The new legislation changes the rules of the game for Autonomi

Autonomi Network Token (incl (e)MAID)
1

Yesterday, American lawmakers introduced a bipartisan bill, the Promoting Innovation In Blockchain Development Act of 2026, protecting developers from criminal liability under Section 1960, and in July 2025, the Geniuss Act was introduced in the US, which fundamentally changes the issues surrounding distributed projects by defining such technologies as DePIN and giving them a completely new status in the legal context.

What does this mean for Autonomi in the context of the provisions of White Paper v. 2.5 and in correlation with the definition of DePIN and the new US law:

  1. The White Paper emphasises that transactions take place directly between the user and the network (as a whole), rather than a specific entity.

Autonomi uses XOR addressing to distribute data and rewards, which means that no node knows for whom it is storing data, which is crucial in terms of protection against criminal liability. Since the Autonomi protocol does not (and cannot) allow for censorship or identification of parties, under the new law, developers are not liable for how the network is used. This enables the launch of a native token that is 100% anonymous and fungible.

  1. Autonomi assumes that the price for data storage is not determined by humans, but by an algorithm that responds to supply and demand.

The network uses a mechanism where the cost of data storage increases or decreases depending on the availability of free space on the network, which means that, in light of the Blockchain Development Act of 2026, such automation protects developers. Since the “price” is the result of autonomous code rather than a management decision (which does not exist), no one can be accused of market manipulation or running an illegal exchange. The network becomes a self-regulating market.

  1. In the new law, it is crucial to distinguish between cryptocurrency mining associated with securities and the provision of infrastructure.

Autonomi defines providers as nodes that do not “mine” blocks but offer measurable resources: Storage, CPU and Bandwidth, and this directly fits the definition of DePIN contained in the Genius Act. The reward mechanism in Autonomi is a direct function of “proof of stake”, thanks to which the native token becomes a digital bill for IT services rather than a share in the network’s profits, allowing for full token autonomy without the risk of classification as a security.

  1. Unlike blockchains, Autonomi assumes that you pay once for data storage forever.

The White Paper states that the token is “burned” or transferred to the reward pool at the time of data storage, which enhances the usability of the network, and the DePIN status legitimises this model. Under US law, burning a token for a service is treated as energy consumption rather than a financial transaction. This allows Autonomi to create its own closed economic system that is legally immune to attacks targeting traditional payment systems.

If these legal acts pass full legislation, they will remove the obstacles that have blocked the establishment of networks such as Autonomi, and the project now has a clear path to classify its token as an infrastructure asset, which is necessary to achieve full liquidity in the markets without fear of criminal sanctions for the foundation or developers.

10 Likes
2

The code is the truth but the white paper and payments are made by a user directly into wallets owned by people running nodes.

As for legal jeopardy, I’ll believe anonymous tokens carry no risk for programmers and participants when I can see, hear and touch it myself.

3 Likes
3

That’s right, but that’s what this law is about, that payment with a native token will not be treated as a payment with a cryptocurrency, but as a settlement for sharing your network resources - at least that’s what the provisions of the law say.

And remember that David had already predicted this: " Fae will run Autonomi nodes, earn ANT, and use ANT for storage — without users needing to interact with any of that directly. The intent is to make the token economy invisible infrastructure." - which eliminates your direct involvement in payments. :wink:

1 Like
4

I think you’re reading something into this that isn’t possible. We don’t have a native token, we have one deliberately chosen so that it exists outside the network - for on and off ramps.

So David literally can’t hermetically seal it inside the network. He can make it invisible to users who take no interest in what’s going on in practice, but those are two different things.

Either way, my point remains that an anonymous token is not going to be ok because of this particular legislation. Meanwhile David has repeatedly vilified people advocating for a native token, and there’s no plan for one. So we are stuck with Arbitrum ANT for the foreseeable future.

I don’t agree with him that this is fine. For example, we’ve seen how ETH fees dominate and break the current pricing model which is essential to matching supply and demand.

For now it’s not clear how Autonomi expect to deliver on the White paper, which promises they still aim to realise and which not, or how they intend to address some fundamental problems.

And anyone putting an LLM in charge of anything valuable is taking a huge risk. I will watch from the sidelines if this ever gets implemented which I doubt.

2 Likes
5

I agree with you, I was going to add an emoticon there :wink: - I’ll correct it right away.

Of course, the token will not be suitable due to US legislation, but it is clearly stated that these changes are intended to attract talent and give the US an advantage, so it must be assumed that Europe will have to wake up and move its legislation in the same direction.
I also published this so that the team could comment on the changes in legislation.

