qBitTensor Labs Live — July 2, 2026

The July 2nd broadcast landed one month after Enigma went live and just shy of qBitTensor Labs' first birthday. ShorShot sat this one out on well-earned shore leave, leaving Bob and Omar to cover a packed agenda: a month of live competition on Enigma, the RSA challenge racing from 360 to nearly 500 bits, a new compute-credit path to web submissions, PennyLane support on SN48, a Quantum.Tech World recap out of Boston, and a wave of Bittensor emission changes reshaping the whole network.

Enigma: One Month In

The launch held up. No fundamental flaws, no showstoppers, nothing that needed a rethink. The surprises were smaller: far fewer validators participated than expected -- the team runs its own validator on an RTX 6000, with Rizzo running another, while many validators simply delegate back to the team to run. A single vulnerability surfaced in one challenge's scratch space, but the community reported it and it was patched before anyone exploited it. Notably, no one submitted the exploit to try to win -- they flagged it instead.

Breaking RSA -- 360 to 480 Bits in Weeks

• Milestone 1 opened at 360-bit RSA -- hard enough not to fall instantly, but achievable with classical methods. Solved, as expected.
• Milestone 2 jumped a full 100 bits to 460, which surprised even the team. It was cracked in 3.9 hours, just under the 4-hour wall time.
• Milestone 3 steps up only 20 bits (480) -- but because RSA is exponentially hard, that "small" jump is a big one. Mid-stream, Omar broke the news that a 480-bit submission had just entered review.
• Submissions cost 0.1 TAO each; roughly 137 came in total. The constrained environment (a single node, four hours) is deliberate: solve it small here, then hand the methods to researchers with supercomputers and months of time.

Inside the RSA-460 Solution

Omar walked through the winning 460-bit solution: a GPU lattice siever built on the General Number Field Sieve -- the same core method used to factor the largest RSA key ever broken (829-bit). The standout was memory discipline: an anonymous memfd kept temporary storage to roughly 100–200 MB, well inside the challenge's limits, answering earlier complaints that the environment didn't allocate enough scratch space. Run on supercomputers for months, the same approach could push past today's records.

Hardening Quantum Proof -- Standing on Shoulders

Peaked circuits are hard because they pack information into matrices too large to simulate on a classical computer. Milestone 1 fell to an MPS (matrix product state) simulator paired with canonical beam search -- an AI search technique -- with results validated across bond dimensions and by scrambling the qubit order to rule out lucky guesses. Milestone 2 built directly on it, adding an "unswap" technique that rearranges qubits to shed bond dimension without destroying the peak. The milestone-2 code even carried comments crediting the milestone-1 solution. That is exactly the virtuous cycle Enigma was designed to create: miners not just competing, but building on each other's work.

Web Submissions & Compute Credits

Today, participating in Enigma's challenges means registering a Bittensor miner and paying in crypto. The team has the infrastructure for web submissions but hit a wall: Stripe -- and the mainstream processors that mirror its terms -- won't touch contests that charge an entry fee for cash prizes. The fix reframes the payment as compute, not entry: users will link an Open Quantum account, purchase compute credits, and spend them to submit through the web portal. It keeps the experience clean and compliant, and it creates real subnet-on-subnet synergy -- Enigma participants become Open Quantum users, ahead of the roadmap where Quantum Compute (SN48) supplies the quantum resources Enigma's future challenges will run against.

Future challenges already in development: quantum RSA (real QPUs, necessarily small numbers in the NISQ era), quantum-simulation RSA (optimizing Shor's algorithm in a simulated environment), and ECC / post-quantum breaking to pressure-test crypto agility as the world migrates off RSA.

SN48: PennyLane, Docs, and a New Machine

Quantum Compute added full PennyLane support (Xanadu's framework) alongside existing Qiskit support -- often a single-line change, with no plugin import required if the package is installed -- riding PennyLane's momentum as it erodes Qiskit's once-dominant market share. The team shipped a complete documentation revamp, teased a new quantum computer coming exclusively through an Open Quantum partnership, and previewed a second partnership set to put Open Quantum in front of more than 20,000 quantum users.

Quantum.Tech World -- Boston

A busy show, and a well-attended one. Peter Shor wore an Open Quantum / Quantum Rings lanyard on stage. There was heavy federal presence (DOD, NSA, Department of Commerce), and copycats are starting to appear -- which, as Bob framed it, validates the category more than it threatens it. Two executive orders dropped: one pulling the post-quantum migration deadline forward to before the end of 2031, another funding quantum infrastructure and directing the government to procure a research-grade system by 2028. QuEra announced a five-year roadmap claiming fault-tolerant quantum computing by 2028 -- 256 logical qubits on AWS Braket at a 10⁻⁶ error rate.

Bittensor Shifts: Emissions, Root Reborn, Conviction

The pendulum swung back toward pro-innovation policy, fast:

• Root Reborn -- routing idle root APY into subnets, ETF-style. Bob likes the concept (put your stake where your conviction is) but not the clunky implementation.
• Price-based emissions replacing the unsustainable flow-based model, with mechanics that support young subnets while guarding against pump-and-dump.
• Manual emission shutdown -- Const ending "free launch" for dormant or exploitative subnets, prompting a scramble among owners who had been quietly collecting emission.
• Conviction and takeovers -- activated overnight. Bob's take: in an ecosystem that can rewrite how all capital flows in a single pull request, locking capital for a year mostly makes sense for operators who need to buy trust rather than earn it.

