of claims on this site are load-bearing — the rest leans on them. In plain terms:
Accidental nuclear winter is a serious near-term risk — and, under business as usual, a long-run near-certainty. A probabilistic forecast built from Cold-War near-misses (the RiskyMAD model, Matheo-b16) estimates roughly a 1-in-40 chance per year — a level of risk no airline or insurer would tolerate anywhere else.
Modest, coordinated action can change the odds. From a pandemic cut roughly 42-fold by widening “germ gaps” (Matheo-b19) to a game-theoretic escape from nuclear roulette (MAP, the Jubilee System).
Seven worldview traditions converge on a shared formal structure about the God–world relationship when their claims are translated into axioms — a convergence that was not designed, but emerged on checking afterward (Matheo-b11).
Claims of that weight deserve to be tested in the open, by many eyes, from many angles — not asserted in a closed room and taken on faith. The honest response to a big claim is not applause and not dismissal. It is scrutiny.
The decisive question — asked of this site too
Across traditions, the one question that separates a trustworthy guide from a confident fake is simple: “will you let others check your work?” (see Matheo-b18c). This page is that promise, kept. The papers are public, the models are described, and the answer to the question is yes — here is how.
What is actually being checked#
You are asked to check the math and the claims — not LLoL’s character, biography, or faith. The argument has to stand or fall on its own, independent of who makes it. The work is layered, so you can enter at whatever depth fits your time and background:
Layer |
Where to read |
What to look for |
|---|---|---|
The condensed visual claims (5 cards) |
Is the core argument honestly compressed, or does the compression hide a sleight of hand? |
|
The plain-language intros (age 12+) |
Do the everyday-language claims actually match the formal ones behind them, or quietly overreach? |
|
All 32 Matheo papers (across the eleven studies), reporting formal axioms and proofs from different perspectives |
Are the axioms coherent? Do the theorems follow? Where does a step get skipped? How do they relate to real-world observations? |
|
The quantitative forecasts |
RiskyMAD (b16, nuclear) · SGIR (b19, pandemic) |
Are the data, assumptions, and sensitivity analyses defensible? Would different assumptions break the result? |
The earlier work the 2026 papers were distilled from |
What needs to be refined based on the 2026 findings? Which core insights from 2025 deserve a separate study not yet in the Matheo Series? Do related insights affect the math core? |
➜ New here? The fastest honest overview is the five-card Nano Flying Scroll Exhibit — the whole call compressed onto business-card-sized pages, free to download and share.
The eleven studies at a glance#
One plain line per study — the shape of the corpus you are invited to check.
Study |
What it puts into checkable logic |
|---|---|
Seven worldview traditions, translated into formal logic, converge on one shared structure for the God–world relationship — with a built-in test (ax14) for any revelation claim. |
|
Why systems destroy themselves (the BABL algorithm) — and the seven-step self-correcting construction that escapes it, read off Genesis 1. |
|
Growth follows a never-ending seven-stage Hero-Journey pattern that keeps a person correctable — a coinductive inoculation against BABL. |
|
An innovation theodicy — why suffering tracks innovation failure — and the Jubilee economy that resets the system before it self-destructs. |
|
A formal clash: “God does not suffer” (divine simplicity) is incompatible with a God genuinely related to the world — and why that idea can be dangerous. |
|
The probabilistic forecast that accidental nuclear winter is a near-term certainty under business-as-usual (~1-in-40 per year) — and the MAP escape ladder. |
|
One agent always holds maximal causal influence at any moment — a theorem, plus a public, falsifiable test, and why that agent is safest when most accountable. |
|
Why every tradition’s defense against false messiahs can block recognizing a real rescue (the Recognition Trap) — and the practical call from MAD to MAP. |
|
A pandemic model showing that small increases in “germ gaps” can cut mass casualties roughly 42-fold — modest coordinated action changes the odds. |
|
The 2020 first-steps design of Work-Logic Cascades for stopping a pandemic — the work/rest (Shabbat-pattern) logic underneath the system. |
|
A framework for honest human–AI co-authorship in scientific publishing after the practical singularity (PraS). |
The series in one line. At any moment, one person holds the most influence — the superhero (h_star). The instant they stop listening, they curdle into the supervillain (h_dark), most dangerous right where they have stopped being able to learn. The only rescue is returning, in time, to real listening (h_zero) — letting reality, not their own certainty, have the last word. Destiny is that influence used well; fate is it gone wrong; the narrow door between them is the willingness to keep being corrected. The math works it out; #AuditTheMath is the open invitation to check it.
