Shape A — committed. Aggressive timeline. GCT goes live mid-July.
Pablo confirmed (2026-04-28 evening): Shape A is the trajectory — public-facing GCT identity, Spark Automations as applied-tech operation underneath. Timeline pulled in by 6-9 weeks vs the original draft. Stage 1: 2026-07-15. Stage 2: 2026-10-31. What follows is the committed plan, not a thinking document.
Where this all lives in 2031
The honest answer is "depends on whether GCT breaks through." That branch happens at year 2-3 based on whether (a) the manuscript gets published in a real venue, (b) at least one independent physicist engages substantively rather than dismissively, and (c) at least one Tier-3 prediction (Zeno Drive in NV-centers, mass ratios, etc.) gets tested experimentally. Three plausible shapes follow.
GCT-centric academic legacy
Pablo, mid-50s, physically still in Mexico (PdC + maybe CDMX or Valle de Bravo). Lattice Institute is 4 years old, has 2-3 co-founders, ~$500k-$2M annual budget, runs an annual conference, has 5-15 papers in peer-reviewed journals — including 2-3 critical responses, which is itself a sign the work is being engaged with. The manuscript is published as a book + as 12-15 papers. ~20 podcast appearances, 3 keynotes. Email list 20-50k. Twitter/X 50-200k followers. SA runs in the background generating $5-15M ARR; RUBISCO2 has shipped 80-200 hotel sites; neither defines you publicly.
SA empire
SA is a 15-30 person applied-AI shop with 4 verticals (fintech / creator content / industrial CAD / conversational AI), each productized from the original bespoke clients (AFI, Miguel, RUBISCO2, Horus). Diego is co-CEO of Horus side. GCT is a serious side project; the manuscript is published, gets cited 50-200 times, doesn't break through to Penrose-tier discourse. Pablo is a respected operator in Latin American applied-tech circles. Less leverage but more concrete revenue. Still a fine outcome — 95% of the practical value with 10% of the controversy.
Polymath Tier 4 most-likely actual
Known as "the person doing X" in 3 different rooms — physics (GCT), Mexican applied-tech (SA + RUBISCO2), Caribbean cleantech (sargassum). Less brand purity but more optionality. Closer to Wolfram (research + Mathematica + Wolfram Alpha) than to a single-track founder. Diego stays partner. Family stable. Spanish at LATAM events, English at Anglo ones.
Twenty holes between here and "no holes"
Brutal honest list. Tier 1-2 throughout. This is what would have to be true before any coordinated public push — across Pablo, SA, Horus, RUBISCO2, GCT, AFI, and Miguel surfaces.
- SA tax structure resolved — RESICO vs salary vs royalty (decisions §1). Going public with revenue and ambiguous structure is a compliance landmine.
- IMPI trademarks filed — Horus + Spark Automations (decisions §4). Currently exposed; a bigger company can register first and force a rename.
- Corporate entity for GCT — none exists. Manuscript IP is in Pablo's personal name by default. Public release without an entity is an attackable hole.
- Domain ownership clean — rubisco2.com, horuslovesvision.com, spark-automations.com. Should all be in named legal entities, not Pablo's personal Squarespace/Cloudflare account.
- Privacy policy + terms + LFPDPPP aviso on every public surface. Currently only the AFI proposal has a template; SA / Horus public sites need them too.
- Pablo personal brand doesn't exist yet. No
pablogonzalezacosta.com, no canonical bio, no headshots, no boilerplate. If a journalist asks "who is this person," there's nothing to point to. - Spokesperson voice — when Pablo is unavailable, who speaks for SA? Diego? Spark itself? Right now: nobody. That's a hole the moment there's inbound press.
- Case studies — Tere/Paola/Luis have great results but no permission yet to be named publicly. AFI and Miguel are too new.
- Pricing transparency or not — decision needed. SA has no public pricing, but if going public, this will be asked. Friend rates (AFI $1,500) become awkward against any public market rate.
- GCT brand — doesn't exist (the gct.html page is internal speculation). Needs lock before any release.
- Twitter/X presence — minimal/none for theory work. Building audience from zero on launch day is hard mode.
- Newsletter — none. The single highest-leverage owned-channel asset for an independent thinker. Should have been started 12 months ago. Can be started now.
- LinkedIn writing — sporadic. Should be one substantive long-form post weekly minimum if going public.
- Reference clients willing to be named — zero confirmed.
- Inbound contact pipeline —
contact@rubisco2.comis a SHARED partner inbox. There's no dedicated public-facing inbox with a documented response SLA. Press inquiry would land in a shared mailbox and get lost. - Reputational scan — what surfaces if someone Googles you today? The FB account thing this morning is an example of something exposed. Need to know what's out there.
- Backup of GCT manuscript in 3+ places — one disk failure away from catastrophe is not OK if going public. Need git, IPFS or arXiv timestamp, and a printed copy.
- 2FA on everything — including the GitHub repos that hold the manuscript and the SA / Horus infra. Pre-launch lock-down.