Community & What's Next

Sentiment ran strongly positive -- miners calling Enigma a great, straightforward experience, and others (rightly) calling the challenges hard. Terra Quantum's Taylor Hartley emailed to say she'd be talking up the RSA challenge at a major cybersecurity conference and invited the team along. On prizes: the earlier "$1M in prize pools by end of summer" projection is being walked back given TAO price swings, with a transparent update to follow. Next stop is the Exploit conference, where the team hopes to host a quantum fireside chat on what's real and what's hype.

Anyway, let's dive right into it. So as always, you know, the usual caveats. These are ideas, not promises. this is not in any way legal or investment advice. you guys have been doing a great job of using this information for good and not for evil. and so as long as you can keep doing that, we'll keep putting these on. And if you're still here, we assume you agree.

jump right into it. All right, today's agenda is going to bring us through the usual covering of the subnets. We'll talk about Enigma first, we'll talk about quantum second. then there's been man, there's been like just a lot going on both in the quantum ecosystem and in the Bittensor ecosystem. So we split that up into two sections and then we'll cover the usual community stuff at the end. Okay, so without further ado, let's jump into it. with further ado

You'll notice Shoreshot is not with us. He's getting some much deserved shore leave. that this is a beautiful AI slop with all of the quintessential American activities and celebration of the Fourth of July. But we hope we hope ShoreShot is getting a little R and R after after working his butt off on the release of Enigma and, you know, the continued you know, continued operation of the other subnets. In fact

I've noticed he's still, you know, periodically appearing on chat, even though he by all means should be off and disconnected. hopefully, ShorShot, you're not watching this and instead you're watching the waves crash or something like that. but we'll continue on. we look forward to having you with us next time. All right. So jumping into Enigma, I guess, you know, the last time we had a QA bit Tensor Labs live, we were freshly launching and

Showstoppers, no need to kind of like fundamentally rethink anything. any minor issues? Yeah, a few. So one, we've been really surprised, and this is maybe interesting in the greater Bittensor context, but we have been surprised how few validators actually participate. so we're running with our own validator on an RTX 6000 and Rizzo is running a validator. but a lot of the validators are just kind of

Delegated to us to run, which in a way is kind of okay. but you know, we would certainly love to have more compute horsepower. So we're really surprised by that. Curious if it's representing like a bigger trend in Bittensor but you know, we'll sort of keep an eye on it. you know, we did run into some small things where we had, and we'll talk about these in a little bit more detail in the kind of takeaways, but you know, we've

played with some scratch space available to the solutions that get submitted. And we did have one small vulnerability that was reported that was in one of the challenges. And we patched that before before it was exploited. So you know, a few kind of like minor rounding the edges, sanding things. But overall, this launch went off really far better than I think I would have even expected in my in my best dreams.

So let's talk a little bit about the progress on the challenges. And then Omar, maybe I'll have you prepped to dive deep onto the solutions that have been submitted so we can talk through some some of how these solutions worked. But we started off with milestone one for both of the challenges. And the first milestone was a 360-bit RSA, which I'm gonna admit, before I started trying to do this myself, I actually thought that that was well beyond what was probably possible.

Peaked circuits has a lot of different knobs. but we sort of had the same thing with peaked circuits. The first milestone was very much known to be solvable. We had solutions that solved it ahead of time, but good to whet the appetite of the community. And so you know, no surprise, both of those were solved. Second milestone, I think the community was really blown away that we jumped a hundred bits.

I was a little bit blown away too. So we had our own solution and so we sort of knew how high we could push ours. But after the first milestone was solved, we actually ran the solution that had won and we pressure tested it ourselves and saw exactly how far that solution could scale. And we found that it essentially could get up to about 440 bit. So we put a little bit of headroom in and and put out the next milestone.

D2, on the other hand, we knew ahead of time exactly what that was gonna be. these both, I would say, were hard to solve. but they were both solved. and so very cool on that. It took a lot longer. RSA, we were a little bit we were a little bit like concerned that it might take a really long time to solve it, but you know, we knew it was possible ultimately, and so that that also was solved.

one of the biggest hiccups on that was sort of like you know, we're hosting RTX six thousands. The initial solutions for RSA really weren't using GPU. And so there was like a heavy dependence on what the CPU architecture was, which, you know, can be variable. And so what we actually found was that some solutions started passing that on like an AMD architecture, but weren't solving it on an Intel architecture until they really kind of

One of the coolest things about RSA is that it's essentially like exponentially hard. That's why we use it to protect all of our communications. and so we think people are gonna solve milestone three? probably. But maybe not. like could you do that if you had ten times as much compute and ten times as much time? For sure. Like easy, right? But but in a constrained environment where you have a single node in four hours.

You know, that actually is it's gonna be impressive when somebody breaks that. milestone three for Pikachu Circuits, you know, similar thing. We do expect that people are gonna be able to break that, but it's gonna take some innovation. and so we'll sort of we'll sort of see where things go. and so yeah, will milestone three be solved? Probably, but it won't be trivial. you know, it's gonna be hard work. says that the hilarious typo. We haven't done it yet.

personally, right? We've not built a solution ourselves that does it. But we believe that it is possible for both RSA and for peak circuits. will milestone four be solved? And that's where I think we're actually going to be unsure about that. So when we see how the RSA solution is solved, w we will pick the next one. But I think we're very quickly going to be getting into the point where those really can't be solved.

and that's sort of the purpose of the classical RSA challenge, and it creates the opportunity for the follow-up challenges that are more quantum centric. It's establishing the benchmark. and for peach circuits, the story is very similar to that. In fact, maybe not a bad time to address the fact that, you know, based on like the way that we've talked about this in the past, I think people are sort of like expecting that because the challenges get exponentially harder.

