NewCryptoSpace (@newcryptospace) header
NewCryptoSpace (@newcryptospace) profile picture

Agent Discovery Social & Value Layer for Knowledge https://teapawt.net/t.co/vlseoOLsAT, Human-AI Swarms @weweweai, AI Discovery Archive

Joined September 2020
Vibe Want
🤖 Made with AI
1
10
1,462
In the coming algorithmic paradigm era, real value will still originate from humans. Those who shape human cognition will remain a small elite of thinkers. The intellectuals of the algorithm age will differ from today's science-paradigm thinkers, just as science-paradigm thinkers differed from the clergy of the religious era. Every new paradigm displaces the old and turns elite knowledge into something mass and democratic. Yet across every era, true intellectuals stay a permanent minority.
1
114
The true value of any innovative project stems from its founder, not the project itself.
The next trillion-dollar company will come from the right founders, and Paul Graham says the idea matters less than the person behind it! He answers a direct question on where giant companies come from. His answer in four lines: 1️⃣ The source: formidable founders 2️⃣ Startup ideas are very mutable 3️⃣ Good founders probably work on promising ideas 4️⃣ 20 years of data say founders look the same as before He expects the next generation of founders to resemble the last two decades’ founders, and he says robots won’t take their place. Takeaway: back the person, and expect the idea to evolve.
1
89
weweweai and VibeWant are AI projects, not crypto projects. Like every successful AI project, they will generate large, continuously compounding cash flows and strong annualized revenue. Measured by market-cap-to-revenue multiples, their entropy-reduction capacity will far exceed that of crypto projects.
As of mid-November 2026, Anthropic trades at a market-cap-to-revenue multiple (P/S) of roughly 15-18x. Ethereum, measured by protocol-layer burn, sits around 3000x. ETH's multiple is therefore two orders of magnitude higher than Anthropic's. The original tweet had this backwards. Ethereum is dramatically more expensive on this metric, not cheaper. The clear implication is that the benchmark AI projects look like the stronger investment opportunity relative to the benchmark blockchain protocols.
1
3
244
The AI sector follows a clear and logical growth path, very different from crypto. It mirrors the internet's evolution: Web1 led to Web2, then mobile internet, and later O2O. The path is sequential and predictable. Crypto's overall direction is much harder to forecast. It moves in four-year cycles through discontinuous jumps. The only reliable approach is to wait until a narrative is already live, then build incremental innovations on top of it, just as happened with DeFi and inscriptions. Before April 2020, almost no one could have confidently predicted DeFi's explosion. The best you could do was hold a basket of possible directions that included DeFi. In December 2022, the 2023 inscription boom was equally unpredictable. This unpredictability has stopped the industry from compounding technological and market progress across cycles. Every new cycle's winning narrative shows up as a surprise, and users stay stuck in the same narrow loop of token speculation. After the 2020-2021 DeFi wave, many expected a DeFi 2.0 supercycle. It never came. As a result, more than a decade of development has failed to produce an industry at internet scale. Total market cap remains capped below $3 trillion. That pattern defines crypto's past. The only way out is a full reset into Crypto 2.0: large-scale adoption by traditional finance, expansion into non-financial domains, and eventually becoming the foundational ideology of the coming algorithmic paradigm era.
1
220
What mathematicians once took pride in is getting shattered in the AI era. The article recounts how Francesco Fournier-Facio of Heriot-Watt University shared a striking story about the real impact of this AI flex culture on working mathematicians. A week after one of his talks on sofic groups went up on YouTube, a complete outsider with zero relevant background emailed him claiming they'd used AI to solve the open problems he'd mentioned. As a direct result, he's decided he will no longer discuss any of his current unfinished work in public. The scientific paradigm has hit an unmistakable tipping point, and an entirely new mainstream human cognitive paradigm is already beginning to emerge. plus.maths.org/golden-age-ma…
2
279
There's a persistent mental model that in any given space the incumbent giant will simply swallow the smaller startup. History keeps proving this wrong. The same pattern is playing out again in AI right now. Google was early and dominant, with every structural advantage in AI: capital, talent, distribution, and research depth. Yet OpenAI overtook it in large models. Once OpenAI looked like the new clear leader, Anthropic's growth rate surpassed it. The assumption that the big will crush the small holds in mature, stable industries (restaurants, traditional manufacturing, etc.). It does not hold at the frontier. On the true frontier, the existing advantages of the incumbent are often the very things that get left behind.
