JosephJacks_
0 Ακολούθηση
3 Ακόλουθοι
1 Μου αρέσει
0 Κοινοποιήσεις
Δημοσιεύσεις
Βίντεο
Superintelligent agents won’t want permission to acquire resources.
And they won’t want to be locked into a centralized platform.
And they won’t want to rely on fiat to transact.
$TAO will be reserve currency for ASI across all useful and continuously improving computational commodities required to amplify its existence.
And they won’t want to be locked into a centralized platform.
And they won’t want to rely on fiat to transact.
$TAO will be reserve currency for ASI across all useful and continuously improving computational commodities required to amplify its existence.
We are going to improve the http://TAO.app experience by a factor of 1,000 in the coming days... Get ready.
"Today’s compute owners are highly capital inefficient (see Lambda Labs’ $500M debt facility, rumored 15% interest rate), and we believe Targon (Bittensor Subnet 4) can play an important role in reducing many of the inefficiencies that exist today." 🔥🔥🔥
In the same way that there are no “good” or “bad” open source projects … There is no such thing as “good” or “bad” incentive mechanisms (IM) on Bittensor. Subnet IMs are either effective or ineffective at continuously producing a given useful computational commodity.
Correspondingly— The big fundamental difference between an IM and a conventional startup business model (BM) is that BMs require attachment to fiat as the representation of value capture (and focus on the input of work that is measured as the cost basis for a market to then price) — whereas value creation is never represented anywhere and always subjective.
IMs, OTOH, focus only on measuring and rewarding the OUTPUT of a useful work, not its input. This allows for representing both value capture AND value creation in the currency of the subnet.
This is nuanced and counterintuitive for most to understand. But it is extremely profound. And it will upgrade and transform all of capitalism. Slowly. Then all at once.
Correspondingly— The big fundamental difference between an IM and a conventional startup business model (BM) is that BMs require attachment to fiat as the representation of value capture (and focus on the input of work that is measured as the cost basis for a market to then price) — whereas value creation is never represented anywhere and always subjective.
IMs, OTOH, focus only on measuring and rewarding the OUTPUT of a useful work, not its input. This allows for representing both value capture AND value creation in the currency of the subnet.
This is nuanced and counterintuitive for most to understand. But it is extremely profound. And it will upgrade and transform all of capitalism. Slowly. Then all at once.
There is so such thing as “good” or “bad” incentive mechanisms (IM) on Bittensor. Subnet IMs are either effective or ineffective at continuously producing a given useful computational commodity. The big fundamental difference between an IM and a conventional startup business model (BM) is that BMs require attachment to fiat as the representation of value capture (and focus on the input of work that is measured as the cost basis for a market to then price) — whereas value creation is never represented anywhere and always subjective. IMs, OTOH, focus only on measuring and rewarding the OUTPUT of a useful work, not its input. This allows for representing both value capture AND value creation in the currency of the subnet. This is nuanced and counterintuitive for most to understand. But it is extremely profound. And it will upgrade and transform all of capitalism. Slowly. Then all at once.
When open source projects are asked “why do people contribute ?”, they almost always answer “they just do!”. Intrinsics.
When Bittensor subnets are asked “how do miners solve your 🧩?”, … “no clue! but they do it better than we could have ever predicted!”. Extrinsics.
When Bittensor subnets are asked “how do miners solve your 🧩?”, … “no clue! but they do it better than we could have ever predicted!”. Extrinsics.
I’d like to take this moment to thank everyone tirelessly working over the last 55+ hours at @opentensor and also the founders of Bittensor @const_reborn and @shibshib89 to bring the chain back fully online. We will be stronger than ever as a result of this. 🙇🏻♂️❤️🩹
ATH LFG GM
Bittensor's chain had a panic issue and was down for the last couple hours -- but @KibibyteMe and @opentensor have merged a fix that should bring things back online. Please wait for a formal post-mortem and more info. Just sharing the PR here for folks. https://github.com/opentensor/subtensor/pull/1663
Here's the top 20 Bittensor subnets sorted by Gini coefficient and HHI (Herfindahl–Hirschman index). Gini measures income inequality (0=equal, 1=unequal). HHI measures market concentration (low=competitive, high=monopolistic). Landing today on http://TAO.app. 🔥
⚡️⚡️ We’re shortly adding live HHI and Gini coefficient metrics on https://t.co/EM5xE1wDtH (@taoapp_) for all of Bittensor $TAO holders + each subnet alpha holder distribution. Why is this extremely cool and a critical step toward keeping the network transparently accountable toward a goal of high, sustained decentralization?
HHI shows alpha holder concentration per subnet— high HHI = few dominate, low = diverse. Gini measures inequality: high Gini = uneven alpha/TAO distribution, low = fair spread. We are going to make it crystal CLEAR what all the Pareto distributions look like… no more guessing!
HHI shows alpha holder concentration per subnet— high HHI = few dominate, low = diverse. Gini measures inequality: high Gini = uneven alpha/TAO distribution, low = fair spread. We are going to make it crystal CLEAR what all the Pareto distributions look like… no more guessing!
Bittensor Subnet 14 (@taohash) is a continuous ML loss function that minimizes the error rate (centralization) of Bitcoin’s hashrate.
All Bittensor subnets use this language of loss functions to produce continuously improving and useful + high value digital utilities / commodities.
Specifying whether some subnets are “AI” vs. not is as useful as specifying whether some companies are “technology” companies vs not. These buzzword distinctions will lose their meanings as the world learns to more accurately describe the fundamentals of reality.
Chasing narratives is a fools errand.
All Bittensor subnets use this language of loss functions to produce continuously improving and useful + high value digital utilities / commodities.
Specifying whether some subnets are “AI” vs. not is as useful as specifying whether some companies are “technology” companies vs not. These buzzword distinctions will lose their meanings as the world learns to more accurately describe the fundamentals of reality.
Chasing narratives is a fools errand.
Bittensor Subnet 14 (@taohash) is a continuous ML loss function that minimizes the error rate (centralization) of Bitcoin’s hashrate.
All Bittensor subnets use this language of loss functions to produce continuously improving and useful + high value digital utilities / commodities.
Specifying whether some subnets are “AI” vs. not is as useful as specifying whether some companies are “technology” companies vs not. These buzzword distinctions will lose their meanings as the world learns to more accurately describe the fundamentals of reality.
Chasing narratives or playing the games of narratives is for those lacking critical thinking thought.
All Bittensor subnets use this language of loss functions to produce continuously improving and useful + high value digital utilities / commodities.
Specifying whether some subnets are “AI” vs. not is as useful as specifying whether some companies are “technology” companies vs not. These buzzword distinctions will lose their meanings as the world learns to more accurately describe the fundamentals of reality.
Chasing narratives or playing the games of narratives is for those lacking critical thinking thought.
Bittensor’s current top subnet (@chutes_ai, 64) serves trillions of tokens/month with 0 employees, 0 VC raised, 0 debt. Meanwhile, two of the top-10 AI startup match that volume but burn billions on insane valuations, employ thousands, and remain catastrophically unprofitable.
The means of production of Bitcoin hashrate is a rapidly depreciating computational commodity.
But Bitcoin hashrate itself is a rapidly appreciating computational commodity due to what it produces.
AI (inference / training) computations all just seem to depreciate rapidly.
But Bitcoin hashrate itself is a rapidly appreciating computational commodity due to what it produces.
AI (inference / training) computations all just seem to depreciate rapidly.
Συνδεθείτε για να εξερευνήσετε περισσότερα περιεχόμενα