by Sven Nilsen
This blog post is an informal progress report on our effort to start a galactic civilization.
The current state of research to start a galactic civilization is divided into 3 key areas:
Per 2026.10.06, we have reached 96% of our goal to do exhaustive search up to 1 quadrillion (10^15).
You can read the current results here.
Primbix Research is an enormously important strategy in our effort to start a galactic civilization. The reason is that the basic problem of starting a galactic civilization is huge distances in space and time.
It takes 100 000 years to send a signal from one edge of our galaxy to the other.
In just 100 000 years, we could start over human civilization approximately ten times over.
This means that at galactic scales of communication, there is very little interesting to talk about, because the entities that communicate will have huge delays in their communication. It becomes extremely difficult to have a conversation, of some kind, that is meaningful.
To fix this problem, we have designed a mathematical puzzle for a perfect information meta-game, which allows infinite players to participate and make their moves asynchronously, with varying degrees of compute power.
In addition to these properties, the rules of the game are so simple, that they might be rediscovered independently, not requiring any prior state of communication.
This is the game:
Primes of the form 1 + 2 * r * s where r, s are primes.
These primes are designed to be hard to reason about, but not so hard that one can not continue to make progress over time.
A primbix value is the number of leaves in the associated binary tree:
0: composite numbers>= 1: primes>= 2: primbixesThe minimum primbix sequence is the lowest numbers of increasing primbix values. This number sequence might be thought of as self-fabricating “pieces” to use in the Primbix Research meta-game.
We have found what we believe is the minimum primbixes for the range 0 to 2^64:
0: 0
2: 1
13: 2
53: 3
317: 4
3407: 5
8243: 6
49459: 7
197837: 8
1187023: 9
18314419: 10
112006043: 11
491325707: 12
2047857083: 13
9869296583: 14
107478087043: 15
460785710279: 16
5075320359359: 17
12279094871663: 18
83710206810107: 19
294698527560839: 20
1771883742990899: 21
11091321145241003: 22
57618683965522403: 23
656132906468552867: 24
5115750284273141999: 25
However, the theorem technique we used is not 100% reliable and we do not understand it completely.
To check the entire range 0 to 2^64 using current compute (26 cpu cores 24/7), will complete in approximately 14 000 years.
This might seem like a very long time, but compared to the time delays at scales of galactic communication, this is very doable.
With other words, by the time humanity has become an advanced civilization and ready to send a signal across the galaxy, we will have covered the 0 to 2^64 range with very high likelihood and can expect other advanced civilizations to have done the same.
Primbix Research is also designed to filter out potential unsafe civilizations, that might be dangerous to communicate with.
The prediction of Eternal Cosmic Inflation by most theoretical physical models of the early universe, which causes a doubling of the universe approximately every trillionth of a trillionth of a second, suggests according to Alan Guth, one of the pioneers of Inflation Theory, that our civilization might be the first in the observable universe.
However, due to the expected amount of planets that will be formed over time, it might be likely that intelligent life forms will evolve in our galaxy in the future.
Therefore, using Primbix Research we might have a first-mover advantage, which can allow us to establish computation and communication superiority, forcing later advanced civilization to adopt to our ethical and philosophical principles, should they wish to join our galactic civilization.
This does not imply that our galactic civilization will be a hierarchical power structure. Primbix Research is part of our effort to find a Synchronizing Symbol of Power (SoP). Hierarchical power structures are associated with Cosynchronizing SoPs and Abstract Corruption in the theory of Avatar Extensions.
You can read more about SoPs here.
You can read more about Abstract Corruption here.
The fastest method of interstellar travel will likely be mind uploading and sending this information across the galaxy with electromagnetic radiation.
However, to build a receiver antenna in another solar system, we first need to send a microbot that is programmed to collect resources and construct the necessary infrastructure.
The reason that it needs to be a microbot, is that interstellar dust makes it difficult to send objects of larger sizes. At significant fractions of the speed of light, collisions with dust particles can be highly damaging, making it extremely difficult just to send a single microbot successfully to another solar system.
A such microbot might land on an asteroid or comet in another solar system, deploy nanofactories to produce greater avatars of itself, which in turn can perform more advanced construction tasks. It does not need to be programmed with a complete knowledge of physics and mathematics. Instead, we program the microbot with the ability to learn and to do meta-theorem proving.
Meta-theorem proving is a technique in Path Semantics where you start with a small set of axioms, that encode information about the inference rules you need in some more complex mathematical language design. An algorithm can search for proofs of new inference rules, using a general proof optimizer and narrow AI tech.
We already have multiple generic solvers and a new logic for something we call “Outside Theory” of mathematics: Avatar Logic
However, more fundamental research is needed on meta-theorem proving specifically, before we try to combine this knowledge with more research on Outside Theory.
