Tendem: A Hybrid AI+Human Platform

2026-02-01

AI Summary

This work addresses the limited reliability of AI systems in complex or uncertain tasks by introducing Tendem, a collaborative platform that integrates autonomous AI agents—equipped with web browsing, tool invocation, and domain-specific reasoning capabilities—with human experts. The system delegates structured tasks to AI while enabling human intervention during model failures or when verification is required, supported by a comprehensive quality assurance mechanism throughout the workflow. Evaluated on 94 real-world tasks, Tendem significantly outperforms purely AI-driven or fully manual approaches in both output quality and turnaround time, at a cost comparable to human-only execution. Furthermore, the AI component achieves near state-of-the-art performance on third-party benchmarks, demonstrating a balanced integration of high quality, efficiency, and cost control.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Multiagent Systems: Teamwork
• Philosophy and Ethics of AI: Safety, Robustness & Trustworthiness

Application Category

• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Tendem is a hybrid system where AI handles structured, repeatable work and Human Experts step in when the models fail or to verify results. Each result undergoes a comprehensive quality review before delivery to the Client. To assess Tendem's performance, we conducted a series of in-house evaluations on 94 real-world tasks, comparing it with AI-only agents and human-only workflows carried out by Upwork freelancers. The results show that Tendem consistently delivers higher-quality outputs with faster turnaround times. At the same time, its operational costs remain comparable to human-only execution. On third-party agentic benchmarks, Tendem's AI Agent (operating autonomously, without human involvement) performs near state-of-the-art on web browsing and tool-use tasks while demonstrating strong results in frontier domain knowledge and reasoning.

Problem

• hybrid AI-human system
• task quality
• turnaround time
• operational cost
• real-world tasks

Innovation

• hybrid AI-human system
• quality review
• autonomous agent
• tool-use
• human-in-the-loop