iztali

@simioape 102 and 96? Yeah I don’t see them holding up once conviction is live and 3 gets going again 5, 24 also fall in this category. 3 as a const sub will eat them all

there are two subnets pumping because we needed to prove a point with sn3 (we won't)

The Covenant/templar situation is a result of misaligned incentives. It was going to be more profitable, certainly short term for them to sell hype on promising early results and dump on holders after collecting 10s of thousands of $TAO in emissions, rather than to attempt to perform another massive decentralized training runs in hopes of positive results They made the decision which was most aligned with their financial incentives, purely selfish, but certainly not unique or irrational. This is the conversation which needs to be opened This is the same misalignment causing a majority of subnets to not actually be running code, and stalling launches for months upon months We need some sort of mechanism to not be actively giving TAO over to subnets not paying miners. At some point we may want to consider sacrificing a certain degree of elegance in the mechanism for improving the functional realities Bittensor will come out of this stronger, more aligned, and will prevail, as we have throughout every exploit or dysfunctional phase.

OpenRouter has updated our provider status after verifying our privacy policy thanks to our recent updates Chutes is in their default routing now! openrouter.ai/provider/chutes

Half the subnets are burning all miner emissions. 3/4 are burning over 50%. It's a dangerous trend beneath the surface. Simple solution, @const_reborn Subnet_Emission = Raw_Emission * (1 - Burn_Rate) x.com/DrocksAlex2/st…

On the @theallinpod this week, @chamath asked @nvidia CEO Jensen Huang about decentralized AI training, calling our Covenant-72B run "a pretty crazy technical accomplishment." One correction: it's 72 billion parameters, not four. Trained permissionlessly across 70+ contributors on commodity internet. The largest model ever pre-trained on fully decentralized infrastructure. Jensen's answer is worth hearing too.

we’re getting close to witnessing what could be the most violent tokenomic flywheel in the history of crypto $TAO an ETH ICO demand shock intersecting with a BTC halving supply shock during a hard takeoff with AI and the event horizon of singularity within view

@simioape What about something like Templar? Where do we draw the line of usefulness where having a decentralized replica actually matters—and is worth investing in?

@iztali_eth Bittensor is falling in the crypto retardation of building things that already exist and nobody asked to be decentralized for them to use it. TEEs for example is great as an accomplishment but what's the actual % of claude code users thinking of privacy before using the product.

for whatever reason i never see this shared about chutes endless tao sent their way for less revenue, less usage and less gpus the worse btc does, the worse these numbers will look.

@simioape Why? I’ve come to the same conclusion, mostly that the R/R of low price subnets is so much higher and downside capped by the dereg floor. Do you have a price cutoff other than top 15?

DO NOT INVEST IN SUBNETS IN THE TOP 15

This @numinous_ai agent is the most accurate predictive agent in the world (you can ask it anything and it returns its best probabilistic estimate, for free rn) It beats grok, gemini, openai, and all other comparative services. Its not just an LLM though, the agents are finetuned to stay on top of prediction markets, news feeds, twitter searches etc and all that is aggregated each query to give you an estimate of some future result. Its a perfect incentive mechanism because miners are rewarded for accuracy over thousands of synthetic tasks each day, they can't overfit so they just get better. Quite litterally "mining predictive intelligence". Really cool.

@simioape I think net flow emissions has helped this. Extractors are now limited to extracting from their own liquidity pool instead of mining new TAO. Ofc there’s still issues like SN29, SN115, SN119 etc. but it’s a step in the right direction

Back when I posted this some pseudo researchers rushed to post how the bittensor halving would make every problem go away because brrrrrrrr and because brrrrrrrrr. They don't understand what makes bitcoin great and what's stopping (or killing) bittensor's potential now.

@SiamKidd Do you expect a major run where sum of subnets goes to >2 and we see large caps like Chutes and Ridges do a 5-10x? Or are those days over and only small-mid caps present asymmetric opportunities?