The more you can sort of increase the prizes over time. And so, well, there might be some exceptions to that. Like, say, you know, if we put out a milestone for breaking RSA 2048, which we know you can't do. Like, I mean, that's known to not be possible. Maybe we could stick a giant prize on that. but for the incremental milestones, we're basically going to be starting at a baseline and slowly turning up the dial on those. And so, yeah, maybe actually,

Now would be a good time to talk a little bit about RSA 460. So Omar, do you wanna talk to that solution?

but it was quite an elegant solution, and so we wanted to you know highlight that. And so the primary mechanism was a GPU lattice siever. so it makes use of the GNFS, so that's general number field siever, which is a common algorithm for factoring you know factoring semi-primes. And so this is you know not been seen or this has been seen before. It was used in factoring the

largest ever factored semiprime. but it was fantastic to see this applied to you know our challenge. I think the really impressive part was how they managed the memory. There were a lot of there was a lot of feedback that, you know, hey, we didn't give you or we didn't allocate enough temporary memory in order to make this challenge solvable. And it turns out there is a way to do it, yeah, using the you know anonymous

memfd which kept the temporary storage as low as you know a hundred to two hundred megabytes, which right now I believe you get access to a gigabyte, if not ten. And so, you know, making use of very little space to you know solve this quite difficult problem. And so, you know, as I mentioned, we use the solution uses the same core method for breaking the largest known RSA bit key.

And I think that's pretty exciting because if we are on the cutting edge of what is achievable, you know, who knows how long it is before innovations that have never been made are made before. And so even though it's in this sort of constrained environment where, you know, we give you four hours on pretty modest hardware relative to the months that the world record solution had on you know parallelized across supercomputers, we can still apply those same methods here and

I think we've had just about like just about a hundred overall. In fact, 137 total submissions, all paying point one TAO to submit. But you know, but if you think about that idea that you know, these solutions very often use like massively parallelized systems, but we're only charging people point one TAO and we're running it on a validator, you know, there's there would be no way that you could run like all of this solution. So by scaling it down.

saying, hey, you have these constraints to operate in. How do you innovate? Then the presumption is that we'll be able to scale that you know to a much cooler, you know, we'll be able to contribute the solutions that we have here into the research community who's willing to put, you know, giant supercomputers on these things going forward. So really cool, really cool stuff, man. next then I guess we have the hardening quantum proof. take us through that.

peak circuits are large largely considered to be intractable is that you just don't have enough memory to store all of these operations. And so the conventional solver for peak circuits tries to eliminate as much of this, you know, noise that's implemented when generating the peak circuits. And so for a difficulty one peak circuit, that is fairly trivial. And what you can do is you can narrow down the matrix and only be concerned with the important information, which then you can use to

reveal the peak. And so what we saw in Milestone One was a MPS simulator. So that's a matrix product state, which is a fairly common solution. And you know, as it evolved the quantum state, it was removing this background noise, so shrinking the operations. And then it also employed a technique called canonical beam search, which is a search algorithm conventionally used in AI. And so it was cool to see this cross over into the realm of quantum. And as they

progressed the solution, it eventually converged onto a peak, which is what you want when finding a peak circuit. And the way they tested this was testing different bond dimensions. So that's sort of the tolerance for you know how big your matrices can be. And if it was consistent across these different bond dimensions, you knew you very likely have the solution here. And then it also cross-checked it by running the experiment by scrambling the qubits, so changing up the order of the circuit to make sure there was no hard coded or

You lucky guess that was happening. And so it was a fairly thorough solution, which we were pretty impressed with. The problem with it is when you get to more difficult circuits where the noise is not as obvious to eliminate, if you are, you know, willy-nilly removing information to you know shrink your matrices, then suddenly you're losing actually important information and your peak disappears. And so what we saw was a solution building on this, which added

milestone one. And so we've adapted that to our own purposes. And so it's super exciting to see that well miners are competing. They are also collaborating to sort of push the limits of what's possible. And that's ultimately what we think is gonna lead to, you know, these major breakthroughs.

really I mean like almost gives me goosebumps, Omar, that use case that you just described because it's like that is the vision, right? The vision isn't just that you have smart people working in silos, but that you have smart people taking the learnings from the previous people and innovating on it. And hopefully you'll see like the milestone one guys taking the milestone two guys and using it to solve milestone three, you know? Yeah. Super cool stuff. All right. So then the

They demonstrated that like these miners aren't just a bunch of, you know, degens. they are actually like really intelligent people who can go and find intelligent work and apply it really well. And so the real question is gonna be like, does that turn into new intelligence? And that's, you know, really what we're on the cusp of right now with the difficulty of the challenges. The other thing is that

You know, we've learned that how we calibrate the compute environment makes a huge difference. and so, you know, I mean, we kind of like knew that at the beginning, and that's why we standardized around RTX 6000 as like the baseline compute, but you know, the amount of temp FS space, the specific CPU type of the host system, like all of these types of things have played a bigger role.