For the full list of all 32 study titles (audience variants and PDFs), see Matheo Study Series.
How an open audit could work#
There is no gatekeeper and no paywall. The mechanism is deliberately simple, so that it works whether one person or ten thousand show up:
The papers are public and citable. Every study has a stable
Matheo-bNNreference and a free PDF. Point to a line, a step, an axiom, or a figure.You send what you find. Use the FeedbackFlow (FF) link at the bottom of any page, or email
FF+audit-the-math@balospe.comdirectly. Questions, doubts, and especially refutations are all in scope.Findings are handled in the open. This project keeps an append-only audit trail of lessons learned (it is literally called HELL — Historically Experienced Lessons Learned). A flaw, once found, is recorded, not buried.
The review rolls forward as support grows. A handful of scattered emails is not a review of 32 interlinked papers. Turning incoming feedback into a systematic, navigable audit — indexed, cross-referenced, answered — takes sustained work. That is the next section.
Embarrassing ideas, tested and rejected, are not failures
They are evidence the system is working. The cycle is: propose ➜ test ➜ catch the error ➜ correct ➜ strengthen. A refutation that retires a bad idea early has done everyone a favour — so it is treated as a contribution, and credited as one.
Why not simply submit it to journals?#
The fair question: if the math is real, why not just publish it through ordinary academic peer review and let that settle it?
Three reasons. Traditional peer review is too slow for a clear-and-present danger; it is too quick to dismiss early-stage, cross-disciplinary work; and, honestly, there is nowhere obvious to submit a body of work this broad. There are of course enough open access publishers who will publish for a fee, but it is not necessarily clear how much value their reviewing would add over keeping the papers well-organized and continually reviewed at Balospe.com. Publishing with any niche-publisher that would take that many papers that are that diverse would likely be the equivalent of burying the work. The other strategy would be to slice up the work according to the respective fields it addresses. Strengthen each resulting paper into a substantial contribution for that field, albeit without the context of the other studies. While that is possible, this would scatter the work across fields and years, making it hard to find (without a website like Balospe.com).
Thus, LLoL fully intends to clear the academic bar eventually. But that goal must be held with caution, because of a difference that is easy to forget:
Writing about a ResearchCity is not the same as building one. Writing good books about existential disasters is not the same as the work of averting them.
So the review here is deliberately ordered:
The core math is reviewed first. There is no point acting on a flawed foundation, so breaking — or confirming — the core comes before anything else.
If the core proves too flawed to build on, the honest move is to write up why, properly and for the record, so no one repeats the mistakes. That is the right time for polished papers: documenting an instructive failure.
If the core holds, then in a race against accidental nuclear winter, acting on it cannot wait for every “i” to be dotted. Crossing the t’s is real work, but it is not on the most critical decision-path and can be perfected on the way. To stall a sound, life-saving move for the sake of publication polish would itself be irresponsible.
This is a discipline a careful scientist already knows: building a real ResearchCity will demand real-time decisions that don’t always leave time to “write the paper” — yet each must still be made with the care of someone who had written it.
Stage 0, Stage 1, and a race against time#
Reviewing the core math is Stage 0 work — together with the practical preparations that only make sense once the core looks sound. Only when enough competent experts judge the core to be sufficiently well-integrated does Stage 1 of the 7-8 stages of scaling up a ResearchCity begin. Without enough reviewing, there is no point in the practical Stage-0 preparations for Stage 1 — which is exactly why this campaign funds buy-in to review first.
