While our Aethir Alpha has come to an end, we are thrilled to now list their token for trading, opening new opportunities for our community to engage with this innovative project.
In November 2022, OpenAI took the world by surprise when it launched its first generative AI bot, ChatGPT. We all remember how mind-blown we were when we first interacted with this bot.
No wonder this application set the record for the fastest growing user base, hitting 100 million users in 2 months. In contrast, it took Instagram ~1,000 days to hit that same milestone. Looking back, it is fair to agree that it was an “iPhone” like moment.
In less than two years, AI moved away from a niche R&D area and became one of the biggest investment priorities for companies. Right now, it seems evident that this technology is bound to have a lasting impact on our world. Yes, the AI revolution is no joke and this technology is on an exponential curve.
But AI is not only about the code. AI is resource-intensive and requires specific physical infrastructure. This accelerated the move from CPUs (central processing units), to parallel processing with GPUs (graphics processing units). With high memory bandwidth, GPUs evolved to tackle other calculations and were proved to be really efficient for tasks such as training, refining and improving AI models.
Given the strong demand for GPUs, there has been a massive supply shortage and as mentioned by AWS CEO, “demand is outstripping supply and that is true for everybody.” Hence, this is creating an arms race for hardware and computing power.
As discussed, AI developers rely on hardware to run their models. Typically, they have two ways to get this hardware capacity: they can either run GPUs locally; or rely on cloud providers. The first solution often proves to be too expensive and not economically worth it, and over time cloud providers will prove to be an interesting alternative.
When we look at the expected growth of the GPU market, it seems that the market opportunity for cloud providers is a no-brainer.
Currently, the market share of cloud computing is dominated by centralised institutions with a lot of resources to acquire and operate these powerful hardware. However, those centralised institutious face many critics: they can be expensive, computing power is not distributed in an efficient and fair manner, and their centralised nature causes security problems.
Having said that, crypto has a role to play in providing access to those multi-node supercomputers in a more efficient, more secure, and decentralised way. This is part of a subfield of crypto called DePIN, or Decentralised Physical Infrastructure and it's going to be one of the key sectors to participate in the AI revolution.
Decentralised Physical Infrastructure represents decentralised hardware networks incentivised by token. The primary goal of this subfield of crypto is to replace monopolised cloud institutions with better access to cloud computing services for all, and to transform every user into essential stakeholders of the network using native token incentives.
• Reduces costs: As there is a more efficient distribution of cloud computing resources and less take rates from middlemen (unlike centralised services), DePIN projects are able to provide cheaper service.
• Rewards network contributors: With blockchain , value flows to the extremities of the network (users, validators, and computing providers) and does not stagnate in the middle (compared to the current centralised cloud computing models). Contributors to the networks are incentivised to engage actively and are rewarded in a fair manner.
• Scales easily: The architecture of DePIN makes scaling easier as there is no barrier to entry and less friction.
• Enhances security and resilience: Decentralised nodes means that there are no central points of failure making the network much more robust.
The idea here is simple: AI developers need more GPUs and there is currently a shortage of them. Hence, there is a lot of reason to believe that crypto DePIN projects can help drive new computing power to the market by activating latent resources with token-based incentives in a reliable and decentralised manner.
To dig deeper on DePIN, here is a great general overview of this sector .
As Aethir is an innovative DePIN project focusing on GPU cloud computing, the investment case for this project is compelling: right at the intersection of the AI revolution and a global shortage of GPUs.
“Aethir is a bit like Airbnb, where there are spare vacation homes that are being underutilized, and there’s obviously a demand for some of that unused vacation home time. [...] So, at a very base level, we service as a kind of Airbnb for GPU computing” - Daniel Wang
Aethir introduces a novel approach to cloud computing infrastructure, focusing on the ownership, distribution, and usage of enterprise-level GPUs. It functions as a marketplace and aggregator, facilitating the connection between supply-side participants—such as data centres, miners and retail GPU providers—and users and organisations from computing-intensive sectors like AI, ML (Machine-Learning), cloud gaming, and other enterprise grade participants.
Aethir achieves this with a two-way mechanism:
“There are a lot of older generation GPUs that are severely underutilised, [..] but they haven’t found consistent steady use case and demand for it to be fully utilised” - Daniel Wang
Today, a significant portion of GPUs, particularly those owned by enterprises and individuals, are not used efficiently, often due to misalignment in resource allocation, where high-capacity GPUs are used for tasks that require much less computational power. Some estimates suggest that the capacity of underutilised GPUs range from 50-75%.
One critical characteristic of Aethir is its focus on repurposing existing potential resources, rather than requiring participants to purchase new hardware.
The team behind Aethir is strong, experienced and well-fitted to properly execute the project mission. Here’s a quick breakdown of the main leadership positions:
Mark Rydon - Co-founder and CEO
Mark has held key roles at NOTA Platform, Flux Capital, Gaas LTD, Kulture Athletics, Inc., and Bechtel Corporation.
Daniel Wang - Co-founder and CBO
Prior roles at IVC (Venture Partner), YGG SEA (CIO), Riot Games (Head of International Publishing Mgmt), Riot Games - China (Head of Operations).