Personally, I am opposed to the automation of life in every sense, and handing over data, resources or payments to an AI assistant is unacceptable to me, but as I understand it, the team is proposing this Fae feature as an option.

2 Likes
6

Although that may or may not be true for Americans, the vast majority of the team are Europeans who have different laws/rules. Shu and I are the only Americans.

I understand many in the community are very passionate about a native token but right now the team has other objectives we are working on. I know its not what you want to hear but even if possible its not a priority or even on the drawing board right now. Could it potentially be developed in the future? Yes. Could it be possible that native token is never developed? Yes.

2 Likes
7

Many in the community are very passionate about anonymity, decentralisation, removing unnecessary dependencies, and reducing the number of middlemen (middlepeople?) involved.

The problem with the current token isn’t that it isn’t “native”, it’s that it gives up on various of these, in different ways.

That said, I do appreciate you stating clearly that the team has no specific interest in making such a thing.

90% of the problem has been the lack of clarity, denials, hand-waving, gaslighting, accusing forum members of being purists and FUD-merchants, etc, instead of just stating very upfront and clear: this project can no longer focus on privacy and anonymity, at all, and will not be doing so. Except, as an afterthought, maybe, here and there.

Would have saved a lot of bother, on both sides.

4 Likes
8

That’s only true on the monetary side. The core principles still apply to data & general network usage.

As long as you can’t link a specific blockchain transaction that pays nodes with a specific file, privacy & anonymity is preserved. If you can, then it’s more tricky.

I think some frustration comes from the fact David is currently interested in anonymity privacy security and freedom for data, not money, whereas many early followers of the project came from Bitcoin and other crypto, wanting the same values not only the data but for money.

It seemed Both were initially project goals, now it’s only data for the foreseeable future.

Despite not loving it, I think it does make sense to prioritise data for now. Getting that working well is a huge challenge in itself without the complications of money. Once the data network is in a stable place, money will be something that can be built on top in a way that brings Autonomi’s privacy, security, and freedom to money, either by the team, or third parties utilising stable data types.

5 Likes
10

We, well I, honestly don’t know what core principles remain, if or how these have changed and tbh I don’t think Autonomi do any more.

To what are they really still committed, short medium and long term?

They’ve just ignored requests to clarify this for a long time now because they want minimal engagement with the community. They prefer to just carry on largely without comment and… well I’m not sure how to characterise what they do now.

Periodically they announce some often vague (for me) goals and appear to be working towards some of them with little handle on what they mean, and then it all changes the next time they do this.

It wasn’t always like this. For a long time I believe we knew what was aimed for and had a good grasp of how it was going to look in the end. At least in terms of core features.

2 Likes
11

If that’s the case (that someone can link an upload payment blockchain transaction with the specific uploaded file), privacy is reduced vs the original vision. Still, the original vision can’t be delivered without a well functioning data network & the team have very limited resources, so getting that in place as a steady base makes sense, and then privacy can be further boosted with a native token in due course.

If there were a chance even the story of the privacy token could cause issues for the team in the coming months/years, they might as well say it’s not on the cards, even if they intended to work on it in the future.

There have been chats about community efforts to work on possible native token solutions. While I know it’s not easy, there seem to be far more people willing to complain the team isn’t doing it than people willing to get stuck in & try to make it happen.

1 Like
13

Privacy and security still remains very important to the team. There is no way to link a EVM address to the data they uploaded. Yes you can see a certain EVM address paid X amount of ETH and ANT to upload but that’s it. Currently if you want to anonymize your EVM address the best approach is to use a mixer (if legal in your jurisdiction) for the time being.

2 Likes
14

In theory these things “cannot” be linked.

In practice we can’t guarantee that, and this argument reminds me of how for years people were told by governments and corporations that their data was anonymised as if that preserved their privacy.

Hopefully most people now realise that this was untrue, and that if you have enough data from different sources, it’s very easy to re-identify so called anonymised data.

Statistical techniques being brought in to identify Blockchain transactions is another example.

So I personally don’t accept the claims Rusty makes, and suggesting people use mixers is to me a remarkable admission that privacy is not guaranteed in the current implementation. I certainly wouldn’t trust this if I were in need of such anonymity.