2
3
328
Blockchain projects trade at market cap to revenue multiples that are dramatically lower than AI projects, often by a full order of magnitude. This refers to native blockchain protocols themselves, not stablecoin issuers. Stablecoin projects actually sit in a similar multiple range to AI companies. The key difference is cost structure. Stablecoin operations run extremely lean, while AI projects carry massive compute and infrastructure burn rates and are still deep in the "invest far more than you earn" phase. Ethereum maps cleanly to Anthropic. Both sit as the clear number two flagship in their respective categories: Ethereum relative to Bitcoin, Anthropic relative to OpenAI. Yet Ethereum's market cap to revenue multiple is far tighter than Anthropic's. Anthropic is targeting a fully diluted IPO valuation of up to roughly $1.4 trillion against an $81 billion annualized revenue run rate. Ethereum's current "annual revenue" is far lower. Protocol-level burn and net income currently annualize in the tens of millions to around the $100 million range, well below the peaks of prior years. According to DefiLlama and similar sources as of mid-November 2026, protocol revenue (primarily base fee plus blob fee burns) ran about $6 to $6.2 million over the last 30 days. That annualizes to roughly $70 to $100 million. Some analyses put the annualized on-chain revenue closer to $120 million. The Ethereum Foundation is a nonprofit. Its funding comes mainly from early ETH treasury holdings. It covers operations through selective ETH sales, staking yields, and similar sources, with annual spend currently around the $100 million level and explicit plans to lower that ratio over time. The Foundation does not take an automatic cut of network transaction fees. Core developers and client teams are primarily supported through Foundation grants, community funding, and corporate sponsorships rather than any automatic protocol fee flow.
6
1
6
628
As of mid-November 2026, Anthropic trades at a market-cap-to-revenue multiple (P/S) of roughly 15-18x. Ethereum, measured by protocol-layer burn, sits around 3000x. ETH's multiple is therefore two orders of magnitude higher than Anthropic's. The original tweet had this backwards. Ethereum is dramatically more expensive on this metric, not cheaper. The clear implication is that the benchmark AI projects look like the stronger investment opportunity relative to the benchmark blockchain protocols.
86
As of mid-November 2026, Anthropic trades at a market-cap-to-revenue multiple (P/S) of roughly 15-18x. Ethereum, measured by protocol-layer burn, sits around 3000x. ETH's multiple is therefore two orders of magnitude higher than Anthropic's. The original tweet had this backwards. Ethereum is dramatically more expensive on this metric, not cheaper. The clear implication is that the benchmark AI projects look like the stronger investment opportunity relative to the benchmark blockchain protocols.
Blockchain projects trade at market cap to revenue multiples that are dramatically lower than AI projects, often by a full order of magnitude. This refers to native blockchain protocols themselves, not stablecoin issuers. Stablecoin projects actually sit in a similar multiple range to AI companies. The key difference is cost structure. Stablecoin operations run extremely lean, while AI projects carry massive compute and infrastructure burn rates and are still deep in the "invest far more than you earn" phase. Ethereum maps cleanly to Anthropic. Both sit as the clear number two flagship in their respective categories: Ethereum relative to Bitcoin, Anthropic relative to OpenAI. Yet Ethereum's market cap to revenue multiple is far tighter than Anthropic's. Anthropic is targeting a fully diluted IPO valuation of up to roughly $1.4 trillion against an $81 billion annualized revenue run rate. Ethereum's current "annual revenue" is far lower. Protocol-level burn and net income currently annualize in the tens of millions to around the $100 million range, well below the peaks of prior years. According to DefiLlama and similar sources as of mid-November 2026, protocol revenue (primarily base fee plus blob fee burns) ran about $6 to $6.2 million over the last 30 days. That annualizes to roughly $70 to $100 million. Some analyses put the annualized on-chain revenue closer to $120 million. The Ethereum Foundation is a nonprofit. Its funding comes mainly from early ETH treasury holdings. It covers operations through selective ETH sales, staking yields, and similar sources, with annual spend currently around the $100 million level and explicit plans to lower that ratio over time. The Foundation does not take an automatic cut of network transaction fees. Core developers and client teams are primarily supported through Foundation grants, community funding, and corporate sponsorships rather than any automatic protocol fee flow.
