Bitcoin has an energy-chewing PR problem. Aussie blockchain firm CloudTech Group knows this, which is why it teamed up with RMIT University to form a research lab to help tackle the issue.

The research unit is called Green CryptoLab, and it’s on a mission to turn the crypto industry a more palatable shade of green. And we’re not talking green candles on price charts (although we wouldn’t complain if the two things went hand in hand).

To learn more, Stockhead got in touch with Dr Hai Dong from RMIT’s School of Computing Technologies and CloudTech Group CEO Kevin He.


Hello, both. Can you please start by giving me the Green CryptoLab mission? What’s the overall goal here?  

RMIT’s Dr Hai Dong: Our mission is to develop the next generation of green cryptocurrency technologies, including what we call Green Bitcoin. The goal is to support large-scale transactions while minimising the consumption of mining resources and rationalising the use of storage resources to create value with the lowest input.

So there’s a distinct focus on Bitcoin with this over other cryptocurrencies?

Yes, and that’s because Bitcoin is the first thing that comes to mind whenever someone thinks of cryptocurrencies. The exponential growth of Bitcoin and the industry as a whole has brought blockchain to the limelight, highlighting its numerous benefits such as transparency, accountability, and potential overhauling of traditional systems.

Kevin He: Bitcoin is digital gold to us, because there is a finite number of total Bitcoin supply, which will be 21,000,000 BTC in total. And BTC, as the first and biggest crypto project – people’s consensus is well formed.

Dr Dong: However, gross adoption of Bitcoin does have its drawbacks, primarily stemming from its consensus algorithm known as PoW (Proof of Work).

Can you explain those drawbacks a little more? 

Dr Dong: The consensus algorithm is done to verify transactions on the blockchain and the execution of PoW is the process known as Bitcoin mining.

Due to a cap in the amount of Bitcoin released, the quicker Bitcoin reaches its end of supply, the more energy consuming it becomes to mine it.

Reports have found that, if Bitcoin were a country it would rank in the top 30 of energy consumers in the world, ahead of the likes of Argentina and Belgium. A stark figure to say the least and it doesn’t bode well due to the rising number of cryptocurrency users, and better reception on a global stage.

[Ed: Regarding the idea of crypto developing a “better reception on a global stage”, this interview was conducted before the FTX exchange implosion saw worldwide sentiment for crypto take a nosedive. Nevertheless, it’s Coinhead’s opinion that inevitable crypto regulations could help spur Bitcoin on to an era of mass adoption. Although…]


No mass adoption until green issues are solved

What brought you together for the collaboration? 

Dr Dong: Late last year I was talking to Kevin about the current problem of Bitcoin energy consumption and we found that we share so many views and values with CloudTech, so we decided to set up this lab.

Kevin He: Yes, and CloudTech Group shares the same vision with many environmentalists, that Bitcoin’s PoW consensus mechanism is energy consuming and lacking efficiency.

So we decided to work with one of the most reputable research institutions in the blockchain research area to come up with a green solution to address the underlying problems with the PoW consensus mechanism.

Got it. What sort of negative perceptions about Bitcoin and do you think many people outside the crypto sphere have? That is, those who don’t really know what PoW is. 

Kevin He: I think there’s an understanding that BTC’s carbon footprint does affect the environment and the industry. They’ve heard or know it consumes a lot of energy.

And, simply, there won’t be mass adoption of Bitcoin until its environmental issues are solved. Most investment banks and institutions won’t hold BTC until that happens. 

What needs to change for the crypto industry as a whole in order to see it flourish? 

Kevin He: It needs to become more compliant, greener, scalable, more secure and just generally have a healthier ecosystem. 

Amen to that. 


An AI-driven consensus mechanism

We understand that Green CryptoLab is building an AI-enabled crypto consensus-algorithm solution for the Proof of Work system used by Bitcoin. But that’s about all we understand. Can you explain it to us? 

[It’s complex stuff, so we needed it sent via email in bullet points to break it down. Here it is, courtesy of RMIT’s Dr Hai Dong.] 

Our proposed AI-enabled PoW consensus (abbreviated as AI-consensus) algorithm comprises the following steps:

• AI-consensus will target a set of data processing and analytics tasks, such as large-scale image, video and natural language processing and analytics etc, which require training complex, large-scale machine learning models for implementing those tasks.

• The machine learning tasks will be advertised with rewards calling for crowdsourcing.

• Miners may choose to participate in any crowdsourcing task advertised.

• Once miners decide to join a crowdsourcing task, they are required to supply their computing resources (i.e. CPU and memory) for training the machine-learning models requested for the data processing task and storage resources for training data storage.

• The participating miners will collaborate to train a machine-learning model required for a data-processing task.

• Once the machine-learning model training is completed, the participating miners will share the reward advertised for the data processing or analytics task. The reward-sharing proportions among the participants will be calculated based on the contributed computing and storage resources of each participant.

… Okay, think we got it, but we’re going to need to read that more than once. How about we cut to the benefits? 

[Again, bullet points via email was the way to go here.] 

 AI-consensus will generate useful artefacts. These machine learning models can be used to perform specific tasks such as image/speech recognition, audio recognition, email spam/fake news detection, and product recommendation, which benefits hundreds or thousands of users.

 • It will greatly reduce energy waste. Traditional consensus algorithms allow multiple participants to compete, but there is only one winner to be selected eventually. The works of non-winners are wasted, since they have to give up their contributions if a winner is selected.

