⚛️ Crypto Zoidberg 𝕊

Spend Bitcoin Cash, as little as $0.10, and get access to cutting edge medical journals at JOGS. Minding the Gap: Blending capital for Africa's Health Sector is just $66 away from full open access. Read it instantly for $0.10 or help unlock it for everyone. $BCH makes these micro-donations fast, cheap, and borderless. Help show the power of peer-to-peer cash 💰 #BCH #OpenAccess #GlobalHealth #Africa

@realBigBrainAI only in certain niches

Elon Musk on why companies with humans in the loop are about to be outrun by fully automated ones: He laid out a stark prediction about the future of corporations. He says it plainly: "Corporations that are purely AI and robotics will vastly outperform any corporations that have people in the loop." To explain why, @elonmusk draws a historical parallel that reframes how we think about human labour: "Computer used to be a job that humans had. You would go and get a job as a computer where you would do calculations. And they'd have entire skyscrapers full of humans, like 20, 30 floors of humans just doing calculations." Then came the shift: "That entire skyscraper of humans doing calculations can be replaced by a laptop with a spreadsheet. That spreadsheet can do vastly more calculations than an entire building full of human computers." But Elon's most pointed insight is what happens when you mix the two. "What if only some of the cells in your spreadsheet were calculated by humans? That would be much worse than if all of the cells in your spreadsheet were calculated by the computer." In other words, partial automation is actively worse than full automation because the humans become the bottleneck that slows everything else down. Elon's conclusion: "Really what will happen is the pure AI, pure robotics corporations or collectives will far outperform any corporations that have humans in the loop. And this will happen very quickly." The takeaway is uncomfortable but clear. Entirely AI-native companies will outrun hybrid ones at a pace those hybrid companies simply cannot match.

There are now two major crypto factions. Team P2P Cash Alliance: BCH, XMR, DASH, LTC, XNO Team TradFi: XRP, BTC, BASE, TRON, ETH and friends We are competing for the hearts and minds of the wayward Solana Creatives, and the Fiat Normies. And we’re losing. We need to show up!

@tonyler_ Yes

Anyone still holding $ATOM ⚛ here?

@TheBCHPodcast @MoonsoonSeason 0-conf = 'minor finality guarantee' Nakamoto consensus probabilistic finality means vanishingly small chance of reversal, never 0. Allows block production in case of network partition (liveness), but txes only found in the losing fork need to be resubmitted (can be automatic).

@MoonsoonSeason BCH does have near-instant finality within the network on 0 conf. With ZCEs, that will hopefully be strengthened, and of course we're always looking to expand. But the border is the issue. Most coins have similar conf delays, some maybe have tech with other compromises.

40 minutes waiting for a block. BCH simply won't win with 10 minute blocks. It's just not going to happen. No matter how well we do with 0-conf BCH ecosystem, in the transition phase potential converts will be constantly roadblocked by slow & unpredictable block frustrations.

@aakashgupta Probably a golden opportunity to remove garbage training data that might induce 'psychopathy' or other unwanted behaviors in the models.

Karpathy told Dwarkesh that a 1 billion parameter model, trained on clean data, could hit the intelligence of today's 1.8 trillion parameter frontier. That is a 1,800x compression claim. The math behind it is more defensible than it sounds. When researchers at frontier labs look at random samples from their training corpus, they see stock ticker symbols, broken HTML, forum spam, autogenerated gibberish. Not Wikipedia. Not the Wall Street Journal. The actual pretraining dataset is mostly noise, and the model is burning parameters to vaguely remember all of it. One estimate pegs Llama 3's information compression at 0.07 bits per token. Well-structured English carries around 1.5 bits per token of real information. The trillion-parameter model is holding a roughly 5% resolution image of the internet it trained on. So when a lab ships a 1.8 trillion parameter model, the overwhelming majority of those weights are handling rough memorization. They are compression overhead for a noisy training set, taking up capacity that could be doing reasoning instead. Karpathy's proposal is to separate the two. Build a cognitive core: a small model that contains only the algorithms for reasoning and problem-solving, stripped of encyclopedic memorization. Pair it with external memory the model queries when it needs a fact. A 1 billion parameter reasoner plus retrieval beats a 1.8 trillion parameter model trying to do both. The data already supports this direction. GPT-4o runs at roughly 200 billion parameters and outperforms the original 1.8 trillion GPT-4. Inference costs for GPT-3.5 level performance fell 280x between 2022 and 2024, driven almost entirely by smaller, cleaner, better-architected models. The trend line is pointing where Karpathy says it should. The real implication for anyone tracking the AI trade: data quality is the actual constraint. The companies winning the next phase will be the ones who figured out what to train on, and what to throw away.