However, note that the race in time against accidental nuclear winter allows for some reasonable distribution of concurrent work that can proceed independently. While reviewing core math remains on the critical path, additional buy-in may allow scaling up two additional teams that will certainly be needed once the math is cleared:
Stage 0 - prep team for Stage 1: This team assumes that the math is sufficiently reviewed to act upon it. Thus it starts with the practical preparations for Stage 1 of scaling up, such as looking for potential places, drafting various ways of how to interpret the architectural ideas in my initial drafts (see the ResearchCity overview), preparing application processes, setting up the necessary web presence for making Stage 1 as transparent as possible, and many more such things. Note that Stage 1 of ResearchCity’s scaling up has so modest space requirements that it would fit almost anywhere. This is not meant to be wherever ResearchCity will later end up being. Likewise for Stage 2, which may be in a different location from Stage 1 (as it needs a larger space). Stage 3 may or may not be in ResearchCity’s designated location. However, if Stage 4-7 are not on-site, then there is a serious gap in the vision described on the ResearchCity page. To avoid such misalignments, it is imperative to complete proper review of the core math first.
Support for open global competition to host ResearchCity: Every nation and district, from Afghanistan to Zululand, is invited to bid. Land will be bought at fair prices, but only with the willing support of those who live on it; clearing land by force disqualifies a bid. If there is no peaceful way to build it, there will be no ResearchCity worth building — and I will say so as plainly as the math does: any city built by force would carry the seed of its own ruin. But success depends on far more than geography. To explain these essential other factors, it may be critical for success to early on scale up a dedicated team to help interested host nations apply. It’s far from clear (and I don’t care) whether ResearchCity lands near a leading research center of today, or of the past (like Timbuktu), or elsewhere — as long as all requirements of Reality are met to allow ResearchCity to live long-term.
The gated stage ladder (Stage 0 to Stage 7)#
Scaling a ResearchCity is not one leap or some continuous growth. Instead it is a ladder of gated steps, each with defined protocols to ensure the overall essential quality is not lost in the process.
Stage 0 has two phases. It starts with the review itself — auditing the core Jubilee-math — which is the biggest gate of all, because if the math core fails, building a city to implement it makes no sense and all other Stages would not be different from any of the many mega-projects that failed.
Yet Stage 0 must also prepare for Stage 1 if there is to be a Stage 1. Hence, given the race against time (nuclear roulette doesn’t wait), it is possible to lay the groundwork for Stage 1 preparations while the review is still running until it can produce a clear No-go/Go signal. Only once enough competent, independent review judges the core sufficiently reliable to build on can Stage 0 do the last remaining preparations for starting Stage 1.
How exactly to divide up that work remains to be seen, but the same is true for each Stage as each depends on all tests to be successfully cleared in the one before.
Each Stage carries its own testing agenda: the full plans for all remaining stages are to be lived in a smaller model, resulting in a live re-test, with results to be refined, and most up-to-date drafts to be accordingly revised at every stage against everything learned so far. No stage starts until the previous one has held. Depending on whether Stage 0 is counted as a full Stage, that makes for a total of seven to eight gated Stages in all (Stage 0, then Stages 1 through 7), which is why the campaign sometimes says “7–8 stages.”
The auditing and re-reviewing is deliberately grueling. The standards are likely in many ways closer to preparing for a trip to the Moon than to typical “city planning”. For auditing the math underpinning a system this complex, I know no credible shortcuts that don’t cripple its mission. Early mistakes are the cheapest to fix and the most dangerous to miss, so the testing is heaviest exactly where consequences compound most. If a later stage uncovers a flaw the earlier review missed, the ladder is built to send the work back down — not to plow ahead on a cracked foundation. But that is only possible if the whole construction stays long enough in “purgatory” to be refined accordingly.
The completion of Stage 7 aims to have all 1600 talent stadia of ResearchCity operational. Whether they require an additional warm-up period to prepare for the Great Jubilee Race they are meant to facilitate remains to be seen.
Each Stage has 6 effective core months, each dedicated to exercising a certain defined core function that is essential to all innovation economies. Most likely it will take 1 effective month to prepare the setup, such that focus within the 6 months is not diluted; likewise, it will likely take another effective month of debriefing, post-processing, and writing up results, in order to ensure that the most important historically experienced lessons learned are not lost.
Therefore, each Stage will likely last 6-8 effective months; if possible in 6, great, if not, then 8, fine. Functional completion matters, yet, this is also a race against time. Therefore, it is at this point somewhat open how exactly the “effective months” translate into actual “calendar months”. However, the expectation is that these should not be too far off.
