Blockchain expands into AI at ETHDenver 2026

Web3 and AI Adoption Takes Center Stage in Denver

-- Abdul Al Ali, Venture Investor at Decasonic

Introduction

There was a noticeable shift in the conversations, panels, and insights at ETHDenver this year. That shift is largely driven by the rapid emergence of AI and the agentic economy as a core design primitive within Web3 - with software agents, physical AI systems, and autonomous coordination increasingly shaping how value is created and transacted on-chain. As agents become first-class economic actors, institutions are beginning to assess how to enter this space at scale, recognizing that crypto-rails may become foundational infrastructure for AI (software or machine)-driven commerce. The pro-regulatory environment is further enabling deployment beyond experimentation, reinforcing the shift in Web3 entering a new phase primarily driven by valuable adoption. 

As investor-operators investing at the intersection of Web3 x AI, the resurgence of attention here feels less cyclical and more structural. In the near term, we are moving through a filtration cycle: AI is compressing differentiation, testing revenue quality, and exposing fragility across both software and token models. “Good enough” no longer clears the bar. AI is good at producing output, but the bar is the quality of output. Some of this reflects rapid progress at the model layer, but the stronger driver is accessibility, AI tools are now easier to deploy, iterate, and integrate, raising the baseline across the industry. For many, this is the signal of the experimentation age. 

At the same time, this filtration is setting up a dispersion cycle. As intelligence becomes cheaper and coordination becomes programmable, variance widens. Elite, AI-native teams operating on crypto rails can scale with structurally lower cost bases and aligned incentives. The role of software AI and physical AI in Web3 is no longer experimental; it is emerging as a first-class design space with a tremendous canvas of asymmetric opportunity. 

Most recently, that accessibility was reinforced by OpenClaw. I remember the first AI agent I created for Decasonic back in December 2024, and how natural it felt to meet users where they already are, across communication channels like Telegram, WhatsApp, and Discord. By February 2025, we moved from “agents as tools” to “agents with identity,” which led to our ‘Jobs’ agent (named after Steve Jobs). OpenClaw expands that design space further, enabling experimentation across individuals, enterprises, entrepreneurs, and institutions, and accelerating the question of what becomes possible when agents interact and move on-chain.

ETHDenver continues to provide a real-time window into how investors, operators, and users are thinking across Web3, AI, and the intersection. If you take only three takeaways from this article, the ones to remember are:

1. The agentic economy is no longer a thought experiment. This is driven by increased accessibility of tools (similar to OpenClaw) that reduce barriers to AI integration and usage.

   

   Event Name: OpenAGI Summit

2. Institutions are shaping product requirements. Participation is shifting, and there is more emphasis on real-world adoption and production-grade systems that are ready for deployment. 

   

   Event Name: TBV Demo Day

3. The frontier of Web3 x AI is expanding into physical systems (with energy coordination emerging as a bottleneck worth designing around). Energy is increasingly viewed as the limiting factor for scaling modern AI across both software and physical systems. Crypto plays an active role in addressing the coordination gap for energy and other tokenized assets that can accelerate the AI economy. 

Event Name: RoboCon 

Some Personal Highlights

1. The x402, ERC8004, and MetaMask teams presented potential upgrades to agentic payments, specifically through execution guardrails and policy layers. At Decasonic, we previously provided market-maps for x402 and Agentic Identity.

2. I also had the opportunity to participate as a panel judge at a tbv event and hear from a strong set of founders. Incredibly appreciative for the opportunity and the conversations.

3. Viewing the PrismaX (a Decasonic portfolio company) tele-operations competition and seeing in real-time the impact crypto can have on coordination within the physical AI market.

Below are my key six takeaways from the event:

The Agentic Economy Is Being Set in Real Time

The Agentic Economy is now an active point of discussion. The shift is primarily operational: more individuals are rapidly experimenting with AI agents as the barrier to entry continues to collapse. Frameworks like OpenClaw, ElizaOS, and Virtuals Protocol are making it easier for individuals to spin up agents that can execute tasks, coordinate, and transact on behalf of users. As accessibility improves, the design space at the intersection of Web3 x AI continues to expand.

This, in turn, increases the number of potential agentic economy participants. Users can spin up OpenClaw agents in under < 10 minutes and give them access to on-chain tools. We can see the early adoption curve reflected in the growth of core infrastructure primitives: ERC8004 registered-agents now exceed ~45.23k. x402 has seen ~75.41M transactions in the past 30D. In parallel, agentic-friendly infrastructure is accelerating across established crypto companies like UniSwap and deBridge, where “skills” and integrations (re: MCP) are emerging that allow agents to execute transactions more natively. Builders are increasingly positioning for a reality where AI , machines and software, becomes the dominant participant on on-chain rails.

