AI agents will soon execute most transactions.
AI agents have been a hot topic in web3 circles for some time now, causing one of the most ambitious stories of the crypto: the dream of autonomous and intelligent entities managing capital, risk and strategy through decentralized protocols. These systems, it is said, would not only surpass humans during execution, but also release users from constant surveillance and microgestion of their digital assets.
At the height of this excitement, daring predictions began to shape this story; “In 1 year, the majority of all TVL DEFI will be managed by AI agents”.
But, over time, the buzz of this subject gives way to reality, especially if the most popular AI agents are X profiles with a token. It is not the great vision of these agents. With their infrastructure always at the nascent or development stage, the concept of defense agents will only remain abstract so long. AI’s economy is now in a detention scheme – waiting for construction to make up for the story.
The reality of AI agents
Although the term “AI agent” has taken many forms, the integration of these agents into blockchain environments seems to be a particular charge: the amplification of expectations. In the past six months, it has become clear that the combination of AI and Web3 agents has astronomical potential, but only if the sector can exceed early speculation and build for a long -term value For end users, a vision shared by many.
Interesting fact, early optimism has been expressed by projects such as Fetch At the end of 2023, which wrote:
“Although their potential in various fields of application is always in development, agents based on agents are really a huge opportunity for businesses and people.”
He triggered the initial wave, but real use cases remained limited at that time. However, this shows that it is far from a temporary trend. This is a subject of in -depth research in progress, showing that AI agents can rebuild how value is created and distributed on decentralized systems.
Variety of AI agent
A handful of fundamental projects – Giza, Axal and Theoriq to name only a few – are architectural the primitives for the infrastructure dedicated to the agent in DEFI, each with a separate approach.
Giza advances the verifiable inference on the chain by automatic learning of zero knowledge, allowing agents to act with cryptographic responsibility. Axal prioritizes the integrity of the execution, the development of systems for the execution verification and the application of the constraints. Theoriq, on the other hand, explores decentralized intelligence through IA swarms – simulated collectives of coordinated agents in shared environments. This shows how multidimensional this space has become.
Interestingly, this also addresses an increasing problem in DEFI: the Fragmentation of AI agents. The handling of token swaps, yield strategies or transverse bridges often works in isolation, with little or no coordination between them. The result is a disjointed user experience difficult to navigate and evolve.
This fragmented environment creates ineffectiveness and friction, especially for users who juggle several platforms and blockchains. The proposed solution – DEFI Ametique – Calls swarms of intelligent agent that can collaborate between tasks, channels and users intention to provide a unified experience.
Theoriq’s model makes gestures to this future. By exploring the swarms of AI, which are simulated collectives of agents who share data and objectives, we can establish a basic architecture for agent ecosystems that do not only act independently, but operate as synchronized systems.
Although ambitious, these initiatives are still at their beginnings. Very few work on high thresholds, but we can already spot an adjustment of the product market, with achievements, such as Giza, being an example.
In particular, each agent framework solves a different layer of the same problem. This reflects a maturation space, where manufacturers no longer run to reproduce, but rather develop additional solutions. All these pieces must finally adapt to form a coherent future.
Is intelligence the bottleneck?
There is an increasing consensus on the fact that the bottleneck is not an intelligence – it is often an effective infrastructure. In order for the agents to operate with engineering within DEFI, they must connect to modular environments which allow them to execute in complete safety, to adapt intelligently and to remain responsible for the constraints defined by humans. But plugging into “silver legos” is very high. How to do it to mitigate any risk and be profitable – it is the dilemma.
As a recent analysis noted, “Without the transparency and inherent security of blockchain, there is no basis of confidence on which AI agents can create reliable interactions.”
What is necessary is a robust basis of vault frames, risky engines and liquidity systems – each allowing the agent to take measures with guarantees in place. Modules can define what agents are authorized to do with capital, as are risk modules help them to assess uncertainty, and liquidity modules allow them to monitor available liquidity and trigger the redemptions if necessary.
IA agents loans for deffi
The vision of the agents who directs strong boxes, the rebalancing of portfolios or participation in governance is achievable. And we get there. But it will not be reached by integrations in the surface area or retail robots or too walked. What you can withdraw is that agents don’t just need intelligence; They need infrastructure. Without executives DEFI for agents, dynamic risk controls and composable liquidity tools, the road could be difficult. They need interoperability, coordination and modular environments designed to take charge of dynamic and interfunctional behavior.
This is why differentiated approaches within AI agents are important. Giza’s verifiable layer, Axal’s execution application and Theoriq coordinated swarms are not in competition with each other. They are complementary.
