b32 — Launch-promotion prompt-set: execution (5 prompts + newsletter reframe)#

Mode: EDEN. Effort: Max (from .claude/ files; reported at session start).
Drafter: Claude Opus 4.7 Max (dv_ClaOp47Max), 2026m05d28_13h54.
Context: Delivers the writing prompt-set LLoL requested on the GoFundMe-
launch home stretch. Supersedes the b31 request — LLoL flagged b31 as
opened for a now-moot tangent (the item-3 disclosure decision is absent from
this prompt); only b31’s #6/#7 location notes were reused. Continues b28/b29.

Redaction note (this llog is a public, indexed page)

LLoL’s give-away clarification (§2.3) names specific media outlets he is weighing and a deliberately cross-spectrum pair of politically-charged example topics — all pre-decisional musings, not commitments. Publishing those specifics verbatim on a public page could misrepresent them as settled and invite needless controversy. Per the b25 §10 / b31 precedent, the specific outlet names and the charged-topic pairing are lightly abstracted (marked [abstracted]); the options, the reasoning, and the urgent-need principle are recorded in full. LLoL can restore any of it. The Vatican/Anthropic detail is recorded but marked unverified.

1. Verbatim prompt (main)#

You are on the home strech for finishing an urgent GoFundMe launch and need to get everything set to polish the final language.

Please read the AHA/buy-in and the DICT participate for basic definitions on the entities discussed here. Then read the last 3 blog posts (dung beetle, eid, auction). Maybe it helps you to read relevant parts of past llogs

source/matheology/hell/ll/infra/b/28 , 29, 31

for context.

Your last session led to the clear disentanglement of a 1- fast launch to meet the hard June1 morning deadeline by calling for All Stadia Backers (email me directly, pay storage company directly) 2- a simultaneous public launch fo the gofundme for averting the 2nd auction and for starting the 80K-800K-8M$ launch until a more permanent buy-in solution can be found. 3- a Sojourners subscription email page (see Newsletter).

Here I repeat the relevant parts of my last prompt that wasn’t completed for reasons that don’t matter here.

Please creat either writing promts to produce the following documents (if that leads to better quality) OR prduce them right away if that is easy enough for you. (my guess is a context for each prompt would be better):

1. A new blog entry that explains 2026-05-28 the realization that a direct ask is necessary due to banking….timelines… (pointing to the related blog-posts, explaining a brief variant of the dung-beetle eid auction urgency. potential image: I’m basically in the same hard spot where Jesus was when asked if he’d pay the “temple tax”; he told peter to throw out his rope to catch a fish, predicting that there’d be a gold-coin in the fishes mouth, which Peter should then give for both of them to pay for the temple. So, maybe the ask explained here (later to be used on my to be set-up twitter account) is this: who will be the FiShFu (Fiduciary Sharing Futures) who has enough funds and is willing to buy in by becoming an All Stadia Backer for the ResearchCity vision of averting accidental nuclear winter? Explain the FF+1600 email (with explicit templates laid out like in FF footers) so it becomes a mouseclick by a rich person who is willing to send me EVERYTHING I need (including the “I will not influence clause” and the SD3a,b Supporting Doc links, so they can study the respective legal side of it if they wish to. (note that I’m not an attorney, but I learned (while selling my house) that anyone can apooint anyone else as “attorney in fact” (or was it advocate in fact… or somthing like it). My SD3 has the right language I think. A way to frame this as give away 50%: would be this: the pope currently struggles to help refine AI towards respecting human dignity: My 50% giveaway woudl be to whatever vatiacan initiative there may exists for getting on top of the AI problem (I don’t know where to direct those funds and if such an initiative exists, but I can mark those funds for going there ASAP or direct donors for doing so. Note that any of these 1600 All Backers counts towards their max; for now I must ask people to keep track of their own limits, because I don’t have the infrastructure to keep counting. But it should all be clear in this blog post, so that anyone reading this can do all it takes to get the auction stopped.

