On October 8, a portfolio manager named Bons typed a four-word verdict onto X about a network worth roughly $30 billion: "XRP is Proof of Authority." Four days later โ October 12, a Monday โ a company called Evernorth was scheduled to ring the Nasdaq bell holding 473 million XRP. And Ripple's CTO Emeritus, David Schwartz, answered the accusation by quoting a tweet he had written in June.
Not October. June.
That four-month gap is the actual story. Not the label. Not the flame war. The gap between when the charge was filed in public and when the defense was pulled from the archive.
Here is what twenty years of watching this sector has taught me: the loudest technical arguments are almost never about technology. They are about legal exposure, priced in real time. When a fund manager calls a payment network "permissioned" eight days before its largest institutional holder goes public, the argument is not academic. It is a risk disclosure wearing a tweet's clothing. It is a pre-mortem, published in public.
So I will do what I always do with these fights. Ignore the adjectives. Read the mechanism. Then decide which side's claim survives contact with the code.
What XRP actually is, underneath the ticker
Most people arguing about this network have never read its consensus documentation. Let me fix that first, because the entire dispute collapses without it.
The XRP Ledger launched in 2012 โ older than Ethereum. Its consensus mechanism is called RPCA, the Ripple Protocol Consensus Algorithm. Structurally, it belongs to the Federated Byzantine Agreement family, the same lineage as Stellar's SCP. It is not proof of work. It is not proof of stake. It is a trust graph.
Here is how it works, stripped of marketing. Validators on XRPL each maintain a Unique Node List โ a UNL. That list is the set of other validators whose votes a given node will count. When enough of a validator's UNL agrees on a transaction set, consensus finalizes. The critical design choice: every operator chooses their own UNL. Nobody is forced to trust anyone.
Ripple, the company, publishes a default UNL โ the dUNL. Think of it as a starter kit. A new validator that does not want to research 150 peers can simply adopt Ripple's curated list and be operational in an afternoon.
That single sentence is the fault line of this entire controversy. Hold onto it.
Now the combatants.
On the prosecution side: Bons, of the investment fund Cyber Capital. His claims, pulled from the exchange directly โ the network runs on a permissioned node list, meaning validators must obey; that Ripple has been running closed-source code for the past two weeks; that this code becomes mandatory tomorrow and operators who disagree get kicked off the chain; that whoever controls the list are the "kingmakers."
On the defense side: Schwartz, the original architect of the ledger. His counter-claims โ anyone can run a node, a validator, a hub, a history server; the UNL grants no actual governance control; honest nodes enforce the rules and software choice resolves disputes, "just like any other network"; and XRPL is more censorship-resistant than Bitcoin or Ethereum.
Both men are intelligent. Both are partly right. That is precisely why this is worth a technical teardown rather than a retweet.
Premise one: is XRPL "Proof of Authority"?
No. Not in the way the term is defined.
Proof of Authority, as implemented by chains like early BSC with its 21 authorized validators, has a protocol-level whitelist. The validator set is fixed inside the software. To join, you are granted permission by a central party. It is a club with a doorman.
RPCA is structurally different. There is no protocol-level whitelist. Any operator can configure any UNL they want, including one that excludes Ripple entirely. In theory, a fully independent validator set could run XRPL tomorrow. Stellar, using a sibling consensus design, has run for years without the same intensity of governance accusations โ because the curator and the treasury holder are different entities there.
So the "PoA" label is rhetorical overreach. That is the honest verdict, and I say it as someone who has taken apart forty-plus consensus designs since the 2017 ICO cycle.
But โ and this is where the prosecution earns its keep โ "not PoA" is a much weaker defense than it sounds. Because the mechanism has a second layer that Schwartz's framing carefully omits.
The dUNL.
When the majority of validators simply adopt Ripple's published default list rather than curating their own, the network's effective trust graph is authored by one company. Nominal openness, factual centralization. This is not a hypothetical. It is the documented state of a large share of XRPL validators. Ripple curates; operators follow; the "kingmakers" Bons describes are real, even if his label for them is imprecise.
Here is the distinction I want you to keep: a permissioned network is one where you cannot join without permission. XRPL is a network where you can join freely โ and then most participants voluntarily defer to one curator. Those are different failure modes. The first is a gate. The second is gravity.
