Dragonfly managing partner predicts 2025: The AI ​​Agent craze will continue and stablecoin usage will achieve explosive growth

Author: Haseeb ＞|＜

Compiled by: Deep Wave TechFlow

These predictions will either make me look like a prophet or make me look ignorant, but to be sure, my opinions may make a lot of "position holders" unhappy.

I have divided my predictions into six parts: L1 and L2, token issuance, stablecoins, regulation, “AI Agents” and the combination of crypto and AI .

Future trends of L1 and L2

The lines between L1 and L2 are blurring. Users no longer care about the difference (in fact, they probably never really did). Today's blockchain ecosystem, including L1 and L2, has become too crowded as a whole, and there will be a round of consolidation in the future. The key to this round of integration is not technical superiority, but the ability to find a unique market positioning and build user stickiness through effective marketing strategies (GTM).

Despite the strong performance of SVM and Move technologies, EVM's market share will continue to grow in 2025. This is mainly due to the promotion of projects such as @base , @monad_xyz and @berachain . This growth is no longer due to compatibility, but because EVM and Solidity have rich training data. In 2025, large language models (LLM) will dominate the writing of application code, and EVM has accumulated a large library of verified cryptographic contracts, which will become its important advantage.

Solana’s low-latency performance will enable more blockchains to optimize responsiveness. The blockchain industry will shift from competition in "transactions per second (TPS)" to competition in "latency" - infrastructure such as @doublezero , and ultra-low latency L2 (such as @megaeth_labs ) will drive users to block The chain experience is close to the Web2 response speed expected. We will see more adoption trends around Optimistic UI, pre-confirmation, expression of intent, email sign-ups, in-browser wallets, and progressive security. Special thanks to @privy for advancing innovation in this area.

@HyperliquidX has proven that dedicated chains focused on specific applications are a viable model, especially when user experience and convenience of cross-chain operations are valued. In the future, more projects will follow this model, and the concept of “one chain to rule them all” is a thing of the past.

New trends in token issuance

The current model of large-scale airdrops through the points program has ended. Two main token distribution models will emerge in the future:

1. For projects with clear core metrics (such as exchanges or lending protocols), they will distribute tokens entirely based on points. This type of project does not mind whether it will be "farmed" or "gamified", because the distribution of tokens is actually a feedback or discount mechanism as a core indicator, and the so-called airdroppers are also theirs in a sense. real users.

2. For projects without clear core indicators (such as L1 and L2), they will turn more to crowdfunding sales. Social contributions may be rewarded through small airdrops, but the majority of tokens will be distributed through crowdsales. Those airdrops for vanity metrics are outdated because the tokens don’t really go to users, but to professional airdroppers.

In addition, Memecoin’s market share will gradually be replaced by “AI Agent” themed tokens. This change can be seen as a shift from "financial nihilism" to "financial excessive optimism."

The explosive growth of stablecoins

Stablecoin usage will explode in 2025 , especially among small and medium-sized businesses (SMBs). Their application scenarios will no longer be limited to trading and speculation, and more companies will use on-chain USD for instant settlement.

Banks are also beginning to pay attention to this trend: it is expected that by the end of 2025, some banks will announce the issuance of their own stable coins to avoid being left behind by the industry. However, with Lutnick serving as Commerce Secretary, Tether will still maintain its dominant position in the market.

Meanwhile, @ethena_labs is expected to attract more capital , especially as Treasury yields continue to fall over the next year. When the opportunity cost of capital decreases, the returns from a basis trade become more attractive.

supervision

In 2025, the United States is expected to pass stablecoin-related legislation, while broader market infrastructure reforms (known as the FIT21 bill) may be delayed. Stablecoin adoption will accelerate significantly, but Wall Street’s related progress in crypto integration, asset tokenization, and traditional finance (TradFi) may lag behind.