4. It may not need a separate session to come up with the changes required for the “Newsletter” section of the website, which currently has some basics online, but would likely much benefit form some revisions or tweaking in light of all this; not least because it ought to become, along with the links above part of what becomes linked into the new modified Buy-in page.

5. Hence reate a prompt for revising the Buy-in page on the few tweaks that need tweaking or adding links in light of all this (albeit WITHOUT destroying the tons of careful work that already went into this).

2. A new prompt for a new context to define the best GoFundMe story text based along the lines we discussed in light of this session. Make it as good as possible so I have to do as few edits as possible. Refer to the urgency of the acution 1 and say that there’s an auction 2 coming up in 2 weeks, hopefully that can by then be settled through GoFundMe funds….

6. Create a prompt for creating an idea-capture infrastructure for the stream of “this and that is a good idea for a blog post - for a tweet - for an Newsletter - …(anything else to do social media wise?)” Currently I sense that there are tons of these ideas, buried in the site.

7. Maybe there ought to be a list of potential people to email/tweet … Please propose a location - or create a prompt together with 6. for collecting such ideas in a central place.

Any questions?

My lastest copy of the GoFundMe Story text is below (I’m not sure if you kept a more recent copy with newer edits that’s easy to find; else go with the below as starting point): [reproduced in full in §3 “LLoL’s latest GoFundMe draft” — not duplicated here to avoid a double-copy in one llog.]

2. Verbatim follow-ups #

2.1 Mid-work clarification #

Do not overvalue the b31 llog; it was opened after I asked a tangential question that no longer matters in this context. It has a few points though on how to do points 6 and 7

2.2 Q2 clarification (sprint scope / prompts-vs-produce)#

Q2 is easy: whatever will give me the best wording in the best time. Running a fresh session at your end is quick; review time of tons of less than optimal drafts from you at my end takes forever. Hence I’d tend to: give me in-depth prompts for best quality, so I don’t have to review as much, but if something is simple and you don’t need a new prompt, then, of course let’s do it in the same session. Is that clear enough?

2.3 Q1 clarification (the give-away) — specifics abstracted per the Redaction note #

Q1: that may be harder.

a. I could stay entirely non-committal. The disadvantage is that people don’t know what “worthy causes” means for me and I am simultaneously gloriously non-committal and also non-divisive, but at the same time Popper-unfalsifiable. That tends to move me to commit to something, even if it’s bound to upset someone. These days it’s impossible to believe or do anything without triggering SOMEONE’s conspiracy theory.

Hence I was toying with option b. That would be e.g. GoFundMe 50% go to struggling investigative journalists who have been hit hard by the “4 horsemen of the media apocalypse” (as someone called it in 2018, predicting irreparable damage to the US news landscape within 5(!) years) - hence me trying to stop the butchering; whether I donate to [abstracted: a specific left-of-centre outlet] or support [abstracted: another specific outlet], whatever I pick will be seen as divisive by someone. Likewise, for the storage materials, I thought about supporting [unverified: a Vatican AI document calling for human dignity (“Humane/Dignitas…”?), at whose launch “someone from Anthropic spoke”] to build momentum for a more humane AI future; but then you flagged it as “Vatican is divisive”. I know. And I don’t even know if they need anything for AI research… or maybe if my donation would be a kick in the butt to do something about it.

So option c. is keep it open by setting up pages where people can apply or propose worthy causes to evaluate. For example: [abstracted: two generally- worthy large institutions] are generally worthy, but unless they face hardship RIGHT NOW (or else have to fire-cut-services…) then I’d [not] want to help. BUT if they do a strategy of “let’s build an endowment portfolio… so we’ll be secure in the long distant future…” - Then I’d rather say, tell me when you need something urgent. I don’t want to feed the banks indirectly by supporting the “worthy causes”.

option d. Maybe I can concoct some mix of worthy causes on the left and right, such that everyone can see themselves in them while realizing that they are not the only ones who struggle. For example: [abstracted: a deliberately cross- spectrum pair — one more left-resonant cause balanced with one more right-resonant cause], maybe make an open call for making AI HUMANE (as ResearchCity aims), while also supporting others in that space. If I’d try to support such a mix, I’d probably want to discuss it with you, maybe in a separate session that needs its own prompt.