Bons accused XRPL of having a gate. It has gravity. He got the mechanism wrong and the concern right.
Premise two: the "closed source" charge
This is the most serious allegation in the entire exchange, and the least evidenced.
The mainstream understanding is that the XRPL client, rippled, is open source under the ISC license. That is verifiable. So when Bons says Ripple "ran closed-source code for two weeks," he cannot mean the whole client. He must mean the disclosure timing of a specific amendment or patch โ code that existed, executed, and was not published until after the fact.
Code doesn't care about press releases. Either the diff was public before activation or it wasn't. This is a binary, checkable fact, and the reporting around it provides no verification โ no commit hash, no repository, no timestamped diff. Just an assertion.
My audit discipline from the 2021 NFT contract work kicks in here. When someone claims a vulnerability, I do not accept the claim or reject it. I ask for the transaction hash. Bons provided none in the material I reviewed. So this charge sits in limbo: potentially the most damaging fact in the case, currently unproven.
If it is true, it is a governance-transparency crisis โ code that governs billions in settlement value, executed before the public could read it. If it is false, Bons's credibility takes the hit instead. Either way, the case turns on a fact nobody has produced.
Mark this as the single open variable. Everything else is commentary.
Premise three: the amendment that "forces" operators off the chain
Bons's claim โ that a change becomes mandatory tomorrow and dissenters get kicked โ sounds authoritarian. The mechanism is more subtle and, in one respect, worse.
XRPL amendments require roughly 80% validator support to activate. Once active, nodes that have not upgraded cannot reach consensus with the upgraded majority. They do not get "kicked." They get left behind. The network moves; they stay.
Schwartz frames this as "software choice resolves disputes." That is technically accurate and rhetorically evasive. Software choice does resolve disputes โ by letting the majority enforce its preference and rendering the minority inert. The 80% threshold means a determined 21% minority cannot veto an upgrade. In a network that markets itself on decentralized governance, that is a meaningful concentration of power, regardless of how gently it is described.
Code doesn't negotiate. It executes. The amendment mechanism is not a debate club; it is a ratchet. Pass the threshold, and the change is permanent, with no minority override.
To be fair: this is not Ripple unilaterally dictating terms. It is a supermajority rule baked into the protocol from early on. But "not unilateral" is not the same as "decentralized." A 21% bloc being structurally unable to block change is a governance property worth naming plainly.
The token economics are a separate war โ and that matters
Here is where most coverage goes wrong. It conflates two independent questions: is XRPL decentralized, and is XRP a sound asset?
They are not the same problem. XRP is a settlement asset. No staking yield. No liquidity mining. No protocol-level token incentive. It does not pay new money to old holders, because it pays nothing at all. The "is this a Ponzi" question โ the one I modeled in spreadsheets across ten DeFi protocols in 2020 โ simply does not apply here. There is no emission schedule to inflate. There is no new capital required to service old promises. Structurally, it cannot be a Ponzi, because it promises no yield to anyone.
What XRP does have is a supply structure. Roughly 40%+ sits with Ripple, much of it in escrow releasing one billion monthly, with unused portions rolled back. That is a long-standing structural overhang, not a new risk. Evernorth's 473 million tokens do not change the economics. They move chips from the secondary market into a corporate treasury โ the "XRP version of MicroStrategy" narrative. Chips change hands. Nothing is produced.
And this is the point the decentralization debate obscures: even if XRPL were proven fully centralized tomorrow, XRP would still be a settlement asset whose value depends on Ripple's ODL corridors, not on-chain activity. The controversy attacks the "censorship-resistant settlement" story, not the settlement function itself. Those are different things, and conflating them is how retail gets whipsawed.
The argument nobody is having: this is a Howey problem
Now the contrarian read, and the reason I think this fight is genuinely important rather than merely loud.
Strip away the technical labels and the securities framework is sitting right underneath. The Howey test asks four things. Money invested โ yes. Common enterprise โ contested. Expectation of profit โ yes. And the fourth prong, the one that matters: profits derived from the efforts of others.