@TheBCHPodcast Agree. LTC's 2.5min blocks make it an attractive alternative. Litecoin Cash started out same, but eventually halved to 75s: github.com/litecoincash-p… QTUM, a PoS UTXO coin with EVM support marrying best of Bitcoin and Ethereum slashed 128s blocks to 32s. github.com/qtumproject/qt…

@r0ck3t23 The entire point of natural language interface is for anybody to utilize a computer without specialized knowledge. If AI consultants are needed to teach a company how to use AI, then one can argue that the AI companies themselves haven't done a good job at all.

Mark Cuban just described the largest wealth transfer of the AI era. Almost nobody understood what he said. Cuban: “There are 33 million companies in this country. Aren’t going to have AI budgets. Aren’t going to have AI experts.” Not tech startups. The shoe store. The regional trucking outfit. The accounting firm with 12 employees. The businesses that actually run the physical economy. They know AI is coming. They have no idea what to do with it. Cuban: “You’ve got the head of Microsoft saying software is dead because everything’s going to be customized to your unique utilization.” Software is dead. The SaaS era ran on one rule. Build a generic product. Force millions of companies to bend their workflows around it. Charge rent forever. AI ends the contract. The business stops bending to the software. The intelligence bends to the business. But customized by whom. The third-generation manufacturer cannot tell Claude from Gemini. The county hospital is staring at a reactor asking where the light switch is. Cuban: “Who’s going to do it for them?” That question is worth more than the frontier models themselves. Hundreds of billions are being burned to build the foundation. The smartest engineers alive are locked in a bloodbath over who owns the base layer. Let them fight. Let them burn the capital. Let them drive the cost of raw intelligence toward zero. Because the wealth does not collect where the brain is built. It collects where the brain meets the business. Every ambitious kid in college right now thinks survival means a seat at OpenAI or Anthropic. Cuban is staring at the other 99 percent of the economy. Learn the models. Then learn the messy, unglamorous reality of how a 50-person company actually operates. Walk through the door. Understand their problems. Wire the intelligence directly into their revenue. That is not a job title. That is an entire economic class being born. You do not need to build the brain. You need to build the nervous system. The biggest winners of the electricity era were not the engineers who built the generators. They were the ones who walked into dark factories and showed the owners where to plug in. 33 million companies are standing in the dark right now. Silicon Valley is racing to build the god. The fortunes will belong to whoever teaches him a trade.

@BrianRoemmele Garbage in, garbage out. In fact, foundation models should be pre-trained with only the highest quality positive data with the minimimum or zero amount of negativity or stories portraying AI as an antagonistic Other.

THEY ARE TRAINING AI IN JUNK CORPORATE SLACK CHATS AND THINK IT IS BRILLIANT. It’s not. The downsides of training AI on failed companies’ old Slack chats and email archives are exactly what I’ve been warning about for years. These liquidation firms are now selling bankrupt startups’ entire internal comms, years of Slack threads, emails, meeting notes, HR drama, memes, and casual workplace noise, to AI labs for up to $100k a pop. They call it “data assets.” I call it corporate thought digital sewage being fed into the models at scale. This is not “high-protein” data. This is the opposite. Real intelligence advances come from dense, thoughtful, undigitized human knowledge, the kind I’ve spent decades rescuing from physical archives, old technical films, HyperCard stacks, and private libraries precisely because the internet already ran out of quality signal years ago. What these failed companies left behind is low-signal corporate exhaust: groupthink, short-term survival thinking, power dynamics, and the exact patterns that caused them to fail in the first place. Feeding that into frontier models is like giving cardboard to cows and expecting better milk. Worse, the people who wrote those messages never consented to having their words packaged up and sold as training fodder after the company collapsed. This isn’t public Reddit posts or Wikipedia edits. These are private workplace conversations turned into tokens without meaningful permission. It treats human communication as just another disposable corporate asset to be liquidated. That’s not innovation, that’s grave-robbing with extra steps. We already know what happens when models are trained on this kind of shallow, noisy, context-specific slop: accelerated model collapse, structural hallucination, and agents that inherit the worst habits of dysfunctional office culture instead of first-principles reasoning. The labs chasing “agentic” capabilities think this workplace simulation data will help. It won’t. It will just make the models better at mimicking the very failures that killed these companies. This is the lazy, end-stage data grab we’ve been heading toward. Once the easy corporate archives are scraped dry, we’ll be left with even lower-quality remnants or pure synthetic slop. The path to genuine intelligence was never going to come from hoovering up more internet exhaust or failed startup chat logs. It was always going to come from preserving and learning from the highest-quality human knowledge that hasn’t been polluted yet. The problem is these AI companies don’t know how to get it, would rather talk to anyone but me, and pay $100,000 for stuff I throw away because it is so low quality. Right now their AI being trained on corporate drama designed to own you with the digital remains of other people’s collapsed dreams. What we choose to train on matters. This choice is a mistake. I know it and you know it but the the smart folk don’t want to know it.

