China’s Ministry of Commerce has just banned two local companies from exporting anti-epidemic products for their quality problems. As one of them has blockchain in its name, the notice from the Ministry of Commerce immediately evokes some criticism calling blockchain a scam.

Amid the coronavirus outbreak, the blockchain technology is meant to track the supply of virus prevention materials. It is a real slap in the face as the blockchain company was disqualified from exporting anti-epidemic products for quality problems. This article will tell you that blockchain shall not take the blame.

In fact, under the current background, traceability and anti-counterfeiting can hardly be realized as a result of the “garbage in, garbage out” (GIGO) problem long existing in computer science.

GIGO is the concept that flawed, or nonsense input data produces nonsense output or “garbage” in computer science, and it is now commonly used to describe failures in human decision-making due to faulty, incomplete, or imprecise data.

In the blockchain space, it turns out to be a fatal problem since the blockchain can only ensure the data is tamperproof, and if the input data is incorrect, the tamperproof feature will be very bad for the credibility of the system. If incorrect data keeps being input, the system will finally become a garbage database, and if you propose to get rid of the “tamperproof feature”, then the system will have nothing to do with blockchain.

Based on the logic above, blockchain seems to be contradictory to traceability, rather than a solution to it.

Let’s first check how the bitcoin blockchain gets around the GIGO problem.

In Satoshi’s bitcoin whitepaper, bitcoin takes the longest chain rule of PoW algorithm, but in fact it is the strongest chain rule which means the chain representing the most work will be deemed as the main chain, and the chain with the smaller workload will be the “fork chain” (represented by Bitcoin Cash).

And the key here is, the workload is contributed by computers (also known as mining machines), which reduces human involvement to a great extent. Then the data computed by computers will reach a consensus through the timestamp and consensus algorithm, and the consensus serves as a standard, violation of the standard will be considered a false bitcoin.

In this way, bitcoin gets around the GIGO problem, thus solving the problem of token data forgery. However, we need to understand here that the data of bitcoin itself comes from the chain, so the problem can be easily solved.

But if we were to upload the offchain data to the chain, the annoying GIGO problem would appear again. That’s because there’s currently no way to ensure that data off the chain is unforgeable or correct.

Take masks, the much-needed anti-epidemic products, for example. As the key material of melt-blown fabric prices keep rising, to make more profits, mask maker A uses ordinary cloth to replace the melt-blown fabric, and operating personnel record the material as melt-blown fabric onto the blockchain. After the mask is produced and rolls into logistics, the logistics personnel will input the logistic info onto the blockchain. At last, the receiver believes that the mask is epidemic prevention after he scans the qr code to learn the mask information, but the mask is actually an inferior product. In this case, we only assume that the input data at the source is incorrect, while in real scenarios, it is possible that the input data is wrong in any steps, and as long as there’re mistakes in one step, the information on the blockchain will be misleading.

It indicates that blockchain cannot solve the “garbage input” problem of data off the chain. So can we take a page from bitcoin to address the authenticity of offchain data by removing human involvement and using consensus algorithms?

This is where the IoT (Internet of things) and AI (artificial intelligence) come into play, in simple terms, it is to use machines or higher-order robots to perform repetitive data collection and then process the data with algorithms to produce more reliable offchain data by using IoT and AI technologies. Finally, the data is uploaded to the blockchain to ensure it’s difficult to tamper with.

The combination of IoT, AI and blockchain could serve as a solution to an anti-counterfeiting and traceability system. As long as one link is missing, the whole system is vulnerable (untrustworthy). For the time being, it is actually very difficult to be realized. A short-term alternative is to strengthen the evaluation and supervision of data governance off the chain.

Therefore, look back on the case at the beginning of the article, it is not blockchain that is a crappy technology, but the company’s human involvement in it.