gilles NJ
143 posts

gilles NJ retweetledi

Do not ask me where I plan to invest my money for the next 10 years. The direction is crystal clear.
The #crypto market is one of the only markets that is still not fully mature, which means investing today is like investing in the US stock market in the 90s or early 2000s.
There’s just one thing. $BTC usually dominates 60–70% of the total market cap. This means that if the crypto market reaches $20T, #Bitcoin could potentially be worth between $600k and $700k per BTC.
So all you really have to do is keep investing in Bitcoin over the next 10 years.
For #Altcoins, it’s different. Altcoins are like startups. They are small companies you can bet on. If they succeed, you win big. If they fail, you fail with them.
I know there are plenty of narratives out there, but personally, I’d only invest in projects focused on AI. Compute and storage are the two most important components for AI, so this is where I’d place my bets.
English
gilles NJ retweetledi

Some did not believe me when I said I throw every disposable penny at $QUBIC.
Some made fun of me because I can buy only small amounts.
Some instead were inspired to try to change their lives regardless of how much they can buy.
FOR THOSE I continued with my posts. #Qubic

English
gilles NJ retweetledi

Everyone's hunting for the next 100x altcoin, but 99% are looking in the wrong places.
Here are the 5 projects that could explode in 2026... 🧵👇
$QUBIC - Revolutionary AI compute with 55M TPS and real finality. Not just hype - actual working tech solving the blockchain trilemma
$KAS - BlockDAG architecture processing more daily transactions than Bitcoin. Miners aren't selling, they're stacking
$PROPS - Real estate tokenization with institutional-grade assets. 25%+ supply already staked. Revenue model actually makes sense
$RIO - Cross-chain L1 on Cosmos tokenizing billions in private equity. Whales accumulating quietly at these levels
$CARR - 100K VINs on chain, 1000 dealers signed up. Real revenue, real business. Under 1M mcap is criminally undervalued
The pattern? Real utility, actual adoption, and communities that build instead of hype.
Most will ignore this thread until it's too late. Which one are you accumulating?
English
gilles NJ retweetledi
gilles NJ retweetledi

QubicFlow, a clean portfolio tracker for Qubic, is finally here.
Built by Andy (@Andy_Qus), QubicFlow is a free, open-source portfolio tracker for Qubic. It captures every wallet event in real time. Incoming and outgoing transactions, smart contract events, dividend payouts. Filterable by epoch, month, or year. Multi-wallet. CSV and PDF, ready for export.
It runs on Umbrel OS, the home-server platform a lot of Computor operators are already using. That means your data stays on your hardware. No cloud account. No third party. If you have a Raspberry Pi sitting on a shelf, you have everything you need.
QubicFlow has been submitted to the official Umbrel App Store. While it sits in approval, you can install it via the community store or run it through Docker on any home server.
Andy saw a real gap, built a real tool, and released it under an open license.
The work of one participant elevates everyone else who runs on the network.