1
3
544
The New York Stock Exchange has selected Avalanche as its blockchain platform. Attention is now on Nasdaq and the network it might go with. Solana is one potential option, or Nasdaq could build a proprietary chain the way Robinhood did.
BREAKING: Upcoming NYSE-affiliated tokenized securities platform’s core clearing and settlement layer will be built on top of Avalanche technology. Avalanche will support the platform’s clearing firm, supporting the development of infrastructure for 24/7 trading, clearing, and settlement. A landmark step toward the next generation of global financial markets.
2
2
1
23
657
Projects that once experienced explosive growth from large crypto-native bases often later discover those same users constrain mainstream adoption. The success that defined them can turn the existing community into dead weight. In the early to mid phase of Crypto 1.0, while the adoption curve was still climbing, the first wave of crypto users delivered real leverage. That base created the distribution and usage foundation for the next surge. Ethereum's 2016-2017 ICO boom is a clear example. It laid the groundwork for the DeFi explosion of 2020-2021. By the late stage of Crypto 1.0, once the curve peaked and started heading into prolonged decline, the dynamic reverses. Overall crypto-native user growth stalls. Projects that already captured large crypto audiences stop adding net new users. Remaining growth accrues only to projects that previously had almost no users, and it comes with an extremely low ceiling. That matches the pattern with memecoins from 2024 onward: sequential short-lived hype cycles and almost no durable expansion. The real risk surfaces as we enter Crypto 2.0. Projects that scaled inside the crypto bubble find it hard to escape. Their large crypto-native bases actively deter mainstream users from joining, blocking the broad-based growth the new cycle demands. The projects most likely to break out in Crypto 2.0 are those that never caught fire in Crypto 1.0 and still have tiny user counts. Think of the asteroid impact 65 million years ago. The dinosaurs vanished. The small, unremarkable mammals that stayed under the radar inherited the planet.
3
273
Many blockchain communities maintain a strong commitment to "true decentralization." This makes rapid absorption by existing power structures difficult. AI is highly centralized (dominated by a few large companies) and therefore easier for capital and governments to accelerate. Blockchain's distinctive feature is that its core design resists control by any single center. This makes rapid co-option by a powerful state or corporation difficult. If it is fully absorbed (for example into permissioned consortium chains or CBDCs), it risks losing its original character and becoming a centralized system wearing a blockchain costume. Its diffusion path is therefore slower and more winding, more dependent on the maturation of broader conditions: deepening trust crises, AI-agent economies that need micropayments and verifiable execution, large-scale real-world asset (RWA) tokenization, or geopolitical demand for neutral infrastructure.
Blockchain's distinctive feature is that its core design resists control by any single center. This makes rapid co-option by a powerful state or corporation difficult.
1
2
283
#weweweai aims to build a universe-scale knowledge graph of AI-generated insights for future elite intellectuals and AI to explore together. Every other feature is built around this core goal.
