Const vs. Striker: The Root Reborn Debate Explained for Everyone Who Missed It

If you weren’t in the 18th June Novelty Search, here’s a rundown of what happened.

Const got on stage to walk through Root Reborn, the proposed Bittensor change everyone’s been arguing about, and then invited Striker to come up and debate him on it, with Mark Jeffrey moderating.

Let’s walk through the whole thing.

First, what is Root Reborn?

Right now, when you stake TAO on root, here’s the mechanical reality. Holding root TAO gives you a tiny slice of economic weight on every subnet in the network. Const’s example: a thousand TAO gives you about 180 alpha worth of exposure spread across every subnet (because root sits at roughly 18%). You’re earning dividends on all of that.

The problem is what the chain does with those dividends. Instead of handing you a messy pile of dust-sized alpha tokens from 100+ subnets, which Const admitted would be “a UI disaster” that nobody would ever bother claiming, the chain just automatically sells all of it into TAO and credits your account. Clean experience, but it creates a brutal side effect: roughly 1,000 TAO worth dTAO sell pressure every single day (around $276K), pouring out of the network like clockwork.

Root Reborn changes that. Instead of auto-selling your yield into TAO, validators can set a “weight vector”, basically a basket, and your yield gets accumulated into that basket of alpha tokens rather than dumped. The basket sits in an escrow account owned by nobody, and you own a proportional share of it that grows over time. Your principal is never touched; only the yield is ever at stake. And if you don’t want to play, you can just keep claiming manually and your experience is identical to today.

Mark summed it up by saying: “So all validators are effectively running a hedge fund.” Const pushed back slightly on the wording; he prefers “yield optimizers,” since they’re not actively day-trading, just curating allocations.

Mark’s opening thoughts

Mark laid out why he thinks part of this is clearly good. In his opinion, a lot of subnet valuations, maybe 20 or 30 of them, are sitting far below where they should be, and one big reason is that the chain is robotically selling 1,000 TAO of subnet tokens a day.

He says: let the market sell if it wants to, but don’t make the chain do it like a robot every single day. Killing that automatic sell pressure, he thinks, stops the network from “kneecapping itself.”

Where Mark got fuzzy about Root Reborn, and where he suspected most people get fuzzy, was the power this hands to validators and what exactly they do with it. That’s what set up the debate.

Const’s economic standpoint

Const’s intends that TAO should behave less like “a token that auto-sells everything it makes” and more like “a bank of assets” that holds and curates a basket.

His core points:

The yield is measurable and competitive. Validators are already doing yield optimization; this just lets them set weights on-chain, where everyone can see and compare them. Stakers chase the best yield, validators compete to deliver it, and bad allocators get abandoned. He kept returning to one question for Striker: over the long term, will stakers make more money or less money with Root Reborn? He thinks they will make more.

It re-engages the validator class. There are smart people on the validator side who currently have almost nothing to do with curating the network. This gives them a real job again, building baskets, making investment cases, and brings some structure and order at the chain level. He compared it to how real markets work: the Nasdaq and S&P aren’t run by day traders; institutions provide ~90% of the liquidity and give the market stability.

It’s strictly non-negative for subnets. This was a point he hammered. If your subnet doesn’t get picked by any validator, your sell pressure is exactly the same as before, and nothing gets worse. If you do get picked, yield flows into your alpha pool as buy pressure instead of being sold. So in his opinion, subnets can only be neutral or better off, plus they get a new credibility signal (validators choosing you means an institutional-style vote of confidence).

Tax efficiency. Right now your yield is realized into liquid TAO every few blocks — a taxable event each time. If it instead sits as alpha in an escrow account owned by no one, Const argues you’re not realizing it until you actually claim, similar to how you’re not taxed on a fund’s NAV until you sell out. Although he doesn’t want to dwell too much on this point, he would rather “leave it to the lawyers and accountants”.

Striker’s rebuttal

Striker came loaded with notes and split his attack into ideology and exploitation vectors.

His main points:

“This doesn’t stop dilution, it just kicks the can.” The chain is an incentive chain; it gives out TAO, and that TAO becomes sell pressure. Deferring a dTAO sale via the current root mechanism doesn’t remove that sell pressure; the same supply still hits the market eventually, and he argued there’d actually be more leakage this way.

Const partly conceded this, agreeing that deferred sell pressure is the same total sell pressure. His counter was that forced, automatic, every-block selling is psychologically worse for price discovery than letting people choose when to sell.

The tax argument doesn’t hold in his jurisdiction. Striker points out that validators aren’t registered funds, so the “you’re not taxed on a fund’s NAV” logic doesn’t apply unless you want every validator to become a registered hedge fund. And if you want tax deferral via an escrow contract nobody owns, you could just do that with plain TAO; you don’t need the whole subnet-basket machinery for it.

This forces work onto people who don’t want it. He compared root stakers to people holding dollars in treasury bills; they just want a safe, passive yield. Root Reborn effectively forces those passive holders to do due diligence on a bunch of validators running hedge funds.

People who want to invest in subnets already can, they can take their root yield and buy whatever subnet they like. Why insert a middleman deciding where the tokens go? “There should not be a middleman that is deciding for you where those tokens are going. It will be a nightmare. It’s too overcomplicated.”

The validator corruption history. Striker was active in the pre-dTAO era and says he fought the shady validators back then. The vast majority of validators had massive conflicts of interest. That corruption, he argued, is the entire reason the network moved from pre-dynamic to dTAO in the first place. So why take multiple steps back toward a pre-dynamic mechanism?

TAO holders are already productive. He rejected the premise that passive root holders “aren’t doing anything.” The TAO price is the backbone of the whole incentive mechanism. If TAO has no value, all of dTAO collapses. Holders buying and holding TAO is providing value, the same way bond buyers provide stability.

The real problem is quality, not sell pressure. He said Bittensor isn’t stagnating because of sell pressure. It’s stagnating because it needs more quality. Artificially inflating prices is “a short-term measure for long-term pain.” Worse, forcing subnets to start pitching and begging validators for allocations injects political games and friction that won’t boost quality. “Demand will come from quality. Demand doesn’t come from doing these weird technical changes.”

Where it netted out

The two sides never fully converged, but the shape of the disagreement got clear:

Const’s position: This standardizes something validators already do manually (Trusted Stake, Mentat Minds, Crucible all do it today), makes it transparent and on-chain, kills a daily robotic sell, and can only be neutral or positive for any given subnet. The feedback loop (stakers can switch validators, validators absorb risk through their stakers) is what makes it different from the old, consequence-free root network.

Striker’s position: It’s an overcomplicated step backward toward a pre-dynamic model that the network already rejected for good reason. It doesn’t actually remove dilution, it forces passive holders and subnet builders into political games and due diligence work they didn’t sign up for, and it distracts from the only thing that actually matters, which is building quality subnets that generate real demand.