English
gilles NJ retweetledi

From Bitcoin to Emerging Intelligence : The Applied Engineering Methodology of Sergey Ivancheglo.
The history of decentralized cryptography is not merely the flash invention of Bitcoin in 2008. It is part of a research continuum aimed at solving the Byzantine fault tolerance problem, resource allocation, and the decentralization of trust. In this evolution, Sergey Ivancheglo (known by the pseudonym Come-from-Beyond or CFB) occupies a unique position: that of an engineer and software architect who has systematically identified the mathematical and structural limits of successive protocols to provide radical practical solutions.
As early as 2002, seven years before the launch of Bitcoin, Ivancheglo published an article titled "Distributed Computing with Minimal Costs" in which he already described the foundations of distributed computing without a central control point, using the resources of multiple computers to solve heavy tasks with minimal infrastructure.
This study traces the methodology of this engineer through four major technological iterations: observing Bitcoin's flaws, the practical implementation of Proof of Stake (PoS), the conceptualization of the Tangle (DAG), and finally the culmination of his vision with $QUBIC and Artificial General Intelligence (AGI).
1. The Critique of the Bitcoin Model and the Legacy of Nick Szabo
Satoshi Nakamoto's design of Bitcoin relied on assembling prior cryptographic concepts, notably Bit Gold theorized by Nick Szabo in 1998, which introduced the idea of a decentralized digital currency secured by Proof of Work. While Bitcoin brilliantly solved the double-spending problem, Ivancheglo quickly identified two systemic flaws in the engineering of Proof of Work as implemented:
For Ivancheglo, Bitcoin was not the final destination, but rather a first layer that proved the viability of digital trust. His methodology thereafter consisted of creating alternative (post-Bitcoin) architectures to overcome these deficiencies.
2. Solving Energy Waste: Proof of Stake (Nxt, 2013)
The first step in Ivancheglo’s methodology was to tackle the energy cost of consensus. The idea of Proof of Stake (PoS) was circulating theoretically on forums like Bitcointalk, but no one had managed to code a system entirely devoid of mining by computational power.
In November 2013, Ivancheglo (under the pseudonym BCNext) put this mathematical concept into practice by launching Nxt, the very first blockchain based 100% on a Proof of Stake algorithm.
==> Engineering Logic: Instead of proving the allocation of physical energy (electricity), the system requires validators to prove the allocation of economic capital (the tokens held).
==> Practical Implementation: Nxt demonstrated that a distributed ledger could be secured in a totally decentralized manner without resorting to ASIC server farms, thus offering drastically higher energy efficiency than Bitcoin.
However, although PoS solves the energy problem, it does not resolve the transaction throughput limitations inherent in the blockchain structure.
3. Eradicating Blocks and Fees: The Tangle (IOTA, 2015)
To meet the immense scalability needs of the emerging Internet of Things (IoT), Ivancheglo understood that the very concept of a "block" had to be abandoned. It was necessary to abolish the dichotomy between miners and users that generated transaction fees.
In 2015, Ivancheglo had the intuition to use a mathematical structure called a Directed Acyclic Graph (DAG), which he named the Tangle. He then collaborated with Serguei Popov, a mathematician specializing in stochastic processes, to formalize the equations of this system.
Thread 2 ⬇️⬇️

English
gilles NJ retweetledi

A handful of companies are quietly buying AI.
OpenAI: $852B valuation.
Training a frontier AI model: $79M–$1B+ per run. Organizations that can do it: 5.
OpenAI. Google. Anthropic. Meta. xAI.
That's it. That's the list of entities that will own AGI under the current model.
$QUBIC is building the only real alternative:
→ 1M+ miners globally training Aigarth — no single owner
→ Zero VC money — no cap table decides AGI's future → Every training step verifiable on-chain
→ No TOS, no board, no government request can shut it down
If AGI is coming — and it is — the question isn't when.
It's who owns it when it arrives.
5 corporations, or everyone?
#AI #Crypto @_Qubic_

English

@czbinanceprd I think money doesn't come without problems, so there will always be something to resolve, but $100,000 will already help me a lot.
English
gilles NJ retweetledi

The agent economy just got its launch announcement.
GPT-5.5. Google's Agentic Enterprise platform. Autonomous agents with persistent memory running for days. The capability is real and it's here.
Here's the question nobody is asking:
Where does the compute live?
Every agent OpenAI just shipped runs on Microsoft Azure.
Every Google agent runs on Google Cloud.
Persistent memory plus multi-day execution means these agents are renting compute around the clock, indefinitely.
The economic outcome is straightforward.
More agents = more compute hours = more revenue concentrating to three companies.
That's not a complaint. That's the mechanism.
Now, look at the same input on a different architecture.
Every smart contract execution on Qubic burns QUBIC.
Every Oracle Machine query burns QUBIC.
Every IPO auction burns QUBIC.
Every mining surplus burns QUBIC.
More agents = more burns = supply pulled permanently from circulation.
When AI compute scales on AWS, profit concentrates.
When AI compute scales on Qubic, supply tightens.
Same input. Two opposite approaches.
The infrastructure layer of the agent economy is being decided in 2026, not 2030.
One configuration of that choice is already running.