In contrast, our AI-consensus algorithm will reward all the participants based on their performance and contributed computing power, which fundamentally addresses the energy waste issue.

 • AI-consensus will have high accessibility. Different from permissioned consensus algorithms such as PoS, AI-consensus will be non-permissioned, which means everyone will be free to join the community to share their learning tasks or become a contributor to share their computing capacity.  

 • And it will be completely decentralised. Another significant difference of AI-consensus is that it will be purely decentralised, meaning that there is no central power to control the learning task advertising, learning capacity crowdsourcing and contribution evaluation processes. A series of AI-based algorithms will be developed to fully facilitate this autonomy.


Industrial adoption of green innovation

What is your research telling you with regards to crypto adoption and the need for green alternatives for current crypto-validation, consensus-mechanism processes?

Dr Dong: Green blockchain technical development has become a critical component in leveraging ecologically favourable opportunities and ecosystems to offset natural resource depletion, which is recognised as the primary source of industry advancement.

Industry is seriously considering making green innovation methods mandatory due to increased demand.

In this regard, the proposed AI-enabled green consensus algorithm will have tremendous impact on traditional blockchain industries. It will fundamentally address the problem of massive energy and resources wasted for computing tasks that are unhelpful to our human society and natural environment.

The outcome of this project is expected to create an environmentally sustainable incentive technique, which would benefit Australia and even global blockchain industry to create a sustainable shared economy environment.

Big. Do you think it’s fair to compare the much larger energy consumption of traditional banking with that of Bitcoin/crypto? 

Dr Dong: Not really. It’s true that a traditional bank’s’s total annual energy consumption is about 140 TWh (Terrawatt-hour) which indeed is higher than Bitcoin network’s energy consumption (130 TWh per annum).

And there are 250,000 daily transactions happening in the Bitcoin network, while for traditional banks, in 2021, the credit card transaction per day was 1.01 billion transactions on average. That’s 4,000 times higher than Bitcoin’s daily transaction.

But worldwide bank account ownership has reached 76% of the global population as of 2021 – 6 billion people. Whereas Bitcoin is estimated to have 106 million users, and a lot of them are holders instead of network users.

So, after Bitcoin gets mass adoption like a traditional bank, the energy consumption will be a lot higher than traditional banks.


The limitations of Proof of Stake

We like that you sound confident about that mass adoption. What’s your thought on Proof of Stake crypto consensus models? Ethereum had its Merge over to PoS recently. Isn’t this already an acceptably greener alternative to the crypto-mining PoW model? 

Dr Dong: The “Ethereum merge” is a significant event indicating the consensus among the major stakeholders in the cryptocurrency industry on the issue of high energy consumption and energy waste.

The Ethereum Foundation has claimed that the transition to PoS reduced Ethereum’s energy consumption by 99.95%. However, PoS is not the panacea for the high-energy consumption issue of the cryptocurrency industry. It still subjects to several limitations.

Can you please explain what those limitations are?

Yes, the first is accessibility. Interested users must own the native cryptocurrency before becoming validators. Interested users must buy ETH tokens using fiat currency or exchanging tokens from cryptocurrency exchanges. Users interested in becoming validators also have to raise at least 32 ETH, which is a considerable amount for most people. [It’s currently about US$41k.]

The second limitation is lack of decentralisation. Since the PoS system relies on delegates chosen to validate transactions, it’s always possible for larger nodes to overpower smaller ones.

The third limitation is centralisation of power. PoS could easily lead to the centralisation of power if a small number of influential delegates or validators gain control over the authority to approve and validate transactions.

In the unlikely situation that a single entity earns the ability to have complete control over the Ethereum network, this could potentially lead to security issues besides going contrary to decentralisation, which is the most significant tenet of blockchain technology.


Understanding the vast potential of blockchain

Let’s finish by talking about RMIT for a sec. It has a reputation for being a leader in cryptocurrency and blockchain education in this country and globally – correct? 

Dr Dong: That’s right. It’s well recognised as a leading institute in blockchain education and research nationally and globally. Recently the CoinDesk rankings, conducted in conjunction with Stanford University, placed RMIT as the 19th Best Global University for Blockchain in 2022.

What is it about blockchain tech, do you think, that attracts people to your courses? 

I think it’s because the technology has the potential to change many areas of our lives – from our urban infrastructure to our financial system even political systems.

It’s a hard concept to grasp, though, wouldn’t you say?

Yes. This innovation remains largely a mystery to most people. When many think of blockchain technology, which is based on different forms of cryptography, they think of dark back rooms, complex algorithms, and indecipherable strings of numbers and letters. But while these clichés are partially true, they are more reminiscent of science fiction movies than real-world businesses.

Do you think blockchain should become part of the curriculum at secondary schools? 

What I think more generally is that we must help younger generations understand the potential of blockchain.

And I mean its potential to disrupt our current economic and social structures and its promise to decentralise our traditional monetary value systems and dissolve the institutional technologies used by centralised mediators for human trust, such as banks and government.

In this regard, yes, I fully believe blockchain should become a mandatory course of tertiary, even secondary education.


This interview was edited lightly for clarity. None of the information contained in this article should be taken as financial advice.

At the time of writing, the Coinhead journalist owns a small amount of Bitcoin and Ethereum, among a handful of other cryptocurrencies.