Does all this help you move forward?

3. LLoL’s latest GoFundMe draft (canonical base for prompt #2)#

Reproduced as given (literal block preserves the [link: …] anchors and the draft’s own formatting). The TITLE and the first story paragraph are marked by LLoL as highly refined — touch only if really needed.

TITLE lock in:
Join Me to Cut Our Odds of Accidental Nuclear Winter

Story:

My name is Laurence Loewe, I'm a scientist, and I'm raising funds because I
don't want to die in nuclear roulette. My latest modeling indicates I'm more
likely to die in accidental nuclear winter than in a car crash (like most
people). That is, unless we learn to leverage the superpower of common sense.
Since 2020 I've spent all I have searching for and testing a way to do exactly
that. By serendipitous accident or divine providence (you judge), I found a
narrow escape hatch: working toward scaling up a "ResearchCity" in 7-8 stages,
whose mission is to take on the existential threats that fall through the
cracks because they're no one else's job. I can't build it alone. I'm not
pretending to be selfless --- but I have a dream of a mathematical system of
checks and balances that can keep me and ResearchCity honest, with your help.
It starts with #AuditTheMath, and I need your buy-in to make it happen. Will
you help me change our odds of survival?

(ABOVE IS HIGHLY REFINED, only touch if REALLY NEEDED and then explain)
(BELOW is close to your last draft...)

Nuclear roulette is unwinnable. There is a way out --- and you can check the
math yourself.

A model calibrated to real Cold-War near-miss data puts the median wait to an
accidental nuclear winter at about 19 years --- roughly a 1-in-40 chance every
single year. Almost no one's full-time job is to take that seriously and do
something about it. This campaign exists to change that.

The ask is small on everyone's purpose: about ~$8 per year per person ---
little over 2 cents a day of everyone working together --- to sustain
independent review of the math behind a proposed way out. That ~$8 is not a bet
that I have the answers. It is a vote of trust *for the work to be audited.* So
please: don't trust me. Audit the math. [link: see the model (SD1)]

---

What your ~$8 actually does

It sustains independent researchers who commit, under fiduciary responsibility,
to fairly review this work on everyone's behalf --- and it keeps me, the person
who spent 7+ years building it, in the race long enough to finish presenting it
for that review. (I'm Laurence Loewe, a research scientist; more about me and
the work is linked below.)

You are not buying answers. You are funding the review that tests whether the
answers hold. If the math fails that test, that is a result worth having too
--- a confirmed flaw is worth more to me than a thousand polite agreements, and
refutations are genuinely welcome.

---

Why only ~$8 --- and why you literally cannot give a fortune here

Most fundraisers want you to give as much as possible. This one is the
opposite, by design. There is a hard per-person cap, and the reason is simple:
so that no billionaire, no foundation, and no government can buy this campaign.
Work meant to speak honestly for *all* sides cannot be owned by the deepest
pocket in the room. The cap is the feature, not a bug.

GoFundMe cannot enforce that cap for me, so I am asking you to keep to it
yourself. And if ~$8 is nothing to you and you would like to do more, the most
valuable things you can give are not money:

- Sponsor people who can't give. Today only about 1 in 8 people worldwide even
has a credit card; the other 7 are shut out of funding a solution that is
arguably in their own interest. You may cover the buy-in of others who
currently cannot.
- Audit the math yourself, or send it to one expert whose disagreement would
carry weight.
- Tell one person who would care. That is often worth more than the money.