5 Likes
16

Perhaps you did not notice that I wrote in response to @happybeing:

Don’t you think it’s very risky to publicly declare such an approach? Guess why…

17

Don’t worry you’re getting a third party LLM to handle the third party blockchain via a third party mixer using a third party wallet on a third party exchange . It will be super! Unlike a native token which would avoid all of this - how boring.

BTW the LLM Fae uses is Alibaba’s qwen models, whose training code and data are not documented. They aren’t open source.

4 Likes
18

Explain how you would organise on/off ramps for the native token?

2 Likes
19

We could always have had limited on/off ramps without exchanges:

  • run nodes → earn tokens
  • store data → spend tokens
  • exchange tokens for …
    etc

That’s pretty much how bitcoin began and was ignored because.. well, we’re told it was so that investors could cash in. That went well.

Now that the value of ANT/eMaid/MAID is negligible for most existing investors, the current off ramps are useless, so we’d be as well off, better off, with a native token. On ramps are equally useless because…

What we need is a functioning network and a network sustaining economy. On and off ramps can and will come later.

If we had a network that worked, I think this community could make it fly. Look at what we had with v1.

Edit: this, despite the community being depleted by the move to Blockchain and Autonomi deciding to move from cooperation with us, to excluding us, and taking the community for granted.

For example, remember Darius? He was very low profile, but a big supporter based on the original values. He ran many tens of thousands of nodes on the 2024 testnets - on servers he personally managed - and dedicated three large servers to support my app development when (despite David appealing to Jim Collinson) Autonomi would not help me with tokens for testing themselves. I only knew Darius because he supported my work directly, but he left because of the move to Blockchain.

9 Likes
20

Native token would have gone down with emissions, let alone the subset of autonomi 1.0.

That is if mutables had even been stable enough in the first place to support a currency.

The truth is, the network isn’t the place to store long term monetary value. Native token wasn’t even an option yet.

Blockchain is and was reliable. If the network, one day, is stable enough to hold a currency, then maybe we will get native.

Here, now? Nah, not even ready yet.

2 Likes
21

This is true, but the third party LLM is there to “help” you if and only if you have a modern machine made by Apple Inc. with 32GB RAM or above. That’s if I’m understanding the Discord correctly (which isn’t certain, the answers can be quite evasive, but I believe that’s what is being said).

The plan AFAIU is to later develop the “agentic” chatbot for first Windows and then Linux. Attempts to get the size down have already been made though and have not worked out thus far, so looks like that 32GB figure could stay, for some time anyway.

It is my understanding that accessing the network will still be possible without running this humongous, resource-intensive, black-box, proprietary, non-deterministic, “female” “assistant”. But, simultaneously, elsewhere it seems to be implied that humans will be doing very little or no config tweaking.

Presumably, everyone who can’t run Fae will in fact be tweaking configs? And, presumably, that’ll be a very large percentage of users?

I don’t know, we’ll see I suppose.

EDIT: Thrilled to learn I was misunderstanding here. @Southside linked to this GitHub repo with more info.

2 Likes
22

I suppose it’s too optimistic to expect:

  • CEXs would be willing to work with DAG token
  • DEX would even be possible for DAG token
  • exchanging tokens on the forum as in the good old time of BTC @ MtGox
  • avg Joe would think it’s a great idea to run nodes 24/7 for a month upfront to upload x GB
  • organizations would even think of running nodes for tokens

I literally can’t understand the workflow of how native token would survive if it is born tomorrow replacing erc-20.

1 Like
23

some more answers/hints can be found at Developers — Fae | Saorsa Labs

LLM Backends
Fae always runs through the internal agent loop (tool calling + sandboxing). The backend setting selects the LLM brain:

Backend Config Notes
local backend = “local” On-device via mistral.rs (Metal on Mac, CUDA on Linux)
agent backend = “agent” Auto-select (local when no creds, API otherwise)
Local model selection is automatic based on system RAM: Qwen3-VL-8B-Instruct for 24GB+ systems, Qwen3-VL-4B-Instruct for lighter hardware. Both support vision and are loaded with ISQ Q4K quantisation.

heres how it all goes together

image
image1036Ă—1060 61.7 KB

image
image914Ă—803 41.3 KB

and heres what David said on Discord
image
image2563Ă—260 45.7 KB

Just hold your nose and just read it for yourself on Discord. :slight_smile:

Contributions are welcome - get tore in :slight_smile:

image
image880Ă—221 14.4 KB

edit: from fae/docs/benchmarks/fae-priority-eval-2026-03-07.md at main · saorsa-labs/fae · GitHub

image
image1447Ă—1038 133 KB

2 Likes