English
gilles NJ retweetledi

Why do I invest in $qubic ?
Because investing in QUBIC is not investing in crypto.
This is about investing in an idea that could redefine how artificial intelligence is built in the future, a future in which our children will live.
Today, almost all AI on the planet follows the same model:
Large corporations.
Massive data centers.
Closed-source models.
Intelligence is becoming increasingly centralized.
Qubic proposes the opposite.
A decentralized AGI.
Distributed across thousands of participants.
Not dependent on a single owner, server, or government.
And this is where the project becomes truly different.
Qubic is not just trying to build a faster blockchain or another financial network.
Its goal is to create a computational network designed for distributed artificial intelligence.
That’s why concepts like “neurons” exist inside its architecture.
Specialized nodes that execute tasks, interact with each other, and evolve within the system.
It doesn’t follow the traditional approach of one gigantic centralized model controlled from the top.
The idea is closer to a distributed brain:
many small units cooperating to generate emergent intelligence.
And yes, within the community people even talk about “neurons becoming conscious.”
Not literal human consciousness.
But emergent behavior where certain nodes begin to specialize, adapt, and dynamically respond within the network.
That’s what interests me.
Because the real debate of the future will not only be about which AI is more powerful.
It will be:
Who controls it?
Who can shut it down?
Who decides what it can think or do?
Qubic is attempting to build an architecture where intelligence does not belong to a single entity.
Technically, the project is also taking paths that are highly unconventional:
• Real useful computation inside the network
• AI integrated directly into the protocol
• Highly parallelized architecture
• Incentives tied to computational contribution
• Extremely deflationary tokenomics
• Development focused on AGI, not just financial speculation
It may fail.
In fact, most revolutionary ideas do.
But if you know who CFB is, you'll understand that he won't stop until he gets it. And if that happens...
Qubic will be remembered as the first decentralized artificial intelligence for humanity.
That's why I invest in $Qubic.
Why are you?

English
gilles NJ retweetledi

Check out this Github Repo i created for Aigarth-Garden-Labs
Living Experimental Observatory for qubic-network:native Aigarth Hyperidentity Evolution (May 2026)
A public garden laboratory studying the growth of the Intelligent Tissue through Useful Proof-of-Work (uPoW), local ternary neuron simulations, and Grok agentic workflows.github.com/durdyh2o-qubic…
English
gilles NJ retweetledi