Why this needs funding#
Here is an important bottleneck. The evidence is not one tidy paper; it is a densely interconnected web — 32 study files (b11–b21), the Good News Pack behind them, and a lifetime of research materials they were distilled from. An audit at that scale is impossible without first organizing the information so reviewers can navigate it efficiently — indexing, cross-referencing, archiving, hosting, and sustaining the people who answer serious questions and route refutations to where they bite.
Left unorganized, even genuinely good evidence stays an unnavigable hairball that no busy expert can afford to dig through. Organizing it — and keeping it from being lost altogether — is itself real work that needs gentle kind reasonable support.
Funding is a vote for the audit — not a vote of trust in untested answers
The two channels run in parallel, on purpose. A contribution does not say “I believe LLoL is right.” It says “this work deserves to be checked properly, so let’s sustain the people doing the checking instead of starving them.” The math still has to earn its standing on its own.
The baseline ask is a voluntary ~$8 per person per year — capped to ~$8/stadion, so no one can buy influence, with half given away to other worthy causes. See how to buy in, or back the work without money.
What you can do#
Pick whatever fits your time and background. The most valuable contribution, at every level, is honest, specific, critical feedback.
For mathematicians, modelers, and other experts#
(theologians, logicians, epidemiologists, game theorists, policy people — welcome too.)
Attack the load-bearing claims. Start where the weight sits: the PET axioms (b11), the RiskyMAD forecast (b16), the h_star theorem (b17), or the SGIR pandemic model (b19). Look for an unsound axiom, a skipped step, a fragile assumption, a forecast that flips under a reasonable reparameterization.
Try to break it, formally. A counterexample, an inconsistency, or a citation that does not say what it is claimed to say is worth more than a page of agreement.
Send a refutation to
FF+audit-the-math@balospe.com. Be as specific as you would want a referee to be with you. Disagreement that carries weight is exactly what is missing.
For teachers, preachers, and other producers#
Stress-test the cross-tradition claims. The series says seven traditions converge on a shared structure and disagree in locatable places (b11, b18c). Is that fair to your tradition? Where is it strained, and where does it ring true?
Teach what you can understand as a worked example of testable thinking — how to state a belief precisely enough that it can be checked, and why “will you let others check your work?” is a question worth teaching everyone to ask of anyone.
Bring your students’ and congregation’s hardest objections. A good classroom or study-group can break an argument faster than a lone reviewer. Send what they find.
Your involvement is essential for translating the findings of experts into a language that beginners can understand. Most experts suffer from the “curse of knowledge”: they worked hard to understand something complicated and after having understood it themselves, they forgot their own struggles with understanding it (and are even less able to understand where someone else may struggle). Your expertise in explaining things and knowing where your audience may struggle as well as actual questions from your audiences are gold for experts, because it may allow them to refine the way the math is presented so that it becomes overall clearer for everyone eventually. But the struggle is hard and the road is long. Thank you for your patience!
For beginners (no mathematics required)#
Read the introductory pages.
Have patience with LLoL’s efforts to explain things better.
Buy-in to support the hire of people who can help explain better.
Pick things you do understand and can agree on to act on them: for example, it doesn’t take an advanced math degree to have a basic understanding of what “gentle-kind-reasonable” might be. Try to live that combination, so that all three conditions remain valid over the long term to the best of your ability to foresee. If LLoL’s findings are reliable, interesting real questions will emerge. Don’t ignore them. Try to find real, reliable answers, not shortcuts. Find others (teachers, experts) to discuss the challenges you encounter. Once Balospe.com becomes better organized, there will hopefully be a place to write in with questions in order to help you get reliable answers. Yes, many questions of beginners are easy to answer for experts (and that makes some experts sometimes sound like know-it-alls!); However, please don’t discount yourself: because you have something that most experts have not: you are free from the curse of knowledge! It means that you can walk right into an expert’s blind spot and ask the question that matters most, and which an experts has overlooked because they were too wrapped up with oversimplifying the world one way or another. Yes, experts oversimplify too. You are there to save them from that trap through your lived real-world experience. If ResearchCity is to work, it must find a way to serve all by finding ways to construct knowledge-pipeline-refineries that help all to learn what Reality is all about (with a capital R, because it’s greater than anyone, beginner or expert can imagine).
Forward a link for testing to one or two persons whose disagreement would carry weight — a mathematician, a teacher, a sceptical friend. One good referral can matter more than a hundred likes.