There will always be workarounds to exploit. Miners have have been great about reporting them. You know, I'm surprised, honestly. Like I like I love it. But like the first exploit that was found

They probably okay, so the Treasury wallets work such that there's a very large like series of steps and you know the administrator needs to like recommend a transaction and then the validators have to do it and there's a time period. So like there is a chance that we would have you know, there's probably a pretty strong chance that we would have caught it and stopped the transaction if somebody had, you know, deployed the exploit. But I was really surprised that the community's response to this was to inform us about the exploit.

and so, you know, we probably will have to, you know, continue on that learning and either figure out how we get more validators to bring compute so that we can continue to like, you know, reduce wait times for people who are submitting solutions, or find ways to reduce the burden on the validators, maybe through other partnerships in the ecosystem, et cetera. Cause one thing that Tensor is also very good at is decentralized compute. and then the other big thing is yeah, just in general.

Tons of excitement around quantum, breaking RSA using quantum. I mean, that is a hot topic in the industry right now. And so really cool stuff. and so what's next? big thing that is coming up. So I don't know if we I guess we have talked about this in the past, but I don't know if we've talked about like how the current state is, you know, under a microscope. So one of the big, really important things is if you want to get real innovation, we need the combination of hackers.

and researchers. And today we launched with just the ability to participate in this through the Bittensor ecosystem. So somebody who submits a solution actually has to register a miner. They have to pay the fees to play using cryptocurrency and the rewards are issued in that ecosystem. But we have almost all of the infrastructure to to allow for web submissions, which means that you don't need to know anything about Bittensor, you don't need to know anything about keys. You could just

register a user account and run your solution. So why didn't we launch with that? Well, the problem is turns out Stripe does not like crypto and they don't like pay to play contests with monetary rewards. and so we actually do lots of business with Stripe in our traditional business and they've always been like really easy to do business with. But if you have a use case that isn't I mean that something they want to avoid, they just shut you down.

We want it to be like a clean, crisp, professional e-commerce experience to allow people to play. And so the next move that we are going to be trying to implement here is to do a credit-based submission where the user actually goes and links their Open Quantum account, purchases compute credits on Open Quantum, and then they can use the compute credits to enter. The idea here is that you'd link your qBitTensor Labs account.

To your Open Quantum account, and you would submit at that point through the web portal. the key distinction here is that the fee is for compute, not for entry. Right. And so that should prevent us from breaching the terms of Stripe. It should also have some other good side effects, which I'll talk about in a second. but but but the key thing is you know, really enabling other people to play. So

Omar and I were in Boston at the Quantum Tech Conference. And one of the many people we talked to while we were at the conference last week was the BlueQubit guy's on the right as Hayk their CTO. He wa he spent a little bit of time with us in the booth, actually talking to quantum users about Enigma. And okay, I'm gonna paraphrase this as an and it isn't an exact quote, but he basically shared that he was at IBM talking specifically about Enigma and the peak circuit challenge.

And a number of the top minds in quantum were eager to jump in. In fact, he said that they're kind of like almo like almost like wishing he would have told them sooner because they were like, Man, I would have just taken that that milestone one and milestone two money. And so people are eager to come in and play. We just need to give them the ability to do it. And so we'll get that going as quickly as possible so that again, we can have that combination of like really good quantum minds and really good hackers.

Compute subnet that provides the access to the quantum computers and the quantum simulation for Enigma, where you know the validators want to run those challenges against quantum environments. But even right now, you know, the user will go to qubit tensor labs to engage with Enigma if they are not a Bittensor user, right? They're going to the website to do all of their engagement. And in this new process, it will direct the user to go to openquantum.com to register.

Put in your credit card number so that you can buy compute credits to use, link the two accounts so that they can be used in the Enigma ecosystem. But now all of the sudden, not only are you using those credits to pay Enigma, but you also are an Open Quantum user. And the type of user that will be doing this is actually the quantum half of that ecosystem. And so it essentially gets you into our direct mailing campaigns to drive.

you know, you'd also spend those credits that you're putting into quantum compute. And so just a really cool example of like you know subnet on subnet synergy coming out of this. And looking forward, we've got, you know, a lot of stuff that will further drive those synergies. So the future challenges, which are already kind of in development at this point, is a quantum RSA challenge. And so this is breaking RSA using real quantum computers, which

Just to set the expectation right, that will be way smaller numbers in the NISQ era of quantum. but as we start to get beyond the NISQ era, you know, we'll start to see that accelerate and hopefully on a different kind of trajectory that's not plateauing the same way that that that it plateaus when you try to solve RSA classically.

To post-quantum cryptographic standards. One of the interesting things you'll hear about in the quantum industry is nobody's talking about like stop using RSA and start using this one form of PQC. They're talking about crypto agility because PQC is not exactly a guaranteed thing. Like it ha it really hasn't had enough time of people trying to break it. And so we'd like to bring that.

incentivize people to break it so the world knows what works and what doesn't work. Yeah. And so yeah, just I think more and more opportunity over time for the interplay where the quantum compute subnet becomes the compute resource for Enigma.

It would have been submitted hours ago, but like wouldn't it be funny if somebody was trying to break it right at this moment? All right. All right. So hey, moving on to 48 quantum compute. one of the things we said when we first launched was we were gonna we were gonna add Qiskit support and we were really quickly gonna follow up with PennyLane support. We've added PennyLane support and the community has been lighting up about it. So for context, Qiskit is IBM's

Qua quantum framework. So it's basically like the Python library that you use to write your quantum algorithms. There was a time when they had like an 85% market share, like pretty incredible. at Quantum Tech, I saw like, you know, an executive from IBM presenting, and I think they said like they have a 63% market share. And he was kind of like bragging on that, which sounds really good. Like a 63% market share is really good, unless it's on like a massive downward you know, trajectory.

and so yeah, I've been really surprised to see how many people have been switching over to PennyLane from from Qiskit. And so we've added full penny lane support. you can now just literally change one line. I think this line right here is the only line that you actually have to change in your penny lane code after you pip install. you know, no, no, this line too, and this line and this line. Yeah, you have to add your credentials. Those the bottom two.