This transition is a catalyst because it enables long-tail consumers, builders, small teams, and non-technical investors to leverage agents as persistent operators both on and off chain. This matters because markets are shaped by participant density. When the number of agents grows, the downstream effects compound:

1. More transaction volume becomes agent-driven.

2. More coordination shifts from human latency to machine execution.

3. More discoverability becomes necessary: agents need to be findable, addressable, and reputation-bearing.

Web3 provides one of the best avenues to enable agents to become persistent, economic actors. As agents evolve into first-class participants, crypto becomes the coordination layer that enables interoperability, persistence, and discoverability.

Agentic Payments Are Showing “Maturity”

The ownership of the execution workflow, and the emergence of a defined ‘control plane’ for AI agent transactions, is a clear sign of market maturity. It signals a transition from experimentation toward valuable adoption. Builders, ecosystems, and infrastructure providers are increasingly moving beyond demo-stage agents into production-grade systems. Real primitives are now enabling agentic payments at scale, including x402, the agentic/smart wallet direction from MetaMask, and execution guardrails and policy frameworks that constrain agentic transactions and interactions. The latter is the more meaningful signal of maturity.

Guardrails and policies are the unlock. They create a pathway toward both scale and economically valuable agents. Agents can operate as first-class economic actors, but their permissions, spending thresholds, counterparties, and execution logic can be explicitly constrained. This shift, from unconstrained experimentation to policy-driven execution, enables real-world deployment and materially accelerates the pathway toward enterprise adoption.

Enterprises are actively seeking control planes: interfaces that allow for the enforcement of permissions, limits, auditability, and policy-based execution for agents acting on their behalf. This layer is not just a security feature, it is a value layer. Ownership of this interface (whether embedded in wallets, middleware, or enterprise tooling) is likely to grow in strategic importance as agent participation on-chain increases. As more economic activity becomes machine-driven, the entities that define and manage the execution surface area will capture disproportionate strategic leverage.

Trust, Verifiability, and Truth as Core AI Infrastructure 

As agents proliferate, trust and verifiability move from abstract ideals to baseline requirements. In a world where AI capabilities continue to accelerate, the differentiator is the quality, reliability, and auditability of that output. As output quality improves and trust compounds, the delegation of meaningful human workloads to AI will continue to scale. Trust is the gating factor for adoption.

Across conversations with founders, several core design questions are now front and center. They primarily revolve around:

1. How do we verify AI outputs in a way that is machine-readable and composable?

2. What is the appropriate trust model when agents are granted access to APIs, wallets, and sensitive workflows?

3. How do we enforce constraints on agent behavior, and verify that those constraints were actually followed?

This is where crypto’s strengths map cleanly onto the problem set: verification, auditability, composable identity, and programmable constraints. As agents begin to transact, coordinate, and act on behalf of users and enterprises, the infrastructure for proving behavior becomes just as important as the behavior itself.

This naturally led to an important observation: both prediction markets and stablecoins appear to have crossed a tipping point in adoption. Prediction markets, in particular, are increasingly viewed as mechanisms for credible, incentive-aligned truth discovery. Their importance compounds when viewed through an agentic lens.

Agents require data they can rely on. Prediction markets (like Opinion Labs) can function as a probabilistic source-of-truth layer that agents can query, price, and act against, especially in environments where social information is noisy, fragmented, or adversarial. In this context, high-quality data inputs can increasingly emerge from incentive-aligned markets, reinforcing the role of crypto-native primitives as foundational infrastructure for AI. 

Institutional Adoption Is Accelerating

One of the strongest recurring signals was the continued acceleration of institutional adoption, particularly from family offices and other forms of durable, “sticky” capital. What stood out even more was the widening disconnect between retail sentiment and institutional sentiment. Retail narratives often change (and rapidly so) with price action. Institutional narratives increasingly revolve around infrastructure readiness, regulatory posture, repeatable deployment pathways, and the acceleration of real, valuable adoption.

This rapid institutionalization is actively shaping what gets built. The underwriting lens is shifting toward products with a clear pathway to durable usage and monetization. The teams that unlock meaningful capital access and enterprise distribution are the ones building production-grade systems that can be deployed repeatedly and generate recurring usage that translates into revenue.

This shift is also contributing to a decoupling within the broader crypto market. Stablecoins sit at the center of this institutionalization arc, with growing sentiment around the decoupling of stablecoins from the broader market, primarily driven by accelerating adoption across institutions, banks, and TradFi-adjacent rails. Stablecoins are being treated less as part of the ‘crypto sector,’ and more as a global settlement primitive that continues compounding regardless of broader market cycles. The same, in recent weeks, can be set about the relative decoupling of Bitcoin as a tradeable asset (and primarily in the form of a narrative) with the wider crypto market. 

