hidden AI gem in the market I strongly believe there is a very asymmetrical investment opportunity in the AI sector - the sector that has retraced the most since its previous high in January. Typically, during a risk-on market, those that have been suppressed the most tend to bounce back the hardest. Previously, Cryptocurrency used to move in general as a whole, but as the industry matures, each move becomes more independent. Because we have too many tokens with extremely high inflation in the market that have been used solely for market purposes to bootstrap early users, unless there is a very strong narrative in one specific sector, we see significant discrepancies in price action. Hence, token selection has been the key part throughout the cycle. After a pure attention-focused meme-driven cycle, we are currently seeing tokens with actual revenue and use cases after the US regulation is getting eased + DAT vehicles coming into the market to onramp boomer retails to crypto Now the narrative is REVENUE. $HYPE and $PUMP have been two of the most standout products this cycle, and the definitive trait these projects shared is their outstanding revenue. HYPE generated over $106M in monthly revenue in August 2025 alone, annualizing to around $1.2 billion. PUMP generated $838M in lifetime revenue since early 2024, with annualized figures hitting $558M in fees. These aren't just speculative; they're cash flow beasts proving that real revenue is the ultimate cycle winner. But there's an asymmetrical opportunity in the AI/DePin space — $ATH. This decentralised GPU cloud platform has been quietly building a revenue engine comparable to the cycle's top performers, but in a sector starved for actual earnings. Revenue @Aethir’s annual recurring revenue (ARR) has skyrocketed to over $155 million, up from $12 million just a year prior. That's almost 13x their revenue in just a year, a feat that few companies have ever achieved in history. Their Q2 revenue hit $32.67 million, with Q3 projections at nearly $39 million, a 20% QoQ jump. That’s real capital costs being paid for its 430,000+ GPU containers across 94 countries, delivering over 955 million compute hours. What sets @AethirCloud apart is that it’s arguably one of the only projects in the AI/DePin narrative generating substantial, verifiable revenue flows. Comparing a basket of AI projects that can generate significant volume in the market (those listed on CB and Binance Spot), it becomes pretty obvious which projects are fundamentally stronger in comparison to others. With such a big portion of the marketcap being accumulated as rewards in the treasury, potential burning mechanisms could trigger very positive. $PUMP and $HYPE have been outerpforming with 23%, 8% Annualized Rev / MC with $1B+ MC. Imagine the impact of $ATH burns with 39% Annaulized Rev / MC. Although there are other projects like $RENDER (5 yrs old), the relative marketcap size of $2 B and the lack of robust revenue flow make them unattractive. $IO may be cheaper in price, but higher inflation and a much smaller revenue flow make $ATH look significantly better. Why has their revenue been so strong lately? AI is simply still an ongoing trend, and Aethir is able to utilise high-quality GPUs while keeping costs low through Checker Nodes. Checker Nodes are the ones behind making sure demand is met with the right GPU providers in an instant. @AethirCloud has been consistently building partnerships across various sectors, including AI and Gaming. Through these strategic partnerships, there is a constant (growing) demand for Aether GPUs. GPUs remain in high demand and continue to offer very competitive prices relative to their Web2 competitors. With a more crypto-friendly environment, perhaps institutions having a growing demand for decentralised GPUs, and Aether is indeed one of the most competitive projects out there. Think point 3 is something to look out for closely in the near future, as we are already seeing massive adoption from institutions on different forms of solutions, such as payment and storage in Web3. It is only a matter of time until institutions realize how much a discount these projects have relative to inefficient Web2 compute providing companies that don’t have proper flywheels. Sooner or later, Web2 will view these projects as “cheap”; thus, I suspect projects with strong fundamentals and demand, like $ATH, should easily start trading at a higher premium. Remember there will always be high demand for GPU. Supply ATH Staking Details Reported circ is around 12.2B (29% of total supply), which includes the 2.5B ATH (22% of CS) staked across 5 different pools, so real circ is around 9.6B ATH (~23% of total supply). There are also two notable wallets (holding 9.85%, 4.85% each) that have been moved and remain untouched. Still not publicly disclosed what those wallets are, but since none of them have moved, even with investor unlocks being distributed over the past few months, we can potentially assume it's controlled by the team. This potentially means that even more of the reported circ is actually locked - making the actual marketcap of ATH much lower, probably to something <20% of FDV. Demand Also, we can see a big demand and love from our fellow Koreans. Generally, during a risk-on market, you want to see sustained strong Upbit volume dominance on an alt. Comparing $ATH chart (left) to $WLD (right) (although WLD was only listed last week on Upbit) $ATH is sustaining strong market demand from Koreans even after the big spike on both price and volume. Vol / MC is also one of our favourite indicators, showing the risk appetite the market has for a token. Sustained strong Vol / MC means more attention, mindshare, and demand from market participants relative to its MC size. With a potentially smaller float, $ATH is looking good in multiple fundamental aspects. Revenue is the meta, and ATH is the no-brainer pick for the AI/DePin sector. It had the strongest bounce from the most sold off sector (AI) and is already breaking out from a multi-year downtrend with strong volume. I personally think this can easily be pitched to Web2 investors and gain institutional demand (not just CT) and continue showing relative strength MTF. Conclusion AI is a massive sector, especially in Web2. We are in the middle of FOMO from Web2 institutions wanting a piece of the Web3 world. It is a matter of time before they start paying attention to the efficiency and the relative discount decentralized compute has in the Web3 space. And as we have seen, Tradfi with tons of $ will only value fundamentals, demand, and the maturity of projects. Considering these points, $ATH is at a perfect spot and is already receiving big attention. And guys.. ticker is literally $ATH 😉