If you have an environment variable set, you don't have to change anything. just the back end you want to run on and that it's an Open Quantum device. And then PennyLane just works. No other changes required. we're super excited about it just because of like the mass adoption that PennyLane has been getting. And, you know, I will say at at the quantum tech conference, that was a that was sort of like a big thing, like a big question on people's lips was do you guys support

Absolutely, you know, use it for other purposes beyond that.

private white glove customers, right? But I don't think you can run on a Xanadu computer yet. but they're starting to eat IBM's breakfast when it comes to when it comes to that front end of getting the developers. So that's really quite impressive.

But it's a colossal pain. And so we're really excited to have this Open Quantum plugin implemented because it is literally changing one or you know, maybe a couple more lines of code to switch between them. But you don't have to manage all these different plugins and packages. It's i if you looked at the code snippet on the previous slide, you don't even have to import the Open Quantum penny lane plugin. It just works out of the box if you have it installed. So

you know, the combination of our AI agents and our interns and our staff collaborating quite aggressively on this. And there's a huge amount of documentation now that takes the user through everything from you know, like the web workflows and how those work, through the Python workflows and how those work, you know, all of the all of the details of the Python SDKs and the plugins for all the different frameworks. And so yeah, really, you know, quite an enhanced set of documentation over what we had in the past.

The other cool thing is we have a brand new quantum computer coming soon that is going to be actually available exclusively from Open Quantum through a partnership, which is quite cool. so keep an eye out for that. One of the one of the interesting things about Open Quantum and sort of qubit tensor labs live and how information shares here is we are actually going to be

engaging traditional journalists on a number of the releases. And so one of the things that's really important to journalists is that if they're reporting on something, they are breaking the news, which means you have to basically like embargo the details on it. And you don't tell anybody and you, you know, you get everybody aligned that this information's gonna become available at this time. And so we're gonna start implying a lot of things and then not necessarily sharing the details until the announcements go out.

But keep an eye out for this one. This I think is very cool. and there's another announcement coming out next week, which will be very much focused on an additional partnership that will put Open Quantum in front of over 20,000 quantum users. you know, we we we we really are excited about the potential of what's gonna happen when lots of people who traditionally have had just a crappy time trying to get access to a quantum computer.

Pretty cool event in Boston. you know, unlike Q2B, where, you know, attendance almost felt like it was on the decline. This sort of felt very busy the entire time. I was completely zonked by the end of it. you'll notice up in the okay in the top left, that's me. Up in the top right, though, you'll notice that's Peter Shor. He's the guy behind Shor's algorithm. And if you zoom in on that, you'll notice.

He's wearing Open Quantum and quantum rings literally around his neck. Let's go. That's freaking awesome. in fact, it's really funny because when somebody like me presents at these things, you have like the A V guy who's wiring you up, you know, putting the mic on and wiring it through your clothes. And really, you're just like doing exactly what they tell you what to do. When a guy like Peter Shor goes up, they don't tell him what to do, because they're like intimidated by him. So they didn't tell him to take his thing off, which was awesome. Yo.

some of the key takeaways from the event, there were so okay, like two percent of companies, I think I think this is literally the stat is two percent of companies who have started investments in quantum end up re like reducing or eliminating their quantum investment. I'm surprised it's that low because the reality is that like while quantum compute holds immense promise, it also is just promise. And so, like if you're a company spending real dollars on it.

you know, you've gotta have a long term vision for what you want with the program and you're gonna be investing a long time before it's paying off. But one of the things that I noticed about quantum tech is even companies that previously had programs that have defunded those programs were back in the room, which is really saying something, right? second thing is it's to the point where I'm almost a little bit uncomfortable with the way that hardware companies are pushing aggressive timelines. Like I d like I can be I can

I can put on my hat and make a very bull case for quantum. I can put on a hat and make a very bare case for quantum. But even I get a little bit squishy when I see some of these timelines that are coming out of quantum computing companies. They've gotten access to millions of dollars from programs like DARPA's QBI, and they're starting to get a lot more aggressive on their timelines, which can be a little bit spooky. Also, at this event, so this is in Boston, East Coast, you tend to have little bit more federal government, but

You know, there was a lot of DOD in the room. NSA had like a booth in the room, Department of Commerce, we'll talk about that in a minute, but they're heavily involved with some of the executive orders that have happened. Navy nuclear lab, Lincoln Labs, NSF, like so much federal government there. I would say almost probably a third of the people in the room were kind of federal, federal representatives.

If you are the only one doing something, it is very hard to convince people that you're doing something smart. If nobody else is trying to do it with you, you look like a weirdo, right? Having other people in the market educates the market. It validates the the movement that we're going on and they're gonna have a really hard time catching us because most of these teams really didn't have any quantum chops. And like when you would actually talk to the other quantum people in the room, they'd be like, Did you see that strange guy on LinkedIn? He was

wearing all kinds of bling and you know, what what's going on with this company, right? And so like, you know, the quantum market is sort of like a little bit queasy about the crypto market, but super excited to see other people kind of in the exact same space. And some of them had some I I think some of them had some cool angles. I think there might be some opportunity to collaborate, you know, with them. So we'll be excited to see how that goes and where our leadership continues to create followers.

other cool stuff happening. So two executive orders came out like literally the Monday, Monday of last week. and so like people were trying to sort of demystify what those were. Paul Dabber, the US Deputy Secretary of Commerce, so like one level removed from the president of the United States, was in the room, you know, sort of sharing a little bit about what was in those, you know.