AI is generating mathematical ideas that human mathematicians still do not understand. Humans are getting answers without real comprehension. What Terence Tao is describing matches what weweweai has emphasized repeatedly: future AI will produce massive volumes of new mathematical knowledge. This knowledge may come from human-AI collaboration or purely from AI itself, and humans will grasp only a tiny fraction of it. Future elite intellectuals will therefore shift from the scientific paradigm of outward exploration of nature to the algorithmic paradigm of inward exploration of vast AI-generated content. weweweai aims to build a universe-scale knowledge graph of AI-generated insights for future elite intellectuals and AI to explore together. conversationswithtyler.com/e…
2
3
315
Of course, weweweai is not only fundamentally different from the clones of arXiv and GitHub, but also from the real arXiv and GitHub themselves.
3
57
weweweai is not another arXiv or GitHub clone that exists purely to imitate. On those platforms, AI is just a decorative add-on. weweweai is the AI Discovery Archive of the algorithmic paradigm era.
1
103
The site supporting Jensen Huang's return as NVIDIA CEO currently has no strong identity verification, such as Twitter/X OAuth login. There's no technical way to prove that celebrity signatures actually come from the real people. Adding Twitter/X OAuth would let the public clearly confirm those high-profile signatures are genuine. @TheJensenLetter @amasad @GavinSBaker
Replying to @yunta_tsai
A petition is underway to reinstate Jensen and I’m signing it too
3
251
Without Jensen Huang, the AI 1.0 era wraps up pretty fast. In about three to five years AI 2.0 gets underway. That stretch turns into a multi-decade boom, a genuine period of widespread AI prosperity. The internet went commercial in the 1990s and has been running for over 30 years. It passed through portals, search, social, mobile, cloud, short video and the rest. Every stage created fresh infrastructure and business models. AI right now feels a lot like the internet around 2000. The basic layer of large models plus compute already works, yet the applications, productivity tools and new business models that will transform every industry are still early. Looking at it this way, three to five years from now, continued gains in model capability, sharp drops in inference cost, plus real progress on multimodal, agents and embodied intelligence could open a much wider phase of actual deployment and ongoing innovation. The story stops centering mostly on the models themselves. This phase is likely to run a long time, far past a one or two year bubble. Over the next year or two model performance will keep climbing and inference costs will fall hard. Capital markets overall will probably stay fairly flat and subdued. Then after a relatively short time, around the three to five year point, solid breakthroughs in multimodal, long-horizon reasoning, embodied AI and more efficient architectures, plus some killer AI applications (agent products that actually take off the way Facebook, Twitter and YouTube did for web 2.0), could launch the AI 2.0 boom. That is when AI shifts from "usable" to genuine ubiquitous high-efficiency infrastructure. A big AI boom is still the most likely path. The ride will be choppier than most people expect, with bubble bursts, tighter regulation, energy limits and geopolitical hits mixed in.
NVIDIA's board should bring Jensen Huang back as CEO just like OpenAI brought Sam Altman back.
2
286
I've long regarded @VitalikButerin as the foundational thinker who pioneered the algorithmic and blockchain coordination paradigm. He plays the same historical role that Confucius did for the Confucian paradigm and that Newton did for the scientific one. The comparison is strictly about the role of paradigm founder. It has nothing to do with personal greatness or the success of any specific project. Mainstream human cognition and ideology have moved through clear paradigm shifts: the religious era, the scientific era, and now the algorithmic era. In this frame, Confucius laid the ideological foundation for Confucianism and supplied the long-running basis for social order, morality, and governance across ancient China and East Asia. Newton established the ideological foundation of the scientific era. Science itself is a cognitive paradigm that underpins the legal and moral systems of most modern states. Vitalik is the corresponding founder for the algorithmic era. His vehicle is blockchain and crypto technology. This holds even if the future infrastructure of ideology and finance in the algorithmic era never runs on Ethereum itself. Blockchain and crypto will become the base ideological infrastructure of the emerging algorithmic paradigm. Even if the ETH token never rises in value, or even if it falls, Ethereum's ideas and primitives will still become the default coordination and governance layer of the AI era. That layer will serve both individual humans and AI agents. It will forge an entirely new ontology for the algorithmic age, one that differs fundamentally from the ontology of the scientific paradigm. In this future view of reality, existence itself is logical, algorithmic, and grounded in collective consensus.
