5Eleven

play #1 for @PlayProphetX march madness contest round 4 🏀 texas tech +7.5 -112

Finding Moat in Web3 AI Agents Too many teams default to ChatGPT-like interfaces instead of embedding their agents into existing popular user funnels. Too many people underestimate the complexity of ChatGPT-like UI/UX. This leads to so many DeFAI abstraction layers operating with minimal users. While the DeFAI vision is great, most teams struggle at the "reasoning" layer—the platform requires users to prompt very specific prompts (instead of the platform trying to interpret & understand user intents like ChatGPT). This begs the question why we need this in the first place vs just sticking to existing DeFi protocols / tools that have far more optimal UI/UX. This is why projects that focus on embedding the interface into existing popular user funnels work a lot better i.e. @clankeronbase & @bankrbot using X as the go-to interface instead of trying to route users through their terminal / WebApp. ——————————— How to Find Your Moat in this market - Put more emphasis on user journey, ease-of-use, and accessibility over copying popular agentic interfaces. - Think about data & insights and how to best deliver them to users (especially if your product hinges on alpha). Best example in Web2 would be @casetext (acquired by Reuters for $650M) where the team built tools embedded into how lawyers draft and review filings, saving time and reducing errors. Their moat was data + workflow ownership. They understand the customers, put themselves right in front of the customers, and enhance their productivity. - Stop overcomplicating—merging buzzwords often muddies your PMF (some teams even call it AI-powered when it's just simple rule-based automations / algo... stop this pls, people aren't dumb). Instead, teams should be very clear on what their agents deliver, what problems they solve, and why it's better than existing non-AI (or other AI) solutions. - When it comes to financial use cases (autonomous DeFAI agents involving trading, yield farming, lending, etc), verifiability, confidentiality, and privacy need to be clearly addressed as there's a "Trust Deficit" for AI Agents i.e. market down bad, people are more skeptical of AI agents, nobody trusts agents enough to put millions of dollars for agents to manage (people would rather trust other people instead of trusting an agent). @gizatechxyz has done a great job highlighting risk management, verifiability, transparency, and confidentiality aspects of their infra in their documentation (not surprised seeing them achieve initial PMF w/ $1M TVL autonomously managed by agents). Checklist for Agent Teams • What data advantage do you have that others can’t easily replicate? • How are you embedding into critical workflows (popular user funnel), beyond being a co-pilot/chatbot? • How defensible is your distribution channel in this vertical? • How does your agent actually improve on existing DeFi protocols / tools in the market? • Is there a transparent audit trail for every AI-driven on-chain action? Can users “see inside” the agent’s logic? For investors, back teams that can answer these questions with confidence. Look for teams that don’t just follow the agentic crowd—but carve out their own lane by owning the full workflow.