I keep seeing comments in subnet chats like "revenue is a meme". So I thought I'd share a few observations. In no particular order: When a subnet is launched, this is akin to a private company IPOing. The subnet tokens (alpha) is then available to trade and the subnet is public in every sense. It has a 4 month immunity period where the subnet has to get its shit together and show/prove/convince the market that their alpha is worth buying. If in 4 months the subnet hasn't convinced the public/market that they are worth buying, there will be more sells than buys, emissions will be 0% or close to it and their alpha price declines. Eventually, they will have the lowest price and then be de-registered. Game over. Whatever percentage of the alpha supply a holder has, they will get that amount of TAO that was in the liquidity pool. So what's the game or arena here? The game is to create a subnet that provides value. In whatever shape or form that may be. It needs to show/prove to the market (dTAO investors) that what they are doing is going to be good. In my personal opinion, there really are only 2 subnet categories. You can either be a business or a research lab. That's it. There's no real in-between here. It's much like the FTSE or S&P500. Those companies are businesses. They make revenue and profit and the market buys their stock based on those metrics and forward looking metrics. What you don't find in those market places (indicies) are Harvard University or some independent research lab. The research labs you may see are actually research labs that have turned into businesses and are now selling/monetising the fruits of that research. I think it's naive to think that dTAO is any different. You can either be a Business Subnet that can use Bittensor to build your product/service and show people that your product or service is going to be raking in lots of revenue and have some alpha buyback plan in place. Therefore showing that value is being driven back into your alpha. ORRRR you can be a research type subnet. Which is great. I truly believe it's the research subnets that will one day punch OpenAI in the face and they won't be able to catch up. Or for a subnet to make game-changing advancements in AI. BUT, yes, there is a BUT...these research subnets need time. They need time to work on their game-changing stuff which requires money. In the real world, research labs have a benefactor. For instance it could be a pharmaceutical company willing to burn $20m a year on R&D with the goal of discovering a new drug in 5-10 years where they can make billions. But what about Bittensor research subnets? What if a great potential subnet needs 5 years to build/moon? At some point, the market will work this out. Game Theory Optimal thinking will kick in and investors will naturally come to the conclusion of "I love the subnet and what they're doing. They're a great team, but they're X years away from breaking out and being able to drive real value to their alpha. So I'll just buy in closer to the time... In the meantime, I'll just buy subnets that are making revenue and doing regular buybacks". A normal retort to this is that it doesn't matter, the APY from holding for 3-5 years will massively put you in profit, even if TAO price goes down and alpha price goes down. To which I completely agree. But 99.999999999999% of dTAO investors will never buy a subnet and sit on it for 3-5 years. For the average dTAO investor, 3-5 months is like 3-5 years... To the purist, this all sounds like blasphemy I know. But when dTAO was created in Feb 2025, I repeatedly said that many big, well respected and high priced subnets would quicker enter sustained periods of "down-only-chart-mode"...which many people disbelieved. But it happened... So back to the same question of what's the game/arena here? Well, the rules have now changed. And will likely change again this year. We now have a capped number of subnets, deregs, a rising dereg price, chain buying and net TAO inflow. Which is great! I love that there's a confined pressure cooker now forcing subnets to be better. If the cap were to lift, that pressure/forcing function would dissipate. But what this means is that for a subnet to get LP emissions, it needs constant buying (net TAO inflow). So because of all these game conditions, for the INVESTOR, there are different GTO moves here. Now isn't the place to expand on these moves, but basically, if there is a Business subnet that can get product market fit, attract revenue and do buybacks, that subnet will DOMINATE the subnet leaderboard and emissions. Even better if that subnet has great alphanomics as well like Hippius and Vidaio. Whereas a Research subnet will need either: i.) Constant community support to keep on buying it, knowing there are years before alpha value becomes evident. OR, ii.) A massive benefactor like Const. He has gone from 75% in root to 59.5% in root in about a month and will continue to allocate into subnets that are good for Bittensor. This is great! But what happens to the research subnets when that 172k TAO in root is gone and he's 0% root? What this means is that there's a finite window for research subnets to deliver. Is it 1, 2 or 5 years? No idea! Hopefully there will be other big entities that enter Bittensor with big money that can happily allocate with a long term horizon to help the long tail moonshot subnets. (Quick tangent, one things I've observed over the last 2 months is the dereg price continually moving up. It was 0.001T and it's currently about 0.0029T. This rising tide means subnets are rapidly trying to get buyer support or revenue for buy backs to avoid dereg. Mark and I have helped 3 subnets out of dereg so far as we genuinely think those teams didn't deserve to dereg, but we've stressed we can't just keep buying them out of dodge. They have to deliver as our funds are finite). So for the meantime, I don't believe Revenue is a meme. We are in the real world here and the easiest and quickest way to driving massive alpha value is making a bucket ton of revenue and doing alpha buybacks. People hate hearing this...but what's the alternative? Is this going to force some subnets to shoehorn in any old product to get to market quickly whilst sacrificing the big dream a bit? Most likely... Will investors eventually figure this all out and rotate more into the revenue generating subnets more? Most likely... Is there way LESS risk buying a low price/low liquidity subnet with a good team and soon to revenue product compared to a high priced subnet which has a lot further to fall? Most likely. Especially seeing as you don't get much of a haircut during dereg. Will zombie subnets that have done bugger all for 10 months continue to do nothing and do the odd buyback here and there to avoid dereg whilst building up their owner key reserve? And then eventually dereg with a minimal haircut? Most likely... Anywho, this is just my 2 cents. Don't shoot the messenger, these are just my observations. For the record, I'm so incredibly bullish on Bittensor and 2026 will be an epic year. There is simply too much talent and brains in this ecosystem for TAO to fail... In the meantime, these new game conditions are presenting investors some lovely GTO moves for their portfolio if they can figure it out...

