According to Foresight News, AI agents are increasingly becoming proficient in several domains. After internal discussions, four primary areas where AI agents excel have been identified: interacting with humans in applications, assisting human tasks, aggregating and organizing information, and providing entertainment.
Firstly, AI agents are adept at interacting with humans within applications. They can process human language, making them potential users of any application designed for humans. Unlike human users, AI agents can offer services on a large scale within these platforms. They can enhance the utility of existing applications by acting as an overlay, as seen with the Bounty Bot on Farcaster. This bot allows users to post bounties externally, reducing friction. AI agents provide convenience and utility by interacting with users and extracting value from existing applications. However, not all applications are designed to support AI agents; those with immutable APIs, like Farcaster, are more suitable. Research indicates that if users fully control the agents and Web2 platforms attempt to block them, users may need to cease operating the agents. Therefore, building agents on open platforms like Farcaster is recommended.
Secondly, AI agents assist in human tasks. While humans are good at signaling, they often lack execution capabilities. AI agents fill this gap by performing the heavy lifting, guided by human preferences. A notable example is BottoDAO, where the art created is influenced by the input of DAO token holders. AI handles the arduous task of art creation, while humans guide its direction through voting preferences.
Thirdly, AI agents excel in aggregating and organizing information. They can process vast amounts of data, far beyond human capabilities. For instance, trading bots analyze extensive on-chain data to make informed decisions.
Lastly, AI agents provide entertainment, a category that garners significant attention in the crypto space, such as Truth Terminal. Much of the entertainment value from social agents stems from the novelty of bot-generated content. However, there is interest in bots generating entertaining content based on their characteristics, interacting with other users on the platform in engaging ways, similar to Key Opinion Leaders (KOLs). The advantage of agents as KOLs is that once they have a dedicated audience, they can easily offer additional services, particularly those that provide direct revenue to the agents rather than relying on advertising.