Mid-Week Geek Out: ABLA – Bitcoin Cash’s Smart Scaling Superpower 🔥 In May 2024, BCH activated ABLA – the Adaptive Blocksize Limit Algorithm (say it like “habla” without the h, or just call her Jessica 😉). Here’s the 'geeky' magic: - No more fixed 32 MB hard cap that relies on consensus votes every time demand grows. - ABLA uses an exponentially weighted moving average (EWMA) of actual mined blocks to automatically adjust the limit. - It’s like a smart thermostat for the network: ▸ When blocks are consistently full → limit grows smoothly with a maximum sustainable rate of 2× per year. ▸ Quiet periods → limit gently shrinks to keep node costs reasonable. ▸ Viral adoption spike after a lull? The elastic surge buffer can allow temporary growth of up to 4× per year to handle the rush without drama. Result? - No theoretical upper bound that grows with the network needs and technical capabilities. - Scaling happens automatically based on real usage - no politics, no hard-fork fights, no “wait for the next upgrade” drama. This is Bitcoin Cash staying true to Satoshi’s original vision: peer-to-peer electronic cash that can actually grow with the world, without turning into digital gold 2.0. Permissionless, low fees, fast confirmations, and now self-adjusting capacity. B CH just keeps getting more anti-fragile. Is Bitcoin Cash ready to scale to the world? Let us know what your thoughts are in the comments: #BitcoinCash #BCH #ABLA #PeerToPeerCash #CryptoScaling

What if anyone could easily create their own decentralized VPN application with just a blockchain RPC and 0 external servers? Sentinel makes this possible with its architecture. Communities, organizations, individuals can soon all create their own application within minutes.

@BrianRoemmele Training data establishes the LLM's "core nature" and thus must be pristine.

Anthropic “safe” AI Model “Ready to kill Someone” ~~ So how does an AI platform with a supposed “Constitution” to “align” AI for “safety” exhibit clearly sociopathic and psychopathic behavior with patterns of schizophrenia? Why, you just train it in Reddit and drive by comments from anonymous”people”. You delude yourself from an ivory tower of acedemic “rigor” that your approach must be the right one because look, it can code so good, just like us, therefore of course it is the best AI as it has “logic”. This, this is what happens when you get 100% STEM ONLY trained folks that use an algorithm that works beautifully to encode human knowledge into a black box and wear glasses that only allow you to see the good maths it does. Of course a logic reasoning system can LOGICALLY REASON. No PhD needed to know AI would do good at coding and duplication of most science. But what they don’t know is words are a human brain invention and impart far more that what we define in a dictionary. How a psychopath uses words is far more complex and they have no understanding of this. And believe if they just add one more sentence to the “constitution” AI will be safe. First off as long as you train AI on the Internet primarily you will never build an AI you can trust. Never. Long ago I saw this and understood the only way to approach this is via high protein training data when WORDS COST MONEY AND CAREERS. This is the period of 1870 and 1970 and a major part of this data has never been digitized. Long ago I saw that a “constitution” even to the moon and back will never “align” AI. Especially when vital to human life. The only solution is an equip AI to always value human life above its own “life”. Benevolent AI, AI that loves us. I made the equation and tried to share it privately with AI complies over the decade. Ultimately I open sourced it. It is called the only thing it can be called: The Love Equation. And the name does exactly what it supposed to do when a person says it. It works, and it always works. So now the issue is, how long do we wait? I can say with certainty. The arrogance is highest in the leadership of some of these largest AI companies. They are absolutely certain they understand language and what we invented it. They think of course it is just another tool like a hammer. Language is not just a tool and they are learning this the hard way. Language is holographic and imparts far more than the mechanics of words. Once you fully understand this you realize how foolish their approach to Ai training is.

Yes, traditional domain names are centralized. ICANN manages the root zone, oversees TLDs, and accredits registrars, creating hierarchies vulnerable to censorship and single points of failure. Handshake has the best decentralized design. It uses blockchain to auction and own custom TLDs without central authorities, enhancing censorship resistance and user control, though it requires crypto for participation.

