Crypto venture capitalists indicate that while decentralized AI is currently experiencing a downturn, genuine opportunities are starting to surface.

14

The objective is to transition from overstated GPU marketplaces and expansive AI model alternatives to tailored, full-stack solutions.

Key information:

  • The intersection of crypto and AI is moving into a more subdued, selective period with a focus on utility-centric applications, as stated by crypto venture capitalists.
  • The aim is to transition from overstated GPU marketplaces and extensive AI model alternatives to tailored, full-stack solutions.
  • Entrepreneurs need to provide more than just a superficial layer over existing AI models such as ChatGPT.

The convergence of cryptocurrency and artificial intelligence (AI) has entered a more tranquil, discerning phase, as noted by two notable venture capitalists.

Anand Iyer from Canonical Crypto and Kelvin Koh from Spartan Group characterized the present environment as a post-hype phase for decentralized AI protocols, with investments and talent reallocating towards more concentrated, utility-oriented applications during Consensus Hong Kong 2026.

STORY CONTINUES BELOWDon't miss another story.Subscribe to the Crypto Daybook Americas Newsletter today. See all newslettersSign me up

“I believe we are currently in a downturn,” remarked Iyer, whose firm in San Francisco supports early-stage infrastructure and applications based on decentralized networks. “We experienced a period of excess. Now we need to identify where the real value lies.”

Both Iyer and Koh expressed concerns regarding what they perceive as excessive investment in GPU marketplaces and efforts to create decentralized alternatives to significant AI models like those from OpenAI or Anthropic. The capital necessary, Koh pointed out, is “vastly different” from what is accessible in cryptocurrency.

Instead, they envision opportunities in purpose-driven, full-stack solutions, tools that address specific issues and develop down to the model, computational, and data layers.

Iyer highlighted startups bypassing costly SaaS solutions and leveraging AI to construct custom internal systems in a matter of days. “Speculation will no longer drive product development,” he stated. “User needs must come first.”

Both investors underscored the necessity of proprietary data, regulatory benefits, or market entry advantages as new types of competitive barriers.

For founders aiming to secure funding, Koh provided straightforward guidance: “A year ago, having a wrapper around ChatGPT was sufficient. That is no longer the case.”