Ian Issa, founder and CEO of Bitcoin mining firm HashNet, says the industry’s rush into AI hosting is a rescue mission for miners who built single-coin, debt-funded businesses, not evidence that mining itself is broken.

In this interview, Issa walks through HashNet’s six-coin, no-debt model, how its Alpha Engine software re-points hardware between coins in 12 milliseconds, and why the company only recently opened a platform that ran for four years with a $2 million institutional minimum to retail investors.

Q1.  You diagnosed pure Bitcoin mining as a failing business model back in 2022, years before companies such as Core Scientific, Riot Platforms, and MARA Holdings began expanding into AI infrastructure. What specific signals were you seeing at the time that much of the industry appeared to overlook?

It was just math. Bitcoin mining halves its revenue roughly every four years, yet many operators responded by buying more machines and taking on debt near market peaks. When markets turned in 2022, that model cracked. We questioned a core assumption: why should mining revolve around one coin? We launched during the downturn because a model that only works in good conditions is not resilient. Today’s AI pivots show much of the industry is only now adapting.

What the industry missed was simpler than any data point. It was a belief nobody thought to question: that mining means Bitcoin, full stop. We just asked a different question. Why are we married to one coin at all? That is the whole company, really. We opened in July 2022, in the middle of the crash, when the network was posting negative difficulty adjustments. Was it a terrible time to start? Yes, but that is exactly why we did. If your model only works in good conditions, you do not have a model. The miners signing AI deals right now to get out of pure mining are just catching up to 2022.

Q2.  CoinShares projects that as much as 70% of public miners’ revenue could come from AI-related activities by the end of 2026. Does that suggest the traditional mining business is becoming obsolete, or do you still see a sustainable future for mining?

The 70% number is real, but look at what it is actually measuring: announced contracts, not the money that has come in yet. Over $70 billion in AI deals have been announced across public mining companies, and right now the market is rewarding them just for announcing those deals, before any of that revenue shows up.

Here is the part people are missing: these companies are not struggling because mining itself stopped working; they are struggling because their version of mining stopped working. If it costs you more than $80,000 to produce one Bitcoin, and Bitcoin is trading below that, you are losing money on every single coin you make.

When your hardware is only built for one coin and you are carrying debt on top of that, there is only one direction that ends. So renting out their power to AI companies makes sense — I would do the same in their position. But look at what they sold to pay for it. Public miners sold a record amount of Bitcoin in the first quarter of this year just to fund that pivot. So I would actually flip the question. It is less about whether mining is obsolete, and more about whether you were ever really in the mining business if you have to sell your own Bitcoin to escape it.

Mining itself is fine. What is dying is single-coin mining built on borrowed money, and honestly, it should. We mine six coins and carry no debt riding on any single one of them. We never needed an escape hatch in the first place.

Q3.  Many public mining companies are increasingly positioning themselves as data center operators. Does that create opportunities for HashNet, or does it introduce a different set of competitive pressures?

It is an opportunity, plainly. Every megawatt that leaves mining for AI is a competitor leaving the network, which means hashrate exits, difficulty drops, and conditions improve for everyone who stayed. The companies converting to data centers are not my competition anymore. They are essentially landlords for AI chips at this point, running a different business entirely.

The real pressure is not competition, it is the narrative. When the biggest names in the sector spend a couple of years telling the world that mining cannot survive on its own, people start assuming that is true across the board. Part of my job now is reminding people that when a company says mining is dying, they are usually describing their own balance sheet rather than the industry as a whole. The companies in trouble borrowed against one coin’s price, and what they are facing is a debt problem dressed up as a mining problem.

Q4.  Walk us through how the Alpha Engine works in practice. When it re-points a machine from one coin to another in roughly 12 milliseconds, what market conditions is it responding to, and how does that decision ultimately translate into more Bitcoin produced?

A machine cannot switch from Bitcoin to Zcash because they use different algorithms. Our switching happens within compatible algorithms. We mine six coins across four algorithms, rescore profitability every 90 seconds using proprietary metrics, and automatically redirect machines in about 12 milliseconds when conditions change. We continuously convert mining proceeds into Bitcoin every eight hours rather than holding mined altcoins.

Here is what really happens. We mine six coins across four algorithms — Bitcoin and Bitcoin Cash on SHA-256, Litecoin and Dogecoin on Scrypt, Zcash on Equihash, and Kaspa on kHeavyHash. Inside each algorithm, the coins are competing for the same machines. The Alpha Engine scores every coin every 90 seconds, and three things go into that score: hash reward, which is what the network pays for your computing power; profitability, which is what is actually left after the cost of producing it — the net, not the gross; and revenue rank, which is how the coin stacks up against the other options on that hardware right now. How we weigh those three stays in the building. When the score flips, the machine re-points in about 12 milliseconds with no downtime.