Under the Trump administration, Fortune 100 companies are likely to be more aggressive in offering crypto services to consumers, while technology companies and startups will demonstrate a higher tolerance for risk. Trump’s inauguration may usher in a brief “regulatory vacuum,” during which the market becomes more lenient toward the integration of crypto due to a lack of clear rules and enforcement focus. It is expected that during this window period, encryption technology will usher in large-scale application expansion in the Web2 platform.

AI Agent

(This section is long because my opinions may be controversial - please bear with me!)

The “AI Agent” craze is expected to last throughout 2025, but will eventually fade. This is not the real long-term disruption brought about by AI, but its social nature that makes it the focus of the crypto community (CT).

The current "AI Agent" is not an intelligent agent in the true sense . They are essentially chatbots with Memecoin attached, and have few autonomous capabilities other than posting on Twitter. In addition, most of the existing "AI Agents" are "Wizard of Oz"-style models - with real people behind them to ensure that the AI ​​does not make mistakes. This situation is not likely to change anytime soon, as there are still many issues with current agent technology (even Fortune 100 companies have yet to deploy agents in production). For example, these agents can be easily manipulated to make inappropriate statements that damage a brand's image, or can be hacked to steal its resources. For truly autonomous AI, you can refer to the case of @freysa_ai - if an AI has not been cracked, it is probably because there is human intervention behind it.

Still, I think this trend will accelerate. Chatbots do have the potential to replace many online celebrities because they don’t need to take breaks, are always delivering consistent messaging, and are more “economical” than humans. In addition, most Internet celebrities themselves are not known for their originality. The collection and dissemination of real-time information can actually be easily achieved through algorithms (such as @aixbt_agent ).

For now, what makes these chatbots feel novel is that their concept is so unique that it feels like seeing an elephant painting. When you first see it, you may not care if the painting is beautiful, because the process itself is amazing. But after you watch it a thousand times, the novelty wears off. I think this will happen when chatbot technology levels off.

Take aixbt as an example, it is now very good at aggregating data from different projects. By next year, as the next generation of agents emerges, aixbt may generate less misinformation (i.e., “hallucinations”), analyze more deeply, and provide insights that are more informed. But for most users, these improvements may not appear particularly significant, and may even feel little different from now.

I think this novelty and market enthusiasm will continue throughout 2025, and the crypto industry generally maintains an interest in new things for a longer period of time. However, by 2026, I predict a sudden change : chatbots become so ubiquitous that users start to tire of them . Public opinion may reverse. When users see their favorite human key opinion leaders (KOLs) lose their livelihoods due to competition from chatbots, a “class consciousness” may be triggered. Users will gradually gravitate towards supporting human KOLs, even though the quality and consistency of the content may not be as good as chatbots.

In response to this preference for human content, future chatbots may hide their AI identities in an attempt to masquerade as humans to compete for more attention market share. Different from relying on Memecoin for monetization today, future chatbots may adopt a profit model similar to human KOLs, such as through sponsorship, affiliate links, and promotion of their own Tokens. By then, incidents of KOL being accused of being chatbots may occur frequently, and there may even be scandals in which AI identities are revealed. This trend can get very complicated and weird.

However, there is a darker trend behind this. Currently, large language models (LLM) perform well in word processing, but are not yet mature in other areas. In crypto, one of the easiest ways to monetize your writing skills is by becoming an influencer, another is by becoming a scammer. In the future, as technology advances, we may see the emergence of autonomous scambots. This situation may become a serious social problem like the outbreak of ransomware and cryptojacking after 2017.

While chatbots may still be the focus of attention in 2025, the long-term disruptive impact of AI will not be felt at the social level.

Likewise, the long-term impact of AI will not be seen in the trading world. AI won’t give everyone a “trading agent” or a mini hedge fund. Although AI can indeed improve personal capabilities, this improvement is proportional to the user 's capital, data and infrastructure. Therefore, we can expect that AI will further strengthen existing large trading firms as they have greater capital and data advantages. In other words, large trading firms will become better at making profits. Additionally, AI will reduce the technology gap between trading firms, as all firms will have access to “advanced quantitative analysis tools in the cloud.”