And half of everything raised is given away --- 50% goes to other worthy causes
(starting with investigative journalists pushed out of their work). This is not
built to make one organization important.

---

Is this "for me"? Honestly, yes --- and for everyone, in the same breath.

Averting accidental nuclear winter is the one place where not wanting to die
myself and not wanting anyone to die are the same wish. I'm not pretending to
be selfless; our interests line up here --- and the cap and the 50% give-away
are there to keep me honest about it. (For the record: contributions come to
me, an individual researcher, because there's no nonprofit yet --- building the
right one is part of the work.)

---

Why you can trust this --- and why I am not asking you to

I would rather show you than tell you:

- It is all auditable. The complete case is public. Bring your most skeptical
expert. [link: balospe.com]
- I have skin in the game. I invested everything I had and maxed out every
credit I could. As I write this, my life's research --- about 2,000 square feet
of it, including hard-to-find nuclear-fallout research I inherited from the
geneticist James F. Crow --- sits at a live public auction. You can see it for
yourself. [link: the live auction]
- I will not game it. The cheapest way out would be to quietly bid on my own
units at a knock-down price. I will not --- that would cheat a storage company
that has been remarkably gracious. The honest path is to raise what I owe, pay
it, and ask them to call off the auction.

The first auction's deadline is Monday, June 1, 2026. The first thing your
buy-in does is help keep this irreplaceable seed material from being scattered
before then. But the materials are only the beginning --- the point is the work
they feed.

---

Who this is for

If you came for the nuclear-winter math: you are in the right place. Everything
above stands on its own --- no belief required. Science is the truth-obsessed
tradition, and it is welcome at this table as a full equal.

If you come from a faith tradition: there is a deeper vision here too --- a case
that reads a concrete survival logic out of older texts, and argues for a
periodic, well-organized renewal (a Jubilee-based reset) for a whole
civilization. It is offered as a lens, never as a gate. There is a personal
reflection on it in a blog post [link: "Two Sons, Two Altars"] --- follow it if
it speaks to you, skip it if it does not. The ~$8 ask is exactly the same
either way.

---

The bigger picture, in one paragraph

The longer-term goal is to scale up a "ResearchCity" --- independent research
aimed squarely at the existential problems that fall through the cracks of
every existing institution precisely because they are no one's job. Think of it
as a *Ministry of Nothing*: the catch-all for everything the powerful wave away
as "that's nothing" --- right up until it isn't. The full scaling math is on
the Buy-In page. [link: balospe.com/buy-in]

---

Where the money goes (and an honest caution)

This GoFundMe is the start-up instrument --- quick, trusted, well-suited to a
one-click ~$8. It is *not* the permanent home. Within roughly 30-60 days I will
add channels that fit a capped, radically transparent, give-half-away model far
better --- most importantly a public, line-by-line ledger so anyone can watch
the 50% give-away actually happen, plus a way to support the ongoing audit over
time.

The honest caution: this is research funding for an unproven framework that is
still in active drafting. There is no guaranteed outcome. Your ~$8 supports the
*work of review*, not a finished product. I think that work is worth about 2
cents a day --- but I'd genuinely rather you didn't give unless you're convinced
it is. You get to decide.

---

Where this is going --- and why this goal is only the first rung.

Let me be honest about both the size of the dream and the smallness of this
first step. This is Stage 0: the seed. From here the plan is to stand up an
independent, fully transparent funding platform, then scale a "ResearchCity" in
7-8 stages --- beginning with a founding team of about 150 people [link: the
plan] and growing toward work aimed squarely at the existential threats no one
else is paid to watch. The full arithmetic --- what 10,000, a million, even a
billion people each giving ~$8 would sustain --- is laid out in the open [link:
the scaling math, on the Buy-In page]. I'm not asking you to fund the whole
forest; I'm asking you to help plant the seed, where you can audit every step.
This is deliberately a *starting* goal --- I'll raise it in stages as we grow,
and move to platforms better suited to transparency and tax-deductible giving
as soon as we can stand them up.

