Is 51% a fallacy? The truth about the 30% Computing Power attack on Monero.

Written by: on-chain view

Thank you for your insightful knowledge, I've learned a lot. I truly find that this matter can be big or small, insiders see the essence, while outsiders just watch the excitement. Let me add a few more points:

1. Seeing overseas big V using this incident to Fud Bitcoin, the reason is that the current mining pool hash rate share shows that the top two mining pools, Foundry USA (33.6%) and AntPool (17.9%), have already exceeded 51% combined, leading to a crude conclusion that if the two major mining pools collude, Bitcoin is doomed. This is a typical case of an outsider looking at the excitement, as he ignores two points:

1. The proportion of two mining pools being 51% and one mining pool exceeding 51% are completely different matters, one above and one below.

2. The hash rate of a mining pool does not represent the complete acquisition of miners' hash rate. When the hash rate of a single mining pool is too high, miners will often take the initiative to switch hash rates to avoid risks.

So Satoshi Nakamoto's POW consensus has reached a subtle balance now, integrating factors like computing power, economics, and interest games, making it almost impossible to break in the short term. So those who are Fudding BTC should calm down.

2. Acknowledge that what Teacher Zhang Ren said about Monero's issues does not equate to the problems of POW. Even if POW has security risks under extreme conditions, it does not mean that POS is the completely optimal solution.

In fact, the problem with POW is that ASIC has made the optimal choice as a moat, avoiding the pitfalls of general CPU/GPU mining. Furthermore, even in a purely CPU/GPU mining chain, attempting to attack using a disclosed miner bribery method faces various challenges. For example, exchanges can increase the number of confirmations, and miners can add checkpoints, all of which can reduce the probability of being attacked.

You see, when discussing the issues of POW, the focus is on POW itself. Comparing cross-consensus may lead to misunderstandings. In fact, there are security risk boundaries that exist outside of each consensus, and of course, the methods of counterbalancing are also different; we cannot favor one over the other.

3. I saw Teacher 0xTodd forwarded my post and talked about the concept of "selfish mining." Simply put, this means that miners should immediately broadcast the blocks they mine, but selfish miners will secretly hide the blocks they have mined, forming a "private chain." When honest miners announce a new block, selfish miners will suddenly release their hidden longer chain, rendering all the work of the honest miners invalid.

This is actually a very rogue approach and is the main attack method of Qubic this time. In fact, its computing power has not truly reached 51% and may only control about 30%, which can briefly achieve the theoretical "double spend attack"? Because, using 30% of the miners for selfish mining, a shadow chain is formed. When honest miners mine a new block, Qubic suddenly releases its hidden longer chain, causing a large number of real miner blocks to become invalid, theoretically causing the destructive effect of over 51% of computing power. Furthermore, if the distribution of the mining nodes controlled by Qubic's mining pool is broad enough, it can also exploit factors such as network latency to further reduce the proportion of computing power, achieving the same effect of controlling the entire network's computing power.

Therefore, the attack on Qubic this time has a very high degree of randomness and concealment, which means that once this method is made public, the threshold for repeating the same trick will become higher.

4. However, I discussed a possibility with the security expert n33k, which is that Qubic might not repeat its old tricks, but instead use a "boiling frog" style attack to operate, further bribing miners to increase its own pool size, and then letting some miners intentionally empty blocks, causing chaos for the normal operation of the Monero network.

If this continues, it will lead to more and more Monero miners fleeing, as their profits decrease and the experience becomes really terrible. As a result, the scale of computational power controlled by Qubic will gradually increase, eventually surpassing 50%. At that point, it will be game over, and no one will be able to play. This method of chronic attack is actually quite scary.

Although there is no reason to prove that Qubic needs to do this, the possibility of such a "parasitic" chronic attack does exist. In the early stages, Qubic doesn’t need to worry about a portion of miners mining empty blocks in Monero; they will still receive $XMR rewards and can also engage in AI training. In the later stages, if Monero's profitability declines, they may attack other chains such as Grin and Beam. Throughout this process, Qubic can always adhere to its main line of AI training, making the logic reasonable.

Because when the demand for AI computing power grows exponentially, and mining is no longer the only place for computing power, the rules of the game have changed. The original cost of attacking the network was simply "burning money," but now there is AI training as an "additional funder" to foot the bill— the cost of attacks is hedged by AI earnings.

This is my biggest concern with that article: AI demand is breaking the fundamental assumption of general-purpose CPU/GPU POW mining - "miners rely on mining profits, so they will maintain the network." When computing power has more profitable avenues, this assumption no longer holds. Although this process will be slow, there is always the possibility.