The conditions that flip a score are usually invisible if you only watch price. A coin’s price can sit flat while its difficulty drops because other miners gave up — that coin just got more profitable, not less. A rally in one coin pulls hashrate off its neighbor, and suddenly the neighbor is the better use of the same machine. These windows last seconds. No human with a dashboard catches them.

The last step is what makes this Bitcoin production rather than altcoin speculation: everything converts to Bitcoin continuously, delivered to operators every 8 hours. We never hold what we mine. The altcoins do the work. Bitcoin is where it ends up.

Q5.  You describe HashNet as one of the most transparent mining operations in the industry. Why do you think that level of transparency remains uncommon, and what does greater transparency mean for operator trust?

Because most of the industry could not survive being transparent. That is the honest answer. If your model depends on quietly using client assets, or delaying obligations when margins get tight, or showing a prettier picture than the real one, the last thing you want is anyone being able to check. The companies that blew up over the past four years all had one thing in common. By the time anyone could actually check what was going on, it was already too late, and that was by design.

We went the other way, and we built it around the one thing that matters most: does the Bitcoin actually show up. It does, every 8 hours, on-chain, delivered directly to operator wallets. Those wallet addresses are public, and every delivery sits on a public block explorer, so nobody needs our permission or has to take our word for the part that actually counts.

People tend to assume that 8-hour cadence is just a service feature. What it really is, is discipline. An operation that has to deliver on-chain every 8 hours cannot quietly develop a hole in its books, because there is nowhere for a hole like that to hide before it becomes visible. Trust stops being a feeling you have about a brand and becomes something you can check yourself, and that is the only kind of trust that survives a bad market.

Q6.  The Zcash example is striking — it rose from roughly $50 to above $600, briefly overtaking Monero by market cap, before surrendering much of those gains. Over the past four years, how much of HashNet’s Bitcoin output has come from capturing opportunities in altcoin volatility versus conventional Bitcoin mining?

I am not going to give you a clean split, and that is not me dodging the question — it genuinely changes constantly, which is the whole point. There is no Bitcoin division and altcoin division inside HashNet. It is one fleet, and every machine points wherever the math says, re-scored every 90 seconds. Some weeks SHA-256 carries everything. During a run like Zcash had, the Equihash lane punches well above its weight. The system does not care which lane wins, only that one always is.

What I will say is that the altcoin lanes are not a side dish, and Zcash is the clearest way to show why. Take its run from around $50 to above $600. Because our machines mine Zcash and convert to Bitcoin continuously, every unit of Zcash we mined on the way up was converted at a far higher value than it had at $50 so the same mining effort produced significantly more Bitcoin. And because we never hold the coin, when it later fell back, none of that captured value was at risk. It was already Bitcoin. That is the entire point of the model: we are not trying to predict the top or bottom, we are turning the movement itself into Bitcoin as it happens.

That kind of volatility is exactly what terrifies single-coin miners, and it is what feeds us. Four years in, we have captured the major moves across the coins we mine, in Bitcoin terms, without ever being left holding the coin when it fell. As for how much of our output comes from the altcoins, it is enough that I would never run this model without them.

Q7.  For four years, HashNet operated as an institutional-only platform with a $2 million minimum entry. What gave you the confidence to open it to retail participants, and what changes, if anything, when everyday customers enter the ecosystem?

Honestly, the confidence came from boredom. Four years of the same record, every 8 hours, not one missed delivery. The institutional years were the real stress test. Clients coming in at a $2 million minimum run due diligence that would make a journalist blush, and we passed it for four years straight, through conditions that bankrupted public companies.

Eventually the question flipped. It stopped being whether this works and became why only people with two million dollars could access it.

What changes now that everyday customers can come in is nothing. They are buying the same machines our institutional clients bought, running on the same Alpha Engine, with the same Bitcoin delivered every 8 hours and the same on-chain proof. The only thing that changes is who gets access, and if anything, everyday customers fit our model better than institutions do.

An institution can hire auditors. A regular person cannot, but they can open a block explorer and watch the deliveries land themselves. We built the whole thing so that checking the part that matters only takes a URL and a wallet address. The people this affects most are the ones who have been let down by this industry already. What they want is something they can verify themselves.

Q8.  You went from a career-ending athletic injury to building at the intersection of AI and crypto before either became mainstream investment narratives. How much of your advantage as a founder comes from approaching the industry as an outsider to traditional finance?

Less than people think, and for a different reason than most expect. The bigger advantage came from sport itself. Sprinting at a national level drills two things into you. The result is decided by work nobody saw, and the clock keeps its own time. Markets work the same way, and in a sense, the Alpha Engine is an athlete’s view of markets turned into infrastructure, built around the idea that whoever reacts fastest and most consistently, without ever getting tired, wins.

The outsider part did help in one specific way. I never inherited the industry’s assumptions. I came into crypto in 2017 without anyone from traditional finance telling me which questions were already settled. So when we applied AI to crypto markets, I did not know that was supposed to be quant desk territory only. We just built it.

When we built dynamic rewards in DeFi, I did not know it was supposed to be impossible. And when I looked at mining in 2022, I did not carry the industry’s conviction that Bitcoin-only was the only serious approach. That is three for three, and it comes down to the same reason each time. I was not carrying anyone else’s blind spots.

Q9.  Both Token Toolkit and Hedge-Finance were acquired by private equity firms. With HashNet, however, you seem to be building for long-term scale rather than an exit. What makes this venture different?

The first two were products. This is a category, and that changes everything about the timeline. Token Toolkit and Hedge-Finance solved real problems inside markets that already existed, and when a private equity firm values a product like that properly, selling can be the right move for everyone involved. Twice, it was.

Liquid Hashrate is different. It is not a product inside an existing market, it is a claim that mining — the last major hard asset with no liquid layer — can work in a completely different way. You do not build a category just to flip it. You build it to define it, and being first only pays off if you are still standing there when the category matures.

There is a structural reason too. HashNet’s credibility comes from its continuity. The four-year record, the on-chain proof, the every-8-hour rhythm, all of that exists because the same operation kept the same promises without a single gap. You cannot sell that and keep it running at the same time. So I am not building toward an exit. I am building toward a version of this industry where what we do becomes the standard, and we are the ones setting it.

Q10.  HashNet has opened a second entity in Dubai and is expanding across the MENA region. Where do you see the company five years from now? Is a public listing part of that vision, or does blockchain-based transparency reduce the need for traditional public-market validation?

Five years out, the goal is simple to say and a lot harder to do. I want HashNet to be the top mining company in the world, full stop. Not the biggest by some vanity metric, but the most transparent, the most technically advanced, and the most accessible to the people who want in. Dubai anchors our move into MENA and the Eastern Hemisphere, and it is the second of several jurisdictions we are building into. I will keep the rest directional on purpose, but the logic stays the same. We go where the operators are, in places that have built real rules for digital assets instead of pretending the industry does not exist.

On a listing, we have no plans to go public as of now. My focus is entirely on making sure our machines and the software running them are the best in the business, and I can say that with confidence today, because of what is happening around us. The same companies that used to push us on mining technology are now converting their facilities into AI data centers. Their engineering attention is leaving mining. Ours is going deeper into it, on purpose, while everyone else heads for the exit. Five years from now, I want that gap to be impossible to miss.

On public markets more broadly: public miners report quarterly and call it transparency. We deliver on-chain every 8 hours, around the clock, for anyone to check whenever they want. We do not need a stock ticker to prove the Bitcoin shows up.

Q11.  You’ve built a system designed to capture the upside of altcoin volatility and convert it into Bitcoin output. Yet you began building it shortly after a career-ending injury took away the pursuit you’d spent most of your life training for. How much of HashNet is a financial system, and how much of it is your personal answer to the question of what comes next when the game you’ve devoted your life to suddenly ends?

Losing athletics forced me to start over, much like anyone recovering from a major setback. I redirected the work ethic I learned in sport into crypto, showing up every day and improving incrementally. HashNet is a business, not therapy, but that experience shaped one principle: the things that matter should be verifiable rather than taken on trust.

What I had left was the work ethic. That was really it. So I pointed it at something new, crypto, in 2017, and treated it the way I used to treat training. Show up every day, do the boring work, get a little better. That part carried over completely. The rest I had to rebuild from scratch like anyone else would.

As for HashNet, I built a business, and I will not dress it up as anything else. But I would be lying if I said the injury taught me nothing. It taught me not to take anything on faith, including the good situations. You can probably see that in how we built the company. Whatever matters, you can verify yourself. That part is personal.

Q12.  The industry has now committed tens of billions of dollars to AI-related infrastructure. HashNet’s position is that it never needed to pivot because it was never built around the model others are now moving away from. But if capital continues flowing into AI at this scale, is there a future in which HashNet incorporates AI workloads as part of its business, or do you view mining as the company’s permanent core focus?

I would not say never — that is how founders end up regretting their own quotes later. The bar I would set is this: it would have to produce Bitcoin more efficiently than pointing the same power at the six networks we already mine. That is the only test anything really has to pass here.

Right now, AI hosting does not clear that bar. The revenue is real, but converting a mining facility into AI-grade infrastructure runs several times the capital per megawatt. The companies doing it are covering that gap with debt and by selling the Bitcoin they spent years stacking, trading the asset itself to fund the infrastructure. That is not a trade we need to make.

The bigger point is that we never needed the exit they needed. The AI pivot is a rescue from single-coin economics, and we do not have single-coin economics. Mining is not our focus out of stubbornness. It stays our focus for as long as it is the most capital-efficient way to produce Bitcoin that exists. The day something genuinely beats it, we will look at that honestly. Right now, nothing comes close.

Three Additional Questions

Q13.  Most mining companies talk about hashrate, energy costs, and infrastructure. HashNet talks about capital efficiency and Bitcoin production. Should investors stop evaluating mining companies based on how much Bitcoin they mine and start evaluating them based on how effectively they convert capital and power into Bitcoin they actually keep?

Hashrate and energy costs still matter — they are the inputs. But judging a miner on how much Bitcoin it digs up is a bit like judging a factory by how fast the machines spin instead of what leaves the loading dock. Production tells you there is activity. It does not tell you whether the operation is healthy. The industry’s own numbers make that point better than I can. Public miners produced record amounts of Bitcoin over the past two years and still ended up holding less of it, simply because of how that production was financed. They had to sell what they mined just to keep going. Mining more and keeping less is not growth, it is a treadmill.

The metric that matters is the one you named. How efficiently does a unit of capital and power turn into Bitcoin, and how much of that Bitcoin does the operation keep? That is what tells you whether a mining company is real in Bitcoin terms, and Bitcoin terms is the only honest scoreboard for this industry. I would genuinely like to see the whole sector measured that way – us included. Especially us.

Q14.  HashNet has been mining ZEC since 2022, and your press kit uses Zcash as the clearest example of how your model captures altcoin rallies. After the Orchard Pool bug was disclosed — a flaw that apparently existed since that same year — Arthur Hayes exited his entire ZEC position. As someone who was mining that coin the entire time, and whose whole brand is built on transparency and on-chain trust, what do you make of what happened, and does it change anything about how you think about ZEC going forward?

What happened with Zcash is serious, and I am not going to spin it smaller than it is. A flaw that could theoretically let someone counterfeit coins sat in the protocol for years, and because of how the shielded pool works, nobody can mathematically prove it was never used.

If you are holding ZEC as a long-term position, that is not noise. It hits the one thing a monetary asset cannot afford to lose: confidence in its own supply. I can understand why some holders, Arthur Hayes among them, chose to step back and reassess — those are fair questions to ask of any asset in that situation.

Here is the distinction that matters for us. We have mined Zcash since 2022, and we have never held it. Output converts to Bitcoin continuously and gets delivered to builders every 8 hours. The question that weighed on those holders — whether you can trust this coin’s supply for the next five years — is one we never had to answer, because we were never betting on the coin in the first place. We collected what its network paid for hashpower, then left.

This is why we built it that way. People hear us say we capture altcoin volatility and assume we only mean the upside. The downside works in our favor too. Zcash is the only coin on the Equihash algorithm, so those machines mine ZEC continuously regardless of price, and the output converts to Bitcoin continuously as well.

When ZEC is rising while Bitcoin is falling or flat, every conversion happens while Bitcoin trades lower, so each unit of ZEC value buys more Bitcoin than it would otherwise. That is when our output grows most. When ZEC comes back down, none of that earlier value is at risk — it was already converted to Bitcoin before the drop.

One more thing, because I do not want this taken the wrong way. None of this is a shot at privacy. On-chain privacy matters. It is one of the most important problems this industry is still working through, and Zcash’s cryptography is some of the most serious work in the space.

The lesson I take from this is simple: whatever a system keeps private, the things people depend on still need to be checkable. That is the balance we built our whole operation around. The deliveries are on-chain, every cycle, and you do not have to take my word for it. You can verify it yourself. Privacy where it belongs. Proof where it counts.

Editorial disclosure: This interview is published as an exclusive with HashNet. The interviewee reviewed their own responses before publication. Bit Gazette maintains editorial control over presentation, editing for clarity and length, and publication decisions. This content is not sponsored.