---

How to buy in

1. Decide what you can happily spare. About ~$8 is the suggestion; less is
genuinely welcome if money is tight; please keep to the cap if you would give
more.
2. Contribute below.
3. Tell one person who might care.

Then, whether you have ten minutes or a lifetime of expertise: audit the math.
Don't trust me --- check it.

[links to place at the foot: Audit the math (balospe.com/audit-the-math) ·
The science / the model (SD1) · Buy-In, full framework (balospe.com/buy-in) ·
The live auction]

The 7 links to wire (the [link: …] anchors): see the model (SD1) -> the SD1
page (or crisis/science); balospe.com -> home; the live auction -> your
2026-05-26 auction post; "Two Sons, Two Altars" -> your Eid post;
balospe.com/buy-in -> Buy-In; the plan ->
balospe.com/action/jobs/153-fishfus-job-positions; the scaling math -> Buy-In.
(One check before launch: don't insert the Eid/auction post links until those
posts are actually published, or they'll be dead.)

4. Session start + clean-slate #

Effort: Max. Mode: EDEN (read from .claude/effort-level + .claude/mode; reported at session start; LLoL did not change them).
Read for context: AHA/buy-in-types.md; source/dict/participation.rst; the b28 prompt + b31 llog; the three blog posts (dung-beetle, auction, Eid); source/buy-in/index.rst; source/action/newsletter/index.rst.
b31 weighting: per LLoL’s mid-work note, treated as superseded except for its #6/#7 location notes. The content cautions below are this session’s own EDEN reading of this prompt, not carried over as b31 dogma.

5. The two forks + resolutions #

Q2 (sprint scope) — RESOLVED by LLoL’s rule: default to in-depth prompts (best wording, least review burden for LLoL — fresh sessions are cheap on his end, reviewing many mediocre drafts is not); do simple things in-session. Mapping applied:

In-session now (simple): #4 newsletter (LLoL: “may not need a separate session”).
In-depth prompts (fresh sessions), run order: #1 blog (FIRST — June-1 critical) → #2 GoFundMe story → #5 Buy-in tweaks (surgical) → #6+#7 (combined) → give-away framing (post-launch).

Q1 (give-away) — RESOLVED into a path. LLoL leans against pure non-commitment (a, Popper-unfalsifiable) and toward committing to something. The strongest ZION option is (c): commit to a falsifiable CRITERION, not a divisive RECIPIENT — half goes to causes in urgent need now, not endowment-building. LLoL’s “I won’t feed the banks’ endowments” rule is the Jubilee anti-hoarding ethic applied to giving — a differentiator. Decision:

Launch pieces (#1, #2) use the (c) criterion as the spine, with light illustrative examples allowed in the blog’s personal voice (flagged as aspiration, not commitment); no named recipient, no Vatican on public pages.
The fuller (b) named-recipient / (d) balanced-mix design + the propose/ apply process + the fact-check of the Vatican/Anthropic detail (unverified; must be web-checked, not asserted) → its own post-launch give-away prompt (LLoL anticipated wanting a separate session for d).

4-option record preserved verbatim (abstracted) in §2.3.

6. EDEN classification #

Grey Edge GE-1 (the launch register). One narrow honest path exists, ringed by BABL look-alikes. Holds when the post stays: capped, transparent, surrender-not-triumph, open-handed, general-give-away. Breaches into BABL via any of: the uncapped rich-saviour, the named/unkeepable give-away promise, the messianic self-frame, or letting the whole launch ride on a June-1 rescue.

Green Meadow GM-1 (the production mechanics). Many non-trap ways to produce/ sequence the artifacts; count = 4 (the Q2 options offered). Chosen: the prompts-default mapping above.