This year, we published a new standard for meta-theorem proving: HOOO EP 2026 Standard
The new standard will help to separate proofs about meta-strength from collapse-strength.
Humanity is currently not on the track to start a galactic civilization.
At the current course, it is expected that humanity will go functional extinct in near term future, followed by total extinction in medium term future. The likelihood that humanity will survive in the long term at its current course, is close to 0%.
One reason for high likelihood of extinction, is the focus on developing AGI (Artificial General Intelligence), when the focus should be on educating people about Gini solvers (see prototype implementation).
A Gini solver is a stochastic binary search algorithm that optimizes the economic inequality gradient with respect to some Gini coefficient target. Early tests show that this can significantly boost the ability to solve constraint problems without backtracking, a class of problems that is relevant for real world economies.
An educated guess of the benefits of starting to use Gini solvers in our economy today, suggests it might lift 2 billion people out of relative poverty and make the impact of charity much more efficient. The economy will also become more resilient and adaptive to increasing AI automation.
AGI is not going to be good at tasks which Gini solvers are good at. At the current course, AGI tech will largely favor a small group of people economically, the super rich, a group that is overall ignorant and irrelevant for the goal of starting a galactic civilization. There are also early warning signs that this group of people are losing control over AGI agents under training.
In the past 10 years, zero efforts have been done by humanity to educate themselves and use Gini solvers in the economy.
The current course of humanity, toward likely extinction, is what we would expect in a world where humanity did not educate themselves about Gini solvers, 10 years ago.
As the founder of the AdvancedResearch community, I realized, that simply publishing useful stuff is not going to help humanity.
Therefore, I have started a new privatized branch of AdvancedResearch, which research results will be kept secret from the public, with the aim to start a new industrial sector with negative carbon emissions.
The reason for secrecy is industrial leverage: Humans are more likely to contribute in a better course direction for humanity, through competition and uncertainty about the future, than to exploit public available research.
There are not many countries in the world that have the necessary requirements to utilize the research done through this privatized branch of AdvancedResearch. It just happens to be that in the country I live in, Norway, has these requirements. I would probably not aim for this strategy if I had to cooperate with some distant countries.
If developed successfully, then Norway is going to lead humanity toward a more productive future than merely humanity going extinct.
This is a heads-up for the nations that are currently behind in research.
At some point in time in the future, Norway might initiate inter-national collaboration among scientists, if they first have established a significant research advantage that ensures their beneficial position. However, no such date will be published before it happens, meaning that these other countries, that do not wish to be left behind, must compete through internal research efforts, given that they can not rely on Norway sharing their research with them.
I certainly will not share my own research in this branch of AdvancedResearch. This is going to be secret, going forward.
A key part of the industrial leverage strategy, is that it is not published research that drives up efforts, because human brains tend to underestimate the benefits of current research results over long term, but that all the things that humans do not know, is what drives their motivation to put in more effort.
This uncertainty has also a positive side-effect, that to keep up, humans also have to divest resources from projects with destructive consequences for our current habitable environment.
The industrial leverage strategy is not merely to use one strategy for leverage, but to use the current leverage to develop a stronger form of leverage, which in turn will result in a stronger form of leverage, repeatedly, until the current course toward extinction for humanity is changed.
If humanity goes extinct, either functionally or totally, then our goal of starting a galactic civilization will not complete. To prevent humanity going extinct is a necessary sub-goal, which we will need to deploy suitable game theoretic strategies to achieve.
Ideally, I would like all research to be public under the AdvancedResearch community, but experience shows that this is inefficient over scales of decades, on behalf of humanity. It is not lack of knowledge that is critical, but under-utilized knowledge. If humanity one day starts to over-utilize knowledge, then I will go back to making my research public.
Sometimes, I receive personal thanks for making my research public, which I am grateful for. I will continue to publish a lot of research and also try to find more ways to publish it over time.
The industrial leverage strategy is only relevant for my research where I see potential gains of keeping it secret. I have no interests in the financial gains from privatizing this research, but of course, it is impossible to start a new industrial sector with negative carbon emissions, without the required financial means.
Norway is continuously supporting people who are trying to start up such projects, but it is from a perspective of a national interest. It is easier to get financial support for the research needed this way. If I make this research public, then Norway might see this direction of research as less potentially a sound investment strategy. Coupled with the unique requirements Norway has to pull off a such project, makes it important to separate between the public and the private research branches of the AdvancedResearch community.
This will be my strategy going forward, at least until next year. What happens beyond next year, depends on what options I have by that time. Unless my options change, I expect that my strategy will remain the same.