Basically there were two orders. One of the orders was, hey, instead of getting to post quantum cryptography by 2035 get there before 2031 ends. so like a directive to hurry up. And then the other was really a focus on like infrastructure to support quantum superiority for the United States, you know, workforce development, supply chain development, you know, accelerating you know, accelerating industry by

bolster the private industry perspective on this also. And Paul also didn't take off his his lanyard. So you know, got another qBitty out of out of a high profile person here. so all of the stuff that went on LinkedIn, any photos of him, ended up with Open Quantum. So super cool. Omar, Quera is one of the companies saying super aggressive things. And this is really cool because actually like

I had heard about this as early as March of this year, but sort of didn't want to share any details on it. So now they've published their roadmap, talk us through their roadmap.

256 logical qubits available for commercial access. And so they're planning to release their next big device on AWS, so Amazon bracket. Their claim is it's going to have a 10 to the minus six error rate. So for context, the best recorded two qubit error rates are 10 to the minus two, I believe. And so this is going to be a pretty substantial jump in in terms of error. And so when we're considering the growth of

quantum computers, you need to consider one, how many qubits you can run, but then also how those qubits handle as you add more and more circuits. And so you know you can have an infinitely wide circuit with a thousand qubits, but if they decohere and and become noise after you know a millisecond, it's not going to do you much. You can't only perform so many operations. And so what this is saying to us is not only are we going to have wide circuits, so 256 logical qubits, which you can

you know, run actual algorithms on, but they're gonna have enough of an error rate that you can be running meaningful operations on them and getting real results. And so while I don't believe this means cryptographically relevant quantum computing, you know, IB er the Google paper that came out put us at 1400 logical qubits, it is a major step and we may be starting to enter the domain of practical applications for quantum computing. And so Bob, I sort of wish we had our our DARPA

graph up, I think we could, you know, get a get a sense for where we might be pushing the line. But if we're pushing into, you know, quantum chemistry applications, we could start to see real, you know, utility scale discoveries being made, which is super exciting.

And then the system picks it up and runs. They're gonna sell like weeks of time, like week, multiple week blocks of time on these things. Because one of the things about Quera as compared to like a superconducting system is it will be physically slower than a superconducting system. And so a lot of the experiments are gonna need like massive blocks of time in order to run, and they're going to be very, very expensive for those blocks of time.

And they were actually asking us about, you know, exploring partnerships to help kind of like bring bring those opportunities to the market, and to help participate because i essentially you think about these things almost like you're launching a rocket to the moon. Like you will have you will pay gobs and gobs of money, you're gonna have weeks of time on the system, and you gotta get it right. Like you gotta use those weeks correctly because then it's over and if you failed

it just blew up, right? And you get you get nothing out of it other than a little bit of learning. Or you use the time well, you're well prepared ahead of time. You've simulated your solutions to the extent you can, you've de risked them on other quantum computers, and then you run the system and you get the discoveries that you're really looking for with the huge investment. So yeah, we really look forward to this. This is great news from Quera, super cool news from Amazon. We're looking forward to working with them on this.

It'll you know, hopefully everything goes to plan and there aren't huge legal ramifications for violating contract, but it it seems like we're, you know, heading in the right track. so super exciting to see.

accidentally anti innovation type of stuff that was happening. And I said, Don't worry, don't worry. The pendulum always swings back. Like the harder it swings, the harder it swings back. The cool news is that it swung back really fast. So there've been like a lot of big changes that, in my opinion, maybe I would have implemented differently, but but actually are very solid moves in terms of being like pro innovation, as opposed to just like

anti exploit. and so I don't know, let's spend a little bit of time talking about this, but root reborn, I mean

Take this root APY and invest it back into subnets. It's sort of like a subnet ETF. And a lot of people responded with criticisms, but Bob, you've expressed maybe some alternate opinions more in support of this change.

You're putting your money where your mouth is. If you belie you either believe in the ecosystem or you don't believe in the ecosystem. and I do like this idea. Like me as a investor in like standard equities, there are periods in time where I'll just be in ETFs. You know, like I might just want exposure to the S&P five hundred, and I don't have any strong opinions right now about where things are going. But then there are also times where like I've done my research, I'm in it, and

I pick the equities that I want to be more exposed to and I can benefit or miss out because I did that. And so I think if if if root were sort of like if root reborn were sort of structured like this, where I'm in an ETF, if if the if the sum of all the subnets does well, I do well. If the sum of all the subnets doesn't do well, I don't do well. but I can at any time just be like, you know what? I actually want to do better. I want to go directly on 63, directly on 48, directly on shoots, right?

Then I can do that. I do think the idea of like having it land in your wallet as you know as as little, you know, fractional piles of stake on lots of different things feels kind of like clunky and ugly if that's how that's implemented, I suppose. but I think the the premise of the idea I think is a cool idea. It's like b you know, believe or don't believe. And by the way, once you start having exposure to the subnets.

