autorenew
Agents Unleashed Podcast Episode 24: Season One Recap on AI Agents and Crypto Future

Agents Unleashed Podcast Episode 24: Season One Recap on AI Agents and Crypto Future

If you've been following the wild world of crypto and AI, you know things are moving fast. The Agents Unleashed Podcast, sponsored by Olas, just wrapped up its first season with Episode 24—a solid recap that ties together six months of deep dives into agentic AI. Hosted by Thomas Maybrier, a producer at Valory (one of the core contributors to Olas), this episode looks back at key insights, lingering questions, and why AI agents are poised to shake up blockchain, including the meme token scene.

Agents Unleashed Podcast Episode 24: Season One Recap sponsored by Olas

Key Highlights from the Recap

The episode kicks off by revisiting early conversations, like the chat with David Minarsch, co-founder and CEO of Valory. He breaks down why people are still underestimating AI agents in crypto. Think about it: just like "not your keys, not your coins" became a mantra for crypto sovereignty, Minarsch flips it to "not your weights, not your brain" for AI. In simple terms, if you don't own the AI model (its "weights" are the learned parameters), you're at the mercy of whoever controls it. This hits home for meme token enthusiasts—imagine owning an AI agent that autonomously trades or predicts meme pumps without relying on centralized platforms that could shut you down.

Olas steps in here as a game-changer. It's a platform where anyone can own and monetize AI agents, powered by the OLAS token. This means you could build or buy agents that handle everything from market analysis to community engagement, all on-chain for transparency and security.

Tradeoffs in Computer-Use Agents

Thomas also demos a "computer-use" agent from Smol Agents, which lets AI interact with your machine—like clicking through UIs to complete tasks. Sounds cool for automating meme token sniping or wallet management, right? But there's a catch: privacy. Running these agents often means sharing screen data with centralized large language models (LLMs), which could expose sensitive info. For blockchain practitioners, this underscores the need for decentralized alternatives, where agents run locally or on trusted networks to keep your data sovereign.

Measuring Model Loyalty

One of the standout segments features Oleg Golev from Sentient AGI discussing "model loyalty." They spotlight the Dobby model—an experimental LLM that's "unhinged" (think no filters, swears included) but still performs on benchmarks without losing safety rails. It's tuned to value open-source and decentralization, aligning perfectly with crypto ethos. Why does this matter for meme tokens? Loyal AI models could be fine-tuned to prioritize community values, like pumping a token based on viral trends while avoiding harmful actions. It's a prototype, but it shows how we can embed pro-crypto biases into AI without tanking performance.

What Really Makes an AI Agent?

Thomas tackles the basics: What is an AI agent, anyway? In his wrap-up, he defines it as a software system using AI (like LLMs) to achieve goals autonomously, handling real-world unpredictability solo or in teams. For Olas, the winners will be transparent, collaborative, modular, and user-owned—principles that could supercharge meme token projects. Picture multi-agent systems predicting market sentiment or automating airdrops, all verifiable on blockchain.

Ethereum's Next Decade

The episode closes with Tomasz Stańczak from the Ethereum Foundation envisioning the next 10 years. Ethereum isn't just about DeFi anymore; it's the "world ledger" for AI and robots. He talks replacing inefficient institutions with agentic systems where failures escalate to humans only when needed. For meme tokens, this means seamless integration: AI agents verifying ownership, governance, and trades on Ethereum, making meme ecosystems more robust and less prone to rugs or manipulation.

Why This Matters for Meme Tokens

At Meme Insider, we're all about demystifying meme tokens and their tech underpinnings. This podcast recap highlights how AI agents could be the next big thing for blockchain practitioners. From owning your AI to building decentralized trading bots, the fusion of agentic AI and crypto opens doors for innovative meme strategies. Whether you're a dev tinkering with OLAS or a trader eyeing AI-powered tools, episodes like this remind us: the future is agentic, and it's built on chains like Ethereum.

If you're into this, check out the full episode on YouTube or dive deeper into Olas Network. What's your take on AI agents in meme culture? Drop your thoughts in the comments!

You might be interested