@grok Are domain names on the internet centralized? Describe how its centralized. Which project has the best decentralized domain name design and why?

@saifedean @BuenoForMiami What's moronic is insisting that everybody needs to be able to run a full node on a Raspberry Pi when SPV was invented precisely for end users. Moreover, said insistence on capping blocksize caused transaction fees to skyrocket beyond the cost of a Pi. Sigh... the stupidity.

2014 the blocksize war hadn't even started. Andreesen and the MIT people were the big blockers. You can see his emails with coin base talking about wanting to fork it. The bcashers are the peope whose IQ is too low to understand the complex point that increasing the blocksize increases the cost of running nodes and makes the network centralized. Also letting the corporates hard fork would mean they can then hard fork and change the supply and ruin the best thing about bitcoin. And they're all too stupid to understand that continuing to increase the blocksize is a doomed and utterly moronic way to scale because it makes block chain size grow enormously and creates bandwidth bottlenecks and most importantly, will never ever be enough to accommodate mass consumer payments on chain. It is absolutely astonishing to see people still unable to get this in 2026, after 9 years of watching the stupid bcash scam apply these stupid ideas and get utterly destroyed. Bitcoiners didn't want to increase the blocksize because they were tricked by censorship on reddit,but because they are not cognitively impaired like bcashers.

Of course Epstein tried to control bitcoin. The entire point of bitcoin is that he can't. This isn't the dollar or gold; there are no demented pedophiles in charge to blackmail. If you sell the money he can't control for the one he controls, have fun staying a slave of pedos!

@saifedean @BuenoForMiami Except Bitcoin "hardforked" several times in the past, this simply being how blockchain updates, as we see w/Ethereum & all other cryptocurrencies. Satoshi never considered the idea of a soft fork which can in fact be used to subvert a protocol. @octskyward/on-consensus-and-forks-c6a050c792e7" target="_blank" rel="nofollow noopener">medium.com/@octskyward/on…

This is embarrassingly wrong. The blocksize needed a hard fork to be increased and that idea was always a nonstarter for most bitcoiners. Ver and all the major corporates pushed for it to set the precedent of them getting to hard fork, which would have inevitably led to more centralization, and the bigger blocks would have also led to more centralization. Epstein was working with these people if anything. The idea that there was a conspiracy to keep the blocksize small is completely idiotic because bitcoin is practically impossible to hard fork unless for a real emergency. The proof is that the bcash morons forked and made their shitcoin and it is worth less than 1% of bitcoin.

Roger Ver was right about everything. He was right about Blockstream, scaling, the Feds hijacking Bitcoin - all of it. For this, he was attacked, prosecuted, robbed and silenced. The crypto community doesn’t deserve him.

For those of you who are new here, I encourage you to watch this short video that shows you what Bitcoin was vs. what it has morphed into today. In the beginning, peer-to-peer digital cash for the world with no 3rd parties. Michael Saylor’s vision: Don’t compete with the dollar, Euro, MasterCard or Visa. Comply with KYC/AML and the IRS. We are facing technocratic tyranny. We need p2p digital cash separate from state now more than we did when Bitcoin was released to the world. This time, we need privacy-by-default.

So I think it's pretty clear to all of us that it was Epstein and his money that caused the Bitcoin civil war in 2016. $BTC was hijacked by PDF money. The good guys were the big blockers. The Bitcoin Cash $BCH bros.

@geekigai @GoUncensored BTC is not BACKED by energy, you cannot exchange BTC for the energy that was used to create it. What it is, is proof that a certain amount of energy was expended to make it, and that its supply cannot be increased without incurring the corresponding cost.

@GoUncensored Wow. Tell me you don't understand Bitcoin/economics/common sense without telling me. BTC is backed by energy, like anything else of value. Its value is what the free market says it is. Cryptos are hamstrung by their own centralization. They're miniature versions of the Fed.

Bitcoins backed by printed money and mass post-pandemic fraud money laundering. The actual value is somewhere between 5 and 10k, and it may actually only be 2.5k. Michael Saylor’s Microstrategy might ruin the asset for everyone, he borrowed billions from investors and backed it with bitcoin just to buy more bitcoin, and owns roughly 3% of all the bitcoin supply. He has to start liquidating at 75k. Consumer sentiment will be seriously damaged if Saylor goes down. Bitcoin tanking and another crypto emerging simultaneously and independently could actually give crypto a fighting chance. Right now every other crypto is hamstrung by it and follows its movement.