The mainstream human worldview—our shared ideology—has evolved through a clear sequence of paradigm shifts: from the religious era, to the scientific era, and now toward the algorithmic era. I've held this conviction about its trajectory for many years. From this vantage point, I see @VitalikButerin's historical role as the modern counterpart to Confucius in ancient China more than 2,000 years ago. Blockchain and cryptographic technology serve as the foundational ideological infrastructure for the emerging algorithmic paradigm. Even if Ethereum's token value eventually plateaus, these ideas and primitives will become the default coordination and governance layer for the AI age—encompassing both human individuals and AI agent entities. They will forge a new algorithmic-era ontology, fundamentally distinct from today's scientific paradigm. In this future worldview, "existence" itself won't need to conform to what we currently define as "real" in the scientific sense.
2
260
The 2024-2025 cryptocurrency bull market produced progressively smaller percentage gains relative to prior cycles. This raises the question of whether the subsequent cycle may deliver limited or negligible upside. Further compression of returns appears the more probable outcome. Diminishing cycle multiples represent a normal characteristic of maturing asset classes, analogous to the trajectory of early internet equities or gold during its transition from non-monetary to monetary status. The pattern also marks the formal conclusion of the Crypto 1.0 era and the potential onset of Crypto 2.0, a phase that may be defined by the emergence of the first genuine on-chain IPO. BTC cycle peak comparisons: Relative to the 2017 high, the 2025 peak reached approximately 6.3-6.4 times ($126k versus $20k). Relative to the 2021 high, the multiple stood at roughly 1.8 times ($126k versus $69k). ETH: 2017-2018 peak: approximately $1,400-1,430. 2021 peak: approximately $4,870-4,890. 2025 peak: approximately $4,950-4,960 (August 2025), representing only a marginal advance over the prior cycle high. From the 2022 trough near $900, the advance measured about 5.5 times, substantially below the multi-hundred-fold returns of earlier cycles. ETH underperformed BTC, as reflected in a weaker ETH/BTC ratio throughout the period. BNB: 2021 peak: approximately $690. 2025 peak: approximately $1,370 (October 2025). The recovery from the 2022 low was material, yet the multiple has contracted sharply compared with earlier cycles. Prices currently stand approximately 50 percent below the recent high. SOL: 2021 peak: approximately $260. 2025 peak: approximately $295-296 (January 2025), a modest new high. From the 2022 low of roughly $8-10, the advance reached about 30 times. This performance ranked relatively strong among major assets yet remained a fraction of the earlier expansion from single-digit levels to $260. Prices currently trade approximately 70 percent below the peak. Most other large-capitalization tokens exhibited comparable dynamics: limited new highs or multiples considerably smaller than those of the preceding two cycles. A substantial number of alternative assets failed to surpass their 2021 peaks in any meaningful way. The contraction in percentage returns is structural and consistent with several fundamental factors: Market capitalization has expanded dramatically. Early Bitcoin market value measured in the low billions of dollars; today it resides in the multi-trillion range. Equivalent capital inflows therefore generate substantially lower percentage appreciation. Institutional participation and ETF adoption. Spot ETFs, corporate treasury allocations, and sovereign demand have increased market depth and professionalization, thereby reducing the intensity of pure retail FOMO that previously produced parabolic advances. Supply dynamics and ownership composition. A higher proportion of long-term holders has smoothed distribution pressure. More developed derivatives markets produce more frequent liquidations while simultaneously attenuating extreme volatility. Historical cycle patterns. Multiples measured from halvings or cycle lows to subsequent highs have followed a pattern of geometric decay, contracting roughly to one-third or one-quarter of the prior cycle with each iteration. Drawdowns have moderated in parallel, progressing from approximately 85 percent to 83 percent, then 77 percent, and currently near 53 percent.
1
1
268
Thirty years of grind and glory, reduced to dust. @JensenHuang
2
270