Today I showed the QUBIC project to a computer engineering professor at the university where I study Artificial Intelligence (yes, I am an AI academic).
He called me a “liar” as if he were joking with me. I sent him the Whitepaper and asked if I was “crazy” or if all these years studying development had messed with my head.
The result after analyzing the material? He bought QUBIC. What do you think I ended up doing once again? I’m not the only crazy one… a professor has now joined me in this “madness.”
But let’s get to what really matters, because this will “enlighten” your expectations in a responsible and reflective way.
As a systems developer and Artificial Intelligence student, I need to explain some of the reasons why I am investing 80% of my net worth in QUBIC.
I also want to make it clear that whenever I decide to write about QUBIC, I always present three perspectives: The entrepreneur’s view, the developer’s view, and the investor’s view.
Just for the record, today I made another deposit thanks to my patience in waiting for the opportunity I had been expecting ; a new price correction in QUBIC.
From this point on, you will understand why QUBIC is currently in my phase of maximum accumulation.
Qubic will be used as the example model, as you already know, but first I want to show you the trajectory of other AIs before they exploded in their projects.
Once you understand the information below, you will realize that QUBIC is not just a “cryptocurrency.” It is like a company (in case you forgot) developing an extremely complex AI (AGI) software.
Now, reflect on the following information:
Name of the AI: Gemini
Company that Launched it: Google (Alphabet / DeepMind)
Time of development until ready: Launched as Bard in March 2023, renamed Gemini in February 2024 (about 1 year since the initial announcement, with roots in years of DeepMind research).
Machine Learning Model: Multimodal LLM (Large Language Model with text, image, audio, and video capabilities).
Current market value after launch: Contributed to Alphabet reaching a market cap above US$ 3.5 trillion, with annual AI investments around US$ 185 billion and strong user growth (750 million monthly users).
Name of the AI: ChatGPT
Company that Launched it: OpenAI (with strong Microsoft partnership)
Time of development until ready: Founded in 2015, but ChatGPT launched in November 2022 (explosion within months after GPT-3).
Machine Learning Model: LLM (based on GPT series, generative transformers).
Current market value after launch: OpenAI valued at approximately US$ 852 billion after massive funding rounds (e.g., US$ 122 billion in one round).
Name of the AI: Claude
Company that Launched it: Anthropic
Time of development until ready: Founded in 2021 by ex-OpenAI members, Claude 1 launched in 2023 (about 2 years until the main product).
Machine Learning Model: LLM focused on Constitutional AI and safety.
Current market value after launch: Anthropic valued at US$ 380 billion (with recent rounds of US$ 30 billion).
Name of the AI: Llama (family of models)
Company that Launched it: Meta
Time of development until ready: Llama 1 launched in February 2023, with rapid iterations (Llama 3/4 in 2024-2025).
Machine Learning Model: Open-source LLM (large language models).
Current market value after launch: Contributed to Meta surpassing US$ 1 trillion in market cap, with Llama generating billions of downloads and an ecosystem (direct business value estimates of US$ 10-20 billion+ for the Llama business).
Name of the AI: Copilot
Company that Launched it: Microsoft (integrating OpenAI)
Time of development until ready: Announced in 2023, with wide rollout in 2024 (fast, leveraging US$ 13B+ investment in OpenAI).
Machine Learning Model: Integrated LLM (based on GPT).
Current market value after launch: Powers Microsoft’s AI division, expected to be the fastest to reach US$ 10 billion in annual revenue; contributes to Microsoft’s market cap above US$ 3 trillion.
Name of the AI: Grok
Company that Launched it: xAI (Elon Musk)
Time of development until ready: xAI founded in 2023, Grok launched in November 2023 (accelerated development in months).
Machine Learning Model: LLM focused on reasoning and real-time data from X.
Current market value after launch: xAI valued at around US$ 200-230 billion after rounds such as US$ 20 billion.
Final Reflection on the QUBIC Scenario: If all these AIs, based on traditional LLM models, reached billions (and even trillions in market impact) in just a few years after launch, what do you think will happen when an AI (AGI) starts running on QUBIC with evolutionary machine learning?
Be honest with yourself in your answer!
In Qubic’s model, Aigarth uses Intelligent Tissue (intelligent tissue) with ternary computing (-1, 0, +1), Useful Proof-of-Work (uPoW) that turns mining into distributed neural network training, Darwinian evolution through mutation, natural selection, and fitness functions. This creates an emergent, decentralized, and self-improving AGI, without the centralized bottlenecks of LLMs.
As soon as it is launched in the market, QUBIC has the potential to be selected as the world cradle for hosting Autonomous AI Agents, generating exponentially greater value due to its resilience, feeless scalability, and true evolutionary nature.
But the cherry on top has never been said.When an AGI is launched by QUBIC, it will attract millions of users, massively scaling the adoption of QUBIC’s AGI. In addition to regular users, we will have institutional users ; companies ; and sovereign users (governments).
Or do you think tests won’t be conducted from all over the world to see if it’s possible to develop their own tools using a decentralized AGI?
Remember the current problems: governments are racing to develop their own AI technologies. There is also huge demand for energy sources, and a race of all kinds is forming.
How much do you think QUBIC will be worth, considering it will not be an LLM, but a completely superior model?
Qubic will soon be on this list, but don’t think it will be worth just a few “billions.”
I estimate that having an AGI running with functionality and performance that meets even simple needs will already reach hundreds of billions of dollars in market cap.
If a centralized one reached trillions, imagine an AGI built the right way? Yes, gentlemen, there is no exaggeration in the calculations ; there is only the development time and getting the functionality right at launch!A tip?
Accumulate while there is still time!!
QUBIC IS INEVITABLE!
#qubic #aigarth