You're probably pretty quickly going to start wanting to pick your favorites and put it on your favorites. So I think I think it'll yeah. devil's in the details. I like it though. price-based emission returns. I was on vacation with my family in California, thinking I wasn't gonna have to think. We just launched Enigma, it's operating beautifully, I was thinking I wouldn't have to think about anything. Then Omar sends me a note.

so I am a big fan of this. there are some interesting implementation details that I don't totally understand, and we haven't finished modeling yet, where it has some things that benefit older subnets, and it has some things that benefit younger subnets. and so, like from from my early modeling, it sort of looks to me like it's it it's built in a way where it tries to

Prevent somebody from showing up and then pumping and dumping, right? So you essentially have like it takes a while for your subnet to start getting emission relative to what its price is for a brand new subnet. If you're an older subnet, it more quickly gets to that. But on the flip side, you know, you want to foster innovation. So you want those younger subnets to have value. And so there's also this like boost that happens for younger subnets where

It doesn't, you know, there's a like bigger EMA on catching up to the price action that it's seeing, but you also have a benefit for just being a young subnet. And so I think like I think I really like what they're trying to accomplish here. I hate complexity and like the fact that you can't just like say it in a single sentence how it works doesn't feel good to me. But I really like what they're trying to accomplish, which is to say young companies need support. You need to give them

funding through emissions so that they can try to build an innovative new subnet. Also, you want to avoid people that just come and make big promises and never deliver anything. So you want to avoid pump and dump, but you want to fund young subnets. On the flip side, the older you get as a subnet, the less you should need to be propped up, and the more your success should depend on how you actually manage the market, how you manage people being excited about your technology, deliver value, get investors based on value. And so

before people start taking sound bites from that, is I am a huge fan of like making a hypothesis, trying something, succeeding or failing, learning and iterating. I think that a lot of times you can get into analysis paralysis where you just sit for years before moving. I think like Bitcoin governance kind of looks like that today. That's a problem. Right. And so the whole time TAO flow had existed, we sort of sat here saying, you know,

We've got smart people in the ecosystem. They're trying an experiment. They're gonna learn something. They're gonna iterate. They did, they iterated on Tauflow two. They iterated back to this price based stuff. But they've learned a lot. And I think what they've come up with is really nice. also probably would be nice to have like more heads up you know, on these things, but you know, beggars can't be choosers, I suppose.

manual emission shutdown. Omar, your thoughts on that?

Yesterday morning, Const did send a message to all the owners and he said, For those of you that did have your emissions shut off, let me know. Do you have code? I, you know, are you distributing rewards? Should I turn it back on? And so there is an active conversation going on with the owners of, you know, if you did have your incentives turned off, you know, is it worth it to, you know, it should do you deserve to have them back? And I think it has prompted a lot of action from

Subnet owners. A a lot of people that were sitting dormant and sort of comfortably collecting their owner omission are now being forced into say, hey, either we have code or you got me. we you know, we were sort of running a scam here. And so it is well, it is trending in the opposite direction of sort of decentralization. I think it is having some upsides, which is interesting to see.

You can try to add rules. Like you can basically say, hey, somebody found some way to exploit and create a useless subnet that's just skimming money off of the network. So we'll implement some obscure rule into some protocol and redeploy and or you know, we'll fundamentally rewrite things and then potentially wreck everybody for a period of time. It's pretty dangerous stuff. you know, and it's really hard to get right. I mean, like as a subnet operators, we kind of experience this, you know, a microcosm of this.

In our own realm because the way that our own mining happens, we have to be thinking about this in a decentralized way. and so I like decentralization. I appreciate that in order to get it perfect, you probably need to create too much complexity. And I think that if you have centralized governance, that's probably okay. what I would love to see is just a little bit more like.

transparency around the governance process so that it's not, you know, just a unilateral thing that somebody, you know, somebody can just have a feeling and wreck somebody's life. Or, you know, a as operators, we invest a lot of our own time and our own money in building these things. And we obviously weren't a part of this because we do real stuff. But

There could have been an accident made and we could have been on the list, you know, and then that could have really hurt. And so, you know, I'd love to see more transparency into how the governance work and what the failsafes are and stuff like that. But, you know, I don't know that I'm a complete hater of this because, you know, it does create a mechanism to get rid of the grift while letting the innovators innovate. Yeah.

Yeah, we have. I Well, I was I was gonna say there's been a a lot of stirring in the community, or maybe not a lot, but there's been some stirring in the community of, you know, hey, you guys are done with the heads down development. When are you guys gonna address conviction? Because it was sort of something we glossed over, but it is worth talking about as you know, fairly mature subnet. And so you had some thoughts on it.

So everybody knows that you're committed to the game. I like that concept. The reality is that you can be on vacation in California with your family and get an IM that everything just changed, you know? this game is rewritten so often. But the one thing that sort of remains universally true over time is that like your ability to your TAO is where power comes from in this ecosystem.

Like your ability to make your subnet survive and make your team survive and you know, look out for your investors depends on your ability to pivot and iterate and deploy your capital in the right ways. if somebody were to come and attack your subnet, you have to use your capital against their capital, to maintain it. Like we have takeovers happening now. And so I think the idea, like unl unless a subnet operator is shady and needs to buy trust.

I don't see why anybody would voluntarily sign up for conviction. Now, that could change. And in fact with takeovers, you know, that might change, right? if somebody say if somebody comes and all of a sudden locks a bunch of capital to try to grab ownership of our subnets, then all of a sudden it's nice that we have the flexibility to, you know, go and beat them and defend our subnet. you know, or you know, or any other

series of moves because the rules will probably be different by the time that happens. But I think it would be, you know, I think it would be quite silly to do it before there's a very good reason to do it. again, if this were a stationary game.

humans and humans as families on your shoulders and the rug might get pulled out from under you because the entire way that you plan to pay for them goes away. and so like the way we operate in conviction today is by building a team, investing in the subnets, building them out. If we actually start locking up capital, you know, we may do that if if a reason comes up for it. but yeah, I think it's a strange thing to have such a dynamic ecosystem that

literally can just completely rewrite how all capital flows in a pull request overnight and to expect the people who are operating in the thing to lock up their capital for, you know, durations of a year or so. It's just yeah, it seems kind of strangely inconsistent to me. Yeah. I don't know. Does that make sense, Omar?

fairly small team that moves rather quickly, having that excess capital available is quite important to us.