Kyle Okamoto - CTO
Kyle has served as the CEO & General Manager at Ericsson's IoT, Automotive, and Security businesses, CEO of Edge Gravity, and Chief Network Officer at Verizon Media.
Paul Thind - CRO
Paul previously co-founded and served as CEO at Triggerspot Inc and was an advisor at Creadits and Trick Studio.
On the investment side, Aethir has previously raised $9 million from a wide array of investors, including venture capitalists and family offices like Merit Circle , Animoca, Maelstrom Fund, and others.
Another noteworthy point is that Aethir also joined the Nvidia inception program in November 2023. This program helps start-ups during critical stages of development and allows them to get access to custom sets of ongoing benefits from the Nvidia ecosystem.
When we analyse different GPU cloud providers, both centralised and decentralised, we clearly see that Aethir is an outlier, offering its service at a fraction of the cost of others. Currently, the cost per A100 on Akash is $1.5 while it will be only $0.3 on Aethir. Clearly a game changer.
If we narrow our focus to the two biggest DePIN projects focusing on GPU computing, we also note that Aethir is in a strong position to become a leader in the sector:
When we look at the respective FDV of Render ($5 billion), and Akash ($2.2 billion) and given how Aethir is positioned, it is fair to say that Aethir has massive room to grow and represents a billions of dollar opportunity. Indeed, smart investors are realising this and Aethir is one of the hottest projects in the DePIN landscape.
$ATH is the token behind Aethir, it is central to the entire ecosystems:
The initial distribution of $ATH will be as follows.
There is a strong case to be made that the best way to capture this opportunity is to become an early node validator of the network.
In addition to contributing to the decentralisation of cloud computing and supporting the build up of a decentralised ecosystem for the largest collection of decentralised AI computational power, a node operator earns rewards in $ATH tokens. The three major forms of revenue for Node Operators are:
In total, there will be 100,000 nodes available and Aethir plans to deliver 15% of its token supply to them over a vesting period of four years.
To reward node operators, $veATH tokens are earned as checker work is done. Here is the expected vesting schedule of $veATH token:
After being unlocked, those $veATH tokens can then be claimed by users and redeemed into $ATH. When claiming, users need to choose a lock period on the token. Six months is the minimum lock up required to avoid a penalty fee. This process is done intentionally to make sure that incentives are aligned from all network participants.
To estimate the potential return on investment of node operators, it is important to understand that the reward for each node is influenced by the portion of nodes that is effectively running and performs correct validation. Therefore to estimate the potential return for this investment we have to simulate a range of different scenarios.
With that in mind, here is a potential return simulation over the whole period:
At the end of 2024, 20% of the total reward allocated to nodes will be released, and 10% will be unlocked and free to be used by the investors.
Using the same hypothetical initial FDV of $800 million (at 100% operating rate), this is the minimum FDV of $ATH required to breakeven in 2024 under different node operation effectiveness rate:
We can see that if the efficiency rate is between 60 and 75%, the $ATH token needs to reach an FDV between 4.8 and 6 billion USD, to break even already in 2024. Aethir’s biggest competitor, Render, currently sits at an FDV of $5.2 billions.
Our Alpha Deal has come to an end, but below you can still find all the main details concerning this opportunity.
https://www.aethir.com/blog-posts/aethir-joins-nvidia-inception-program
https://aethir.gitbook.io/aethir/ljvx8d8ee4ElPliP31K1/
https://ar5iv.labs.arxiv.org/html/2303.08774
https://www.binance.com/en/research/analysis/depin-an-emerging-narrative
https://blog.impossible.finance/aethir-research-report-2/
https://medium.com/@Aethir_/a-look-back-at-aethirs-remarkable-2023-00ab5286e6c5
https://middleware.io/blog/efficient-server-utilization/
https://mirror.xyz/sevenxventures.eth/Hx4AScWLZf4HrCl1IoumFTq1L20e3SzF-9XEkgWmrG4
https://www.precedenceresearch.com/graphic-processing-unit-market
https://www.theverge.com/23824200/ai-cloud-amazon-aws-adam-selipsky
Disclaimer: The information contained in or provided from or through this article (the "Article") is not intended to be and does not constitute financial advice, trading advice, or any other type of advice, and should not be interpreted or understood as any form of promotion, recommendation, inducement, offer or invitation to (i) buy or sell any product, (ii) carry out transactions, or (iii) engage in any other legal transaction. This article should be considered as marketing material and not as the result of financial research/independent investments.
Neither SBorg SA nor its affiliates (“Entities”) make any representation or warranty or guarantee as to the completeness, accuracy, timeliness or suitability of any information contained within any part of the Article, nor to it being free from error. The Entities reserve the right to change any information contained in this Article without restriction or notice. The Entities do not accept any liability (whether in contract, tort or otherwise howsoever and whether or not they have been negligent) for any loss or damage (including, without limitation, loss of profit), which may arise directly or indirectly from use of or reliance on such information and/or from the Article.