Agents Are Not Enough
Chirag Shah, Ryen W. White
TL;DR
The paper addresses the problem that current agentic AI, despite its promise, fails to deliver scalable, trustworthy, and broadly applicable automation. It surveys historical eras of agents, identifies five core failures, and proposes a multi-faceted remediation plan that includes symbolic-ML integration, new architectures, better coordination, robust learning, and ethical design. It then introduces a concrete ecosystem around Agents, Sims, and Assistants, plus an agent store, to preserve user privacy, personalize interactions, and coordinate tasks across heterogeneous components. This work provides a practical blueprint for building sustainable, privacy-preserving agentic AI and informs designers, policymakers, and platform builders about the necessary components and governance for broad adoption.
Abstract
In the midst of the growing integration of Artificial Intelligence (AI) into various aspects of our lives, agents are experiencing a resurgence. These autonomous programs that act on behalf of humans are neither new nor exclusive to the mainstream AI movement. By exploring past incarnations of agents, we can understand what has been done previously, what worked, and more importantly, what did not pan out and why. This understanding lets us to examine what distinguishes the current focus on agents. While generative AI is appealing, this technology alone is insufficient to make new generations of agents more successful. To make the current wave of agents effective and sustainable, we envision an ecosystem that includes not only agents but also Sims, which represent user preferences and behaviors, as well as Assistants, which directly interact with the user and coordinate the execution of user tasks with the help of the agents.