BABL Dangers flagged + how handled (HUMANE requires stating these):

Temple-tax image — handled: framed as “the spot Jesus faced”, never “I am Jesus” (encoded as a hard caution in prompt #1).
Vatican give-away target — BABL (speculative + divisive + unkeepable public promise). Handled: kept off public pages; routed to the give-away prompt with a verify-don’t-assert instruction.
“attorney-in-fact” / POAATAD — handled: no invented legal language; kept behind SD3a/b links; prompt #1 adds a “have a real lawyer check SD3 before wiring a large sum” line (honesty that builds trust).
“send me EVERYTHING I need” / single-rich-patron — reconciled, not rejected (see §7): the cap + “I will not influence” clause make it on-strategy; June-1 money flow set to direct-pay / fastest-verified-route, not “to LLoL”.
Deadline-panic over-reach — handled: prompt #1 explicitly forbids letting the launch hinge on saving June 1; the post is held with open hands.

7. Key finding — the FiShFu ask is ON-strategy (the cap reconciliation)#

LLoL’s worry-shaped instinct (ask a rich “FiShFu” to fund things) does not contradict “no billionaire can buy this campaign” — because of the cap. Even the largest single buy-in is bounded by the absolute ceiling already on the Buy-In page (~$716,800 = $8 × 1,600 stadia × 7 yr × (1 self + 7 sponsored)), and the backer signs the “I will not influence” clause. So a FiShFu steps fully into the cap (enough to clear the storage debt and seed Stage 0) without buying influence. This is rail 3 (“All-Stadia Backers”, FF+1600xN) in AHA/buy-in-types.md, braided with rail 2 (direct-pay-the-creditor) for the June-1 emergency. The blog must make the cap + the clause loud and early so the ask reads as stepping into the cap, not buying the campaign.

8. What was produced this session #

Item	Form	File (`b/32/`)	Note
#1	prompt	`b32-prompt-1-direct-ask-blog`	Run FIRST. Direct-ask “FiShFu” blog; June-1 critical; held open-handed.
#2	prompt	`b32-prompt-2-gofundme-story`	Final polish of LLoL’s draft (§3); adds auction-2; give-away criterion.
#5	prompt	`b32-prompt-5-buy-in-tweaks`	Surgical additive link/consistency pass; do-no-harm guardrails.
#6+#7	prompt	`b32-prompt-6-7-idea-capture-contacts`	Post-launch; idea capture + (private) contacts; privacy-flagged.
give-away	prompt	`b32-prompt-giveaway-framing`	Post-launch; designs (c)+(d); verify Vatican/Anthropic, don’t assert.
#4	in-session edit	`source/action/newsletter/index.rst`	Reframed as the Sojourner lane; added Back/Audit/Tell rails + links.

9. Summary + recommendations (decision-relevant)#

Run #1 immediately. It is the only channel that could affect June 1 (money must clear ~Fri 2026m05d29). But the realistic odds of a wealthy reader acting in <72 h are low; the post’s durable value is the second auction, the GoFundMe, and the Twitter seed. Hold June 1 with open hands (matches the published three-outcomes surrender) — do not let the launch ride on it.
Then #2, then #5 (after #1 exists, so it can be linked), then the two post-launch prompts (#6+#7, give-away) once the launch pieces are out — do not build them under the deadline clock.
Give-away framing: launch uses the criterion (urgent-need, not endowment); the named-recipient / balanced-mix / Vatican decisions are deferred to the give-away prompt (and need a web fact-check first).
Newsletter: reframed in-session as the Sojourner subscription; LLoL to review (it is a draft, NN_draft_v1).
Open item for LLoL: the storage-debt dollar figure is a [LLoL to fill] placeholder in prompt #1 (not invented).
AAA QuickRef: not updated — these are INFRA launch prompts, outside the HEAVEN b11–b18 scope of the AAA rule (indexed instead in the INFRA index).