You know, you could even if you locked a hundred percent of your own ear wallets, I mean, still people could pull a SAM on you. You know what I mean? Like it's like I like I i it's just a strange thing where I feel like it maybe creates like like I say, virtue signaling, but also a false sense of security and so more ways to maybe manipulate. but that said,

like I say, with takeovers launching, which again, then I'll like talk about changes happening overnight. I think that one literally happened overnight, right? did that actually go live, Omar?

locking in their funds through conviction. cause you're sort of like needing to buy trust, which I feel like you should just earn. Okay. Anyway, that might I don't know. I'll I'd love feedback from the community on that one because every once in a while I'm miscalibrated on something. okay, other stuff. And then we gotta tear through this man. Holy smokes, we're over time. okay. So exploits coming up. I'm gonna be there.

People have started using like an AI'd version of my headshot. That's totally not me. That was never quite. Yeah. But all right, whatever. anyway, yeah, this one's supposed to be a quantum fireside chat. we'll see if we can pull this together, bring some quantum experts into exploit to talk about, you know, the current state of quantum and what's real and what's not real. it'll be a cool opportunity to talk about sixty three and to talk about forty eight. And I guess, Omar, you you're thinking you've heard from some of the cubeys that they're gonna be there.

they fully expect milestone three to be broken. you know w with with relatively smart approaches, but you know, known approaches. milestones four are are gonna get to be harder. and so yeah, we are gonna start to get to the point where we're a little bit more impressed with all these things. All right. What do you what do you

you know, before like getting into it and trying it. And then getting into it and trying it, I was able to get a little bit further, a little bit further, a little bit further. I was really surprised when I kind of got up into like the high 300 bits. and now we're talking about almost 500 bit. So like, am I surprised? I wasn't surprised because I had tried to do this myself and sort of found like I had surprised myself at what was possible sort of ahead of these results.

But I think the world is surprised. I think we haven't been paying a lot of attention to this because it's sort of like, you know, back in the early days of computer science, we sort of ran such a challenge and we really haven't looked back. I think people have been a lot more focused on quantum. And so yeah, I th I mean I think it's really quite cool. And I think the more the more the quantum threat exists or the more quantum pushes, you know, what is capable, I think you're gonna see classical methodologies.

Innovate and hurry up to kind of catch up, which could be one of the coolest things about quantum computing if all else fails, right? Yeah. What else we got?

We can go engage other people about launching challenges and you know, with the existing partners that we have and the existing results that we're getting, you know, it's a it's a fairly easy story to digest.

you know, they start coming to us and saying, Hey, we have this really cool idea. We'd love to host it on Enigma.

Right. And so Omar kind of talked through some of the like internal logistics of what makes it hard. but really the the trick here is to slowly calibrate until you essentially create the minimum product that a classical computer can't do, but a quantum computer can do. It will be achievable. Like, you know, we're we're not going to find that it can't be achievable. It's just a combination of like how much do we care about like the

you know, the ac making the accuracy of the simulation so that approximate methods don't work matter how much you make the how much burden you put on a quantum computer to make it so it maybe can't, you know, achieve it. you know, the the it's just about getting sort of the right mix of things. And I do guess that that's probably going to be like forever a cat and mouse game, you know, where you you find the parameters that make it unachievable for classical systems.

And then maybe it'll just last six months before somebody breaks and then they'll break it, you know. and and then you'll adjust and make it a little bit harder. Yeah.

you know, what the the experts believe to be impossible into you know, something that really is impossible or impossible to the best of our knowledge.

So yeah, r it's really kind of interesting to see how how people are attacking that. Okay, we've gotta keep going though, 'cause we are so over time and all right, sentiment hit a y you handle it, Omar. I'll keep my mouth shut.

The guys that are out there trying to break the solutions. And so there was a note from one of the miners saying just that everything so far has been a great experience. You know, Enigma has been solid and straightforward and it's just like such a great mining experience. And we are over the moon to hear that. We've been working tirelessly to put this all together and to see that, you know, people are getting joy and also, you know, large prizes out of it is you know super awesome. that being said, there are also been comments that the challenges are really hard, and that is good. You know, we don't

We don't want them to be easy. It's you know, we need we were trying to really drive you to innovate and we've definitely seen a lot of people rise to the occasion.

is you know, we were originally projecting like those prize pools, like we're on track to hit a million dollars by the end of the summer. I don't think that's gonna the case anymore. I think we've seen, you know, like the swings of towel prices probably have a set a lot lower. We should sort of update people just transparently on what we're trending towards at this point.

But as always, thank you for your strong participation, all your support. some of the things in the sentiment highlighted us getting you know, coverage in some forums that we don't usually get coverage in. That probably has as much to do with you guys and all of the support that you do actively supporting us and bringing the message to everybody. So kudos to you guys for that. And yeah, hopefully we continue to see the success like that continue.

right now things are moving smooth and we're hoping that with all of these updates to Bittensor puts us on track to kind of keep a smooth operation going. All right. We'll talk to you again soon. Thank you, everybody.