Chinese artificial intelligence company DeepSeek is developing a system designed to increase the autonomy of AI coