Monero Hashrate Centralization: What Qubic Taught Miners

For most of Monero’s history, “hashrate centralization” was a theoretical worry. In 2025 it became a practical one, when the Qubic project steered a large block of CPU power at Monero and used it to orphan other miners’ blocks. The campaign is over, but the episode showed exactly where Monero’s mining is fragile and what ordinary miners can do about it. This post focuses on the mining lessons; for a news timeline, see our news category.

Why hashrate distribution matters

Monero, like Bitcoin, decides which chain is valid by accumulated proof of work. Whoever controls a large share of hashrate gains power over the most recent blocks:

• Reorganizations. They can build a private chain and release it to replace blocks others already accepted.
• Double spends of their own payments. By reorganizing away a transaction they made, they can try to spend the same coins twice.
• Censorship. They can refuse to include certain transactions and orphan blocks from other miners that do.
• Starving honest miners. Orphaned blocks pay nothing, so honest miners earn less and may leave, which strengthens the attacker further.

What a majority miner cannot do is equally important. It cannot spend coins it does not own, forge ring signatures or change consensus rules for nodes that do not upgrade. Your existing XMR remains safe; the risk is to recent, unconfirmed history.

What Qubic did, briefly

Qubic is a separate blockchain whose “useful proof of work” lets its computors earn by doing other work. In mid-2025 it redirected that system toward Monero mining. According to The Block, the mined XMR was sold for USDT, which was then used to buy and burn QUBIC tokens.

Key points from published reporting and research:

The strategy was selfish mining: rather than publishing each block immediately, the attacker keeps found blocks private and releases them when doing so will orphan honest miners’ blocks. Done well, selfish mining can earn more than a fair share with less than 50% of hashrate.

Did it work?

The research paper “Inside Qubic’s Selfish Mining Campaign on Monero” by Suhyeon Lee and Hyeongyeong Kim (arXiv, December 2025, revised February 2026) concluded that:

• Qubic did not hold a sustained majority, despite its framing.
• It alternated between classic selfish mining and more defensive release decisions instead of following one optimized strategy.
• It gained no reward advantage over honest mining. The disruption was real, but it did not pay off in block rewards.

Other estimates, such as one reported by Protos, put Qubic’s real share during the September reorg at roughly 28–35%.

Key point: The 18-block reorg happened without a confirmed majority. Under selfish mining, a well-connected minority with luck can rewrite surprisingly deep history. That is why concentration below 50% still matters.

The bigger problem: centralized pools

Qubic was a single entity directing many machines. But most of Monero’s hashrate already flows through a handful of ordinary pools, and a pool operator, not the individual miner, decides what goes into each block template.

At the time of writing, MiningPoolStats shows a picture like this (shares of self-reported pool hashrate, which add up to more than the network estimate, so treat them as approximate):

The top three pools together report around two thirds of pool hashrate. None of them is malicious as far as anyone knows, but a compromised, coerced or simply buggy operator could cause the same kind of reorgs Qubic did. Miners rarely notice, because their dashboards keep showing payouts either way.

Note: A higher total network hashrate makes attacks more expensive, but only if that hashrate is spread out. A big network dominated by three operators is still only three operators.

How the network responded

The community and developers reacted on several fronts:

• Pressure on pool share. Community members urged miners away from Qubic’s pool, and its share fell from its July peak before climbing again in August.
• DNS checkpoints. Monero has long had an opt-in mechanism called MoneroPulse that publishes checkpoint block hashes in DNSSEC-signed DNS records. After the September reorg, researchers argued that temporary checkpoints were the quickest defense against deep reorgs, with the trade-off of trusting a small set of operators. P2Pool’s README now recommends running monerod with --enforce-dns-checkpointing, which makes a node roll back and follow the checkpointed chain if it disagrees.
• Longer confirmations. Merchants and exchanges were advised to wait well beyond the usual 10 blocks for large deposits during the reorg period.
• Research. Researchers studied the campaign in detail, which is how we know it did not out-earn honest mining. Longer-term protocol defenses against deep reorgs remain a research topic rather than a shipped feature.

What you can do as a miner

1. Mine on P2Pool

P2Pool has no operator. Every participant runs their own node and builds their own block templates, so nobody can redirect the combined hashrate. It is also 0% fee. A single desktop belongs on the mini or nano sidechain. Our P2Pool setup guide in the mining category walks through it.

2. If you use a centralized pool, avoid the biggest

Check MiningPoolStats and pick a pool that is not near the top. Fees and payout thresholds differ, but between reputable pools the expected income difference is small compared with the systemic benefit of spreading hashrate.

3. Run your own node with sensible flags

A node you control lets you see reorgs and verify the chain yourself. The P2Pool README’s recommended flags include --enforce-dns-checkpointing and --enable-dns-blocklist, and priority nodes run by well-connected pools to reduce your own orphan rate.

4. Watch the network, not just your dashboard

Bookmark MiningPoolStats and the P2Pool Observer. If one pool or an unknown entity starts approaching 40% or more, it is worth moving your hashrate and talking about it in community channels.

Tip: Merge mining Tari through P2Pool adds a little extra income without reducing your Monero hashrate. It removes one of the reasons people give for staying on a centralized pool that offers Tari merge mining.

Warning signs a miner can spot

You do not need special tools to notice when something is off. During the Qubic campaign, several signals showed up in places ordinary miners already look:

• Rising orphan or uncle counts. On P2Pool Observer and on most pool dashboards, a sudden run of orphaned Monero blocks is a red flag that someone is withholding and releasing blocks.
• A new name near the top of the pool list. Qubic went from a minor entry to one of the largest pools within weeks. Fast growth by an entity with an outside agenda deserves attention.
• Exchange notices. When an exchange pauses deposits or raises its confirmation count, it is usually reacting to reorg risk.
• Node log warnings. A monerod that reports reorganizations of more than a block or two, or MoneroPulse checkpoint mismatches, is telling you the chain tip is being contested.

None of these signals requires you to do anything drastic. They are prompts to check where your own hashrate is going and to make sure your node and P2Pool are up to date.

Lessons for the future

• Incentives can be bought. Qubic showed that an outside project can rent CPU power at scale and aim it at Monero. RandomX keeps mining open to ordinary computers, but that also means any large fleet of computers can join.
• Selfish mining is a practical tool, not just a paper idea. Real reorgs happened with a minority of hashrate.
• Hashrate spread is a shared responsibility. Protocol fixes take years; miners can move their hashrate in minutes.

Qubic has moved on, but the pool distribution that made the episode possible is still with us. Choosing P2Pool or a smaller pool is the most direct thing a Monero miner can do to keep the network resilient.