- Operating runway — going public increases burn (legal, brand, photography, travel, possibly hires). Need 9-12 months of buffer at planned spend, not 3.
- Revenue separation — RUBISCO2 should be its own entity for liability. Currently mixed with SA. If a hotel cleaning machine kills someone, GCT IP and SA revenue shouldn't be exposed.
20 holes total. Some are 30-min fixes (privacy policy template). Some are months of work (newsletter audience, reference clients). The ordered checklist is in §4.
What "release" means, and what's missing for the right combo
Definitions of "release" are not equivalent. Naming the fork before recommending:
| Release path | What it is | Time |
|---|---|---|
| (a) arXiv preprint | Manuscript posted publicly. Establishes priority. No approval needed. | ~2 weeks once LaTeX-clean |
| (b) Peer-reviewed paper(s) | Submitted to journal, reviewer rounds, eventually accepted. | 6-18 months per paper |
| (c) Book | Trade or academic press; or self-publish. | 18-24 months academic, 6-12 self-pub |
| (d) Public-facing release | Coordinated launch with press kit, podcast tour, social. | Pairs with (a) or (c) |
| (e) Experimental validation | Independent lab tests a key prediction (Zeno Drive in NV-centers, etc.). | 5-10+ years, Tier 4 |
Ten missing pieces for that combo
- Markdown → LaTeX conversion. Manuscript is currently markdown. arXiv requires LaTeX or PDF.
- Figures pass. 20-40 publication-quality figures probably needed at this scope. Manim is already in the stack — perfect tool.
- Citation hygiene. Every claim cited to existing literature. Explicit comparison sections vs IIT, GWT, Orch-OR.
- Falsifiability statements explicit. Each prediction needs an "experimental protocol that would falsify this" paragraph. Physicists respect that move.
- Independent pre-review. 2-3 physicists outside the circle willing to read and give honest feedback before public release. Highest-leverage missing thing. Trade-off: exposes the theory pre-publication. Mitigation: NDA + endorsement letter at the end.
- Code repo public. Lattice geometry simulations, mass spectrum calculations. GitHub under (future) Lattice Institute org.
- arXiv endorsement. First-time submitters need an endorsement from an established arXiv user in the relevant category. Need to identify who.
- Strategy on which paper goes first to peer review. Tier 3 lean: derivation-of-Standard-Model from zero free parameters — that's the headline result.
- Tier-labeled language → journal voice. The 4-tier system is great for internal clarity but journals expect "we derive X" not "Tier 2 claim that X." Need a translation pass.
- Manuscript backup hygiene. Per §2 #17.
Time estimate (Tier 3): 12-16 weeks of focused work for one person to take the manuscript from current state to arXiv-ready + 1 paper submission. That's a quarter at full-time, OR ~6 months at part-time around other commitments.
Two stages, dated, ambitious but bounded
GCT goes live
Single coordinated public moment. The single highest-leverage event we can engineer in 2026.
- GCT v1.0 manuscript on Zenodo with DOI — canonical archive (no endorser gating; arXiv pursuit deferred per Pablo)
- 1 strongest-derivation paper submitted to peer review — target: Foundations of Physics or Classical & Quantum Gravity
lattice.institute(or final name) live with simple landing page + manuscript download + biopablogonzalezacosta.comlive with bio, links, contact- Twitter/X announcement thread (10-15 tweets) on launch day
- LinkedIn long-form post on launch day
- Newsletter live with 200+ subscribers (built over the prior 8 weeks)
- 3 podcast outreach pitches sent — targets: Mindscape / Closer to Truth / Lex / TOE
- Manuscript timestamp registered on git + Bitcoin (cheap priority insurance)
- Lattice Institute registered as Estonian OÜ via Pablo's e-Residency (solo founder, fully digital, EU-domiciled)
Public-ready across the stack
Maturation + ecosystem. By Dec 31: published theory, foundation, public personal brand, profitable applied-tech company, named client roster, trajectory toward Shape A / C set.
- First peer-review round feedback received and addressed
- 3+ podcast appearances done or scheduled
- Book proposal in flight (academic press preferred) OR self-publish underway
- Newsletter at 1000+ subscribers
- SA + Horus public sites live with case studies (named clients, with permission)
- IMPI trademarks filed for SA + Horus + Lattice Institute
- Tax structure resolved (decisions §1) and corporate setup clean
- RUBISCO2 demo done (June 9), AFI shipped (Sept-Oct), Miguel pipeline running automatically
- At least 1 Spanish-language interview or feature (Mexico — own that)
Six decisions — four resolved, two still open
One more thing
You said this is a massive moment. The actual big shift isn't any single deliverable — it's committing to GCT as the public identity rather than the private work. That's irreversible in a soft way: once you're publicly the GCT person, you can't quietly retreat to being only the SA founder.
The gain is alignment between what you actually care about and what people see. The cost is exposure.
You don't have to decide that today. But the goal post above implicitly assumes you do. If you'd rather keep GCT private and use the next year to scale SA + Horus + RUBISCO2 instead, that's a different goal post — equally valid, much smaller risk, smaller upside. Worth naming the fork before we commit.