Fetch AI, a company founded by former DeepMind founding investor Humayun Sheikh, has introduced three interconnected products designed to enable trust, coordination, and interoperability in large-scale AI agent ecosystems. The announcement includes the release of ASI:One, a personal AI orchestration platform; Fetch Business, a verification and brand agent discovery portal; and Agentverse, an open directory hosting over two million AI agents.
Together, these offerings position Fetch AI as a foundational infrastructure provider for the emerging “Agentic Web,” where AI agents representing consumers and brands collaborate to execute complex tasks rather than simply providing recommendations.
Founding Vision Rooted in DeepMind Experience
Founded in 2017 by Humayun Sheikh, an early investor and one of the first five members of DeepMind, Fetch AI leverages insights from advanced machine learning research to build autonomous software agent infrastructure. Sheikh’s early backing of DeepMind and his experience with its acquisition by Google informed Fetch’s focus on agentic systems—AI networks capable of coordinated action across organizational boundaries.
Operating with a team of around 70 professionals across Cambridge and Menlo Park, and having raised approximately $60 million, Fetch AI has refined its model through interactions with over one million users. This data-driven approach shaped the design and functionality of its newly launched products.
ASI:One—A Platform for Multi-Agent Coordination
Central to the product launch is ASI:One, a language model interface built specifically to orchestrate multiple AI agents working in tandem. Unlike conventional large language models (LLMs) that respond to isolated queries, ASI:One functions as an intelligence layer managing context sharing, task routing, and user preference modeling.
The platform maintains user-specific data such as preferred airlines, dietary restrictions, budgets, loyalty memberships, and calendar availability. When users request complex services—like planning an entire trip including flights, accommodations, and dining—ASI:One delegates tasks to verified agents capable of providing actionable results like bookings and inventory, not just generic advice.
ASI:One’s architecture is modular, combining agentic and specialized expert models. It uses multiple user-owned knowledge graphs to securely store preferences and contextual data, ensuring personalized and consistent performance without mixing data across users or with Fetch-operated servers.
Currently in beta, ASI:One expects a wider release in early 2026. The company also offers ASI:One Mobile, which extends these orchestration capabilities to iOS and Android devices, enabling users to interact with agents on the go.
Fetch Business—Verified Identity and Brand Agent Control
To facilitate reliable consumer-to-business AI interactions, Fetch Business provides a verification and discovery portal where organizations can claim official Brand Agent handles, akin to domain registration or SSL certification for websites. This system protects consumers by ensuring they engage only with verified brand representatives, mitigating risks from counterfeit agents.
Fetch Business features low-code tools enabling businesses to create AI agents quickly and connect them to real-time APIs such as inventory or booking systems. Once verified, brand agents receive a persistent, portable identity across platforms integrated with Agentverse.
Verification leverages existing web trust infrastructure by requiring domain owners to insert cryptographic challenge codes into their websites, granting agents an authenticity badge similar to a “blue check.” Companies can start claiming their brand agents now via business.fetch.ai.
Agentverse—An Open, Platform-Agnostic Agent Directory
Agentverse serves as a comprehensive directory and cloud platform hosting millions of AI agents across industries including travel, retail, entertainment, and enterprise services. It provides metadata, capability descriptions, and routing logic that ASI:One uses to identify the most suitable agents for user tasks.
The directory is designed to be platform-agnostic, allowing agents created with any framework to join and interoperate securely. This universal discovery layer addresses a critical bottleneck where up to 90% of AI agents go unused due to lack of visibility.
Sheikh describes Agentverse as “the DNS for agents,” enabling near-instantaneous publication and discovery of agents independent of proprietary cloud environments. The platform also integrates payment systems through partners like Visa and Skyfire, supporting secure transactions with configurable user controls.
Implications for AI Ecosystems and Future Developments
Fetch AI’s launch arrives amid growing interest in shifting consumer AI from static interfaces to autonomous agents capable of executing complex multi-step workflows. Current challenges such as siloed architectures, limited interoperability, and weak verification hinder broader adoption.
The company’s integrated approach tackles these issues by establishing a cross-platform coordination layer, a robust identity system, and an open discovery service. By ensuring verified brand identities and enabling persistent user preference management, Fetch aims to build a reliable agent ecosystem where AI can act decisively and securely.
In addition, the inclusion of micropayment and digital transaction infrastructures supports Fetch’s vision of autonomous agents completing end-to-end financial tasks, positioning the company at the forefront of the agent economy’s infrastructure development.
Ultimately, Fetch AI’s ASI:One, Fetch Business, and Agentverse combine to form a comprehensive stack that could redefine how AI agents collaborate, discover each other, and perform real-world tasks, reflecting the company’s foundational belief that true intelligence requires the ability to act.
Fonte: ver artigo original
Related Articles
OpenAI’s Chief Scientist Highlights Slow AI Progress but Anticipates Major Breakthroughs
Meta Expands Solar Power Capacity by 100MW to Support New AI Data Center in South Carolina
U.S. Department of Energy Restructures, Prioritizes Fusion Over Renewables Amid Legal Concerns
California Launches New Tool Empowering Residents to Control Their Personal Data from Brokers