Crypto as the Coordination Layer for Physical AI

One of the core constraints in the advancement of Physical AI is the energy bottleneck. Addressing this bottleneck is essential to enable full-scale deployment of Physical AI systems and to sustain the accelerating demand for software-driven AI. Energy, and more specifically programmable energy, sits at the intersection of two accelerating trends: (1) the tokenization of commodities and the movement of real-world assets on-chain, and (2) the expansion of crypto into physical AI systems. Energy is increasingly recognized as a primary bottleneck to AI advancement in the U.S., and in conversations with investor-partners there is a growing emphasis on identifying opportunities that can directly alleviate this constraint and unlock the next phase of AI scaling.

As AI capabilities continue to advance and expand from purely software-based systems into physical environments, the energy constraint becomes more visible. This pressure will manifest across physical systems that require compute, sensing, actuation, and persistent execution. In this context, crypto functions as a coordination layer for physical AI resources. It enables the pooling, ownership, and programmable allocation of scarce inputs, aligning resource owners with beneficiaries. Through market-based allocation, incentive design, and programmable access controls, crypto provides a framework to address energy constraints with a lens on scale and capital-efficiency. 

A tangible example of this coordination layer was expressed at PrismaX’s RoboCon event. The team hosted a live teleoperation competition, where coordination extended to human resources, specifically teleoperators controlling physical systems in real time. In this case, crypto can serve as the coordination layer that identifies top-performing operators, aligns incentives, and rewards them transparently. The same mechanism that coordinates capital and compute can coordinate human and machine collaboration.

If you zoom out, the pattern is consistent:

1. Physical AI requires data, coordination, and infrastructure.

2. Energy is a structural constraint that will increasingly determine what scales.

Crypto is uniquely suited to coordinate scarce resources across actors, geographies, and systems. Its ability to assign ownership, delegate access, and align incentives across contributors positions it as a foundational layer for scaling Physical AI and in addressing the resource bottleneck heads-on. 

Token Timing, PMF, and Incentive Alignment

For founders aiming to learn more about the timing for token launches, this learning is for you. Another recurring signal was the increasing clarity founders are drawing between tokens as a product and the core company product. This distinction reflects real market maturation and was closely tied to a metric that surfaced repeatedly in conversations: time to revenue. As the space institutionalizes, valuation dynamics are adjusting accordingly. Crypto is moving beyond experimentation, vision, and ideals into deployment. As a result, both valuations and the frameworks used to underwrite companies are evolving.

The market timing for token launches is shifting. In most discussions, tokens are increasingly viewed as a post-PMF accelerant rather than a mechanism for jump-starting PMF. The sequencing matters. Product-market fit validates demand, use-case durability, and monetization pathways. In a world increasingly shaped by AI-driven systems and on-chain coordination, durability also means ensuring your product and token can serve both human users and machine-native participants - including agents, and eventually physical AI systems. A token, when introduced at the right stage, can amplify growth, deepen network effects, and align incentives across users, developers, operators, and autonomous actors.

Time-to-revenue acts as a forcing function. It signals whether a project can move toward self-sufficiency regardless of token timing, liquidity cycles, or broader market conditions. Teams are placing greater emphasis on accelerating revenue pipelines and building valuable adoption before introducing token complexity. AI meaningfully compresses time-to-revenue by increasing capital efficiency, accelerating experimentation around PMF, and enabling more sophisticated simulation and optimization before large-scale deployment. Token launches are increasingly treated as a strategic lever, deployed once durability is proven, to accelerate adoption and align ownership with real usage.

Conclusion

ETH Denver reinforced a clear structural shift: crypto is entering a phase where adoption and value is accruing to real-world deployment, and that deployment is increasingly AI-driven. Conversations have shifted beyond experimentation towards deployment of production systems, where AI can transact, coordinate, and execute. This includes crypto being actively assessed as the core layer enabling the agentic economy to operate at scale. 

At Decasonic, we partner with founders driving this intersection forward. We partner with teams and founders to define market category leaders. If you’re building at the frontier of Web3 x AI, reach out, let’s help transform your vision into reality. 

The content of these blog posts is strictly for informational and educational purposes and is not intended as investment advice, or as a recommendation or solicitation to buy or sell any asset. Nothing herein should be considered legal or tax advice. You should consult your own professional advisor before making any financial decision. Decasonic makes no warranties regarding the accuracy, completeness, or reliability of the content in these blog posts. The opinions expressed are those of the authors and do not necessarily reflect the views of Decasonic. Decasonic disclaims liability for any errors or omissions in these blog posts and for any actions taken based on the information provided.