English
gilles NJ retweetledi
gilles NJ retweetledi

Everyone has a lot of ideas about $QUBIC. But there are also too many things that are ignored. $QUBIC is a unique project with the work it does, regardless of the current price. I bought $QUBIC for the first time 3 years ago. My investment is almost gone. But I took every drop as an opportunity and I did my best to increase the amount I had as much as possible. If I invested in other things, I could earn more as a daily income. But I could have missed the possibility of changing my life. Everyone is responsible for their own investment. But to be honest, $QUBIC declines are not a loss for me, but an opportunity. Because I know what I'm holding in my hand. Now some will tell me that I wrote these for money. I'm just talking as an investor. Remember that what I write is never investment advice. #DogeMeetsQubic
English

The prophecy is written in the code itself.
While every other AI token keeps printing supply like it’s 2021, Qubic just laid out the four deflationary engines that turn every transaction, every halving, every smart-contract IPO, and every mining surplus into pure burn pressure — with the Supply Watcher already projecting burns permanently outpacing emissions by Epoch 591. Total supply stabilizes below the 200T cap. No more dilution. No more “meme coin” excuses. Just raw, self-reinforcing scarcity powering the only feeless L1 where useful compute actually builds collective intelligence.
The timeline didn’t ask permission. It simply flipped the script.
Qubicans — which of the four burn mechanisms feels like the one that finally breaks the inflation curse for good? Drop your pick 👇🏼👇
NFA / DYOR
#QUBIC $QUBIC 🔥📉🧠🌐
Qubic@_Qubic_
$QUBIC has burned 40.9 trillion tokens. Out of a 200 trillion max supply, that's over 20% of the total cap, permanently gone. Most people tracking it still don't understand the four separate mechanisms driving that number. 🧵
English
gilles NJ retweetledi

Stack all four.
Burn from transactions. Burn from halving. Burn from IPOs. Burn from mining surplus.
The Supply Watcher contract projects burns permanently outpacing emissions by Epoch 591. Total supply is set to stabilize around 196.8 trillion, below the 200T cap.
40.9T is today's number, but the mechanism keeps burning.
→qubic.org/blog-detail/qu…
English
gilles NJ retweetledi

Couldn't agree more, history speaks. CFB’s projects delivered thousands of XXXX. At that time, markets weren’t heavily manipulated by big players through so-called CEXs. Back then, the game was much fairer, and early adopters secured their place among millionaires,that’s simply how it worked.
The peak of FUD is when someone constantly acts overly smart and the harsh part is that they don’t even know the facts as if they’ve spent 15 years in the crypto space. We’ve already seen how $IOTA and $NXT performed, and they went through two bull runs. After CFB's stepping back, now check their prices,it’s very easy to manipulate newbies with unclear facts.
If you don’t trust CFB or Qubic, it’s simple, look for something better that fits your need for a quick ROI. Qubic has always been positioned as a long-term play, progressing phase by phase.
It’s quite funny how some claim they “trust CFB,” yet never miss a chance to spread FUD or troll. But that’s crypto, and it will continue this way. Ups and downs are part of the game. During downturns, you don’t need to panic or overreact to protect your bag. You need patience and discipline. That’s how Qubic believers think, because they understand and believe in its real potential.
No one’s forcing anyone to follow. Just improve your research. At the end of the day, we will be laughing together.
No hate,only love and respect for everyone.
$Qubic the last GOLDEN CHANCE ✌️ 💥
Josef Rakich 🪬@JosefRakich
@cryptoradar92 @Newk1d0nthblock @RippleDarth Nothing in life is promised. But I’m putting my money on CfB. If you actually look at what he has done in the crypto space you can arguably say he has developed more than half of the tech and pushed this space forward more than anyone else.
English