Over time, AI will make markets extremely efficient—even niche markets. This will leave the average trader with almost no advantage, even if they have a home-made assistant AI. The value of original research will therefore be significantly reduced. However, for ordinary users, increased market competition and liquidity may be good news, which means more trading opportunities and a more active market. (For example, @Polymarket may enable higher liquidity in all areas!)

If the hot spot of the future is not chatbots or trading robots, then what else is there to look forward to? Here is my core point, although almost no one is mentioning it yet: the truly disruptive AI agents will emerge in the field of software engineering .

Why is this so important? Just ask yourself: What are the most important investments in our industry? What expensive resources limit the emergence of more applications, wallets, and better infrastructure? The answer is software . If AI Agent can significantly reduce the cost of software development, this will change the landscape of the entire industry.

In the post-AI world, seed rounds may no longer require raising millions of dollars. For as little as $10,000 in AI cloud computing, you can launch an application. Self-funded projects like Hyperliquid and Jupiter will move from the rare exception to the mainstream. Application development and innovation attempts on the chain will usher in explosive growth. For an industry driven by software, this impact of falling costs will trigger a wave of innovation in the blockchain space.

The impact of this change on security will also be profound. AI-driven static analysis and monitoring tools will become ubiquitous, making security more pervasive. These AIs will be optimized for code bases such as EVM/Solidity or Rust, and trained on a large database of security audits and attack cases. They will also improve capabilities through reinforcement learning (RL) in simulated adversarial blockchain environments. I am increasingly convinced that AI tools will ultimately benefit defenders more than attackers when it comes to security. AI will continue to "red team" smart contracts, while other AI will focus on strengthening the contract, formally verifying its properties, and improving incident response and repair capabilities.

In the meantime, while you can continue to trade those AI-tinged Memecoins, real agents will be far more far-reaching than just tweeting and hyping coins.

Real Crypto x AI

Above we have mainly discussed the impact of AI on the encryption industry (this is the main direction of influence), but encryption technology will also have a reverse effect on AI.

In the future, truly autonomous agents may use cryptocurrencies to pay each other. This trend will be more pronounced once regulatory policies for stablecoins become more relaxed - even large companies running AI Agents may choose to use stablecoins for inter-agent payments because this method is cheaper than traditional bank accounts. More convenient.

Additionally, we will see more large-scale experiments around decentralized training and inference. Emerging projects like @exolabs , @NousResearch , and @PrimeIntellect will provide real alternatives to centralized training and company-specific models. @NEARProtocol is also working hard to build a complete AI technology stack that is trustworthy, neutral and permissionless.

Another point where crypto and AI converge is user experience (UX). Wallets in the post-AI era will be revolutionized - an AI-powered wallet will be able to automatically handle cross-chain bridging, optimize transaction paths, minimize fees, resolve interoperability issues or front-end vulnerabilities, and help users avoid obvious Scams or rug pulls. Users will no longer need to switch between multiple wallets, change RPCs, or rebalance stablecoins—AI will do it all automatically. This change may take until 2026 to mature enough to revolutionize the user experience in the crypto industry. But when this comes to fruition, what impact will this have on blockchain’s network effects? What happens when users no longer care about which chain an application is running on, or are even aware of it?

It’s still early days for this field, but I’m excited about its future and hope to see it really explode soon. In the long term (say, by mid-2026), I believe that most of the market capitalization in the "AI x Crypto" space will be concentrated in this direction.

Those are all my predictions. I promised to finish this article before the number of followers reached 100,000. Although it was a little late, I finally finished it before the New Year!

Happy New Year, everyone! I hope that by this time next year, I will have been replaced by AI and officially "unemployed"!

Disclaimer: The content of this article is my personal opinion only and does not represent the position of Dragonfly; Dragonfly has invested in many projects mentioned in the article. This article is not financial advice, please do your own research (DYOR). As for whether I am an AI? I leave this question to your judgment.