That fourth prong is exactly where the decentralization debate lands. Ripple has spent years arguing XRPL is decentralized precisely so that XRP is not a security โ because if the network is genuinely independent, then XRP's value does not derive "from the efforts of Ripple." The 2023 ruling leaned on that logic for programmatic sales.
So watch the inversion. Bons accuses XRPL of being effectively controlled by Ripple. If a court ever agreed, that accusation does not merely damage the decentralization story โ it strengthens the argument that XRP's value derives from Ripple's efforts, which strengthens the securities case. The prosecution's technical claim, if true, is a legal grenade aimed at the defense's own position.
Schwartz is not defending a consensus mechanism. He is defending a securities exemption. That is why a June tweet resurfaces in October. That is why the tone is defensive rather than demonstrative. The stakes are not ideological.
Code doesn't grant legal legitimacy. Marketing doesn't either. Only verifiable decentralization does โ and the burden is now on the defense to prove it with validator-distribution data neither side has published.
One more structural fact the debate buries. Ripple wears three hats simultaneously: it publishes the dUNL, it holds the largest XRP treasury, and it is an investor in Evernorth. Curator, holder, and financier in one entity. That triple role โ not any single technical feature โ is the actual economic root of the centralization concern. It is why "just run your own UNL" rings hollow: the operator who does so is choosing to diverge from the entity that also controls the largest supply and the newest institutional vehicle.
The timing is not a coincidence
Let me be explicit about the thing the material implies but never states.
The criticism lands in the window right before Evernorth's Nasdaq listing โ a SPAC merger with Armada Acquisition Corp II, backed by Ripple, Pantera, and Kraken. A fund manager (Cyber Capital) publishes a damaging technical framing days before a major institutional XRP holder goes public.
I have seen this pattern before. During the 2022 Terra collapse, the telling detail was never the peg mechanism alone โ it was who was positioned for the break. I do not know Bons's book. I am not accusing him of a short. But a fund criticizing an asset it may be positioned against, at the exact moment of maximum narrative leverage, is a conflict of interest worth naming. Motive does not make the question wrong. It does mean you verify the claim before you amplify it.
Conversely, the defense's reliance on a four-month-old tweet suggests Ripple chose silence on the October specifics. Silence is not confession. But it is a choice, and in a securities-adjacent fight, silence is a strategy.
The competition that actually threatens XRP
Here is the blind spot the entire debate ignores.
XRP's real competitor is not Bitcoin or Ethereum. It never was. It is stablecoin settlement and SWIFT's upgrade path. USDC and USDT already dominate cross-border settlement volume. Institutional clients want compliance and liquidity โ not decentralization. Nobody wiring nine figures across a border cares whether the settlement rail is permissionless. They care whether it clears, on time, under regulation.
This is why the decentralization charge has limited fundamental bite. XRP's institutional thesis was never "we are the most decentralized." It was "we are the fastest compliant bridge asset." Bons attacked the pitch the network barely sells. And the broader market, judging by how little XRP's price moved on the accusation relative to the noise it generated, seems to have priced that in years ago.
The genuine long-term risk โ stablecoin erosion of the ODL corridor โ is invisible in this fight. Which is exactly the point. The loud argument is a distraction from the quiet one. Every hour spent litigating the PoA label is an hour not spent asking why a bank would route a payment through XRP instead of a tokenized dollar that settles in the same second, on the same chain, with a compliance wrapper it already trusts.
Takeaway
Watch three things, not the tweets.
First, the diff. If Bons produces a commit hash or a timestamped repository showing code activated before disclosure, this becomes a governance crisis. If he does not, it becomes a credibility footnote. One fact resolves the entire technical dispute.
Second, Evernorth's first day of trading. That is the real market referendum on the "XRP treasury company" thesis โ not the flame war. A premium that holds validates the structure; a break below issuance value signals the whole narrative is borrowed, not earned.
Third, validator distribution data. Until someone publishes how many XRPL validators actually run a non-Ripple UNL, "nominal openness versus factual centralization" remains unfalsifiable. Whoever publishes that dataset first wins the argument that matters.
The most revealing question is not whether XRPL is decentralized. It is why a network this mature, this well-funded, and this old still cannot answer that question with a number. Code doesn't care which side is louder. It only cares which side can produce the evidence.