Audit the plain-language layer. Read an “intro” paper or the Nano Flying Scroll and ask: do you see empty persuasion or an honest argument? Where did it lose you? Which of your “obvious questions” did it not answer? Obvious to you may not be obvious to everyone, hence, if you’re willing to share your reaction please send it in. It’s what enables Balospe.com to improve.
Check consistency. You do not need a proof to notice when page A and page B seem to say different things. That’s either a mistake or something that has an explanation that should be reachable. It helps if you can catch such cases.
FAQ — about auditing specifically#
I’m not a mathematician. Is there really anything for me to audit?
Yes. Most of the work has a plain-language layer, and a lot of auditing is not about equations at all: does the everyday claim match the formal one? Is the argument real-life honest or merely empty words? Do two pages contradict each other? Anyone careful can help with that — and can forward a paper to someone who knows more and who can test more. And if you can’t directly help with that, your buy in helps others to do more of that work. The end product envisioned by LLoL is not only core math that satisfies the arcane exacting requirements of mathematicians, but that is also translated into a plain-language layer that is such that anyone can understand it without the need for a math-degree. Hence, both buy-in and engaging others in respective discussions helps.
What exactly am I being asked to check — and what not?
The math and the claims that rest on it. You are not asked to judge LLoL’s character, faith, or biography; the argument is meant to stand or fall independently of who makes it. If a claim only holds if you already trust the author, that itself is a bug worth reporting. LLoL is a flawed human being, like everyone, arguably more so, because he couldn’t have discovered the BABL traps without having fallen so hard for them. The reason for his call to AuditTheMath is that before seriously scaling up a ResearchCity, he must be certain that he didn’t fall for some complicated trap that he so far couldn’t see.
Where do I send what I find?
Ideally use the FeedbackFlow (FF) link at the bottom of the most relevant page, or email
FF+audit-the-math@balospe.com. Specific, critical feedback —
especially detailed, well-argued refutations — is the most welcome kind.
Do I have to pay or “buy in” to audit?
No. Auditing is free, open, and entirely independent of funding. Money is never a condition for checking the math or for being heard. Buying in sustains the separate work of organizing the FF material and the review so that audits can scale — it is a vote for the work to be checked, not a toll on checking it. Your buy in supports someone who listens who would otherwise either starve or have to look for other work.
Isn’t “#AuditTheMath” just a slogan?
It is meant as a checkable commitment, not a mood. The papers are public, the models are described, refutations are invited and as best possible recorded in the open, and the project’s own position (see Matheo-b18b) is that the safest person in any decisive seat is the one who submits to scrutiny rather than hiding from it. If the commitment is ever not kept, that is itself a fair thing to call out — here.
What happens if someone finds a genuine, fatal flaw?
Then the project has done its job. A refutation that retires a bad idea early is a success of the method, not a defeat — it gets recorded, not buried, and the person who found it is credited. The aim is reliable results over the long term, which is impossible without people willing to say “this part is broken.”
Flaws come in many kinds, not all of them touch the core of an argument. Some flaws can be easily fixed by refining the theory. A fatal flaw essentially exposes the core argument as either false or unreliable in a way that cannot be fixed. For a math model as complex as presented here, that will be worth writing up in the clearest possible way to help others avert the same traps that trapped LLoL.
If the core math underpinning ResearchCity fails, then no ResearchCity will be built on it as it would be irresponsible to do so. In that case, LLoL’s vision will have failed and he will find some way to wind down this overall operation, likely by taking a tiny aspect of it that is still deemed useful by others. The buy-in of everyone will be honored inasmuch as it will be going towards improving common goods in other forms than the original vision.
Broader questions?
For wider questions — “Is this a doom cult?” (No), “Who is LLoL?” (a scientist and believer in Reality), “Do I have to believe in God?” (Which one?), “Is LLoL claiming to be a messiah?” (No), “How is AI used here?” (as a challenge) — see the general FAQ. This page covers the questions specific to auditing; the general FAQ links back here for those.
A personal framing#
For LLoL’s own framing of why this campaign exists — told through a stumbling dung beetle, Aesop’s fable of the Beetle and the Eagle, and a very practical ask — see the blog post Prophetic Dung Beetle. Its plea is the same as this page’s: don’t take the dung ball on trust — check whether there is fertilizer in it.