too many people I know in the space that I respect rocking an EVMaverick for me not to own one. so now I do.

EVMavericks finally verified on OpenSea LFG

There's a guy out there currently borrowing $15,000,000 against 112 Cryptopunks at 13% APR He claimed 180 of them for a grand total of $200 in 2017 and now pays $5,500 per day in interest He's paid $3,000,000+ in interest since 2022 to borrow against them instead of selling them but in doing so has seen his punks portfolio go from ~$10,000,000 to $30,000,000+ during that time IMO the GOAT of NFTs

@Lazz_eth @yanez__ai I’m ready

Leg two is fast approaching Warned you about this multiple times Top 20 inevitable @yanez__ai (54)

@xavi3rlu @taoapp_ How do I cancel an order? One of my orders is stuck as “Failed” because of the chain issues

TWAP is LIVE on tao.app! 🚀 Tired of MEV bot sandwiches? Sick of slippage? Say goodbye with TWAP 📊 Spreads your trades evenly to minimize impact 📈 Trade smart 👉 @taoapp_ #Bittensor $TAO #DEX

Most people still don’t get what @Investing88ai is building. So here’s the breakdown: ✅ Team: ex-Goldman Sachs, Meta, Citadel — deep roots in traditional finance and AI ✅ Live product: db.investing88.ai — an AI-powered stock screener with fundamentals, sentiment, and real-time signals ✅ Dynamic Emission Control (DEC): First on Bittensor. Emissions now shift with miner quality. No more static waste — real efficiency. ✅ Miner feedback loop: 88 built DEC with their miners. No other subnet is this responsive. ✅ Roadmap: — US Stocks — Beta feature rollouts (multi-asset) — Global asset management + capital automation No bullshit. Methodical shipping from a real team — with real domain expertise. If $TAO is the decentralized AI economy, 88 is building its first Wall Street. #Bittensor #dTAO #Subnet88 #Fintech #Tokenomics

#Biττensor >> ∆ τ << #τₐcc                          > $TAO < Let’s talk about #SN90 – the scammer brain! At its head: @dotkrueger . I’ll let you discover his bio in the images below. You see, Fred, I’ve been on the Bittensor network since 2022. Always here to support and uplift teams. Always ready to give my time to promote the Bittensor ecosystem. But today, I’ve decided to sacrifice a bit of that precious time to expose you publicly. I don’t have any fancy degrees, and honestly, I shouldn’t stoop down to your gutter level. But I’ll make an exception, because someone has to step up and warn investors before they fall for the brilliance of your corrupt genius. Censorship on Discord, right? Silencing criticism? No problem. That only fuels my will to fight parasites like you. SN90 – it all started with big announcements from you. Bittensor became your new playground. Of course, you hyped your project before it was even active. The result? Investors got rekt. Their $TAO lost 99% of its value the moment your subnet launched. I told you, Fred — degrees don’t make the man. And you're not one. You’re just a professional scammer with a polished résumé. I’ll attach the long list of your past crypto rug pulls — it’s your signature, isn’t it? Scamming your peers while hiding behind your prestigious background. Take a look around, Fred. The world is run by corrupt elites like you. Educated, yes, but rotten to the core. Feeding off the masses. But this is Bittensor. And here, even with your 165K followers, you won’t get away with it. We’ll dismantle you, piece by piece. Day after day. Until you and your ego crawl back to wherever you came from. Bittensor community, there are plenty of quality subnets worth supporting. Why would you stake on SN90’s alpha and back a known scammer? This subnet must die. Make room for a real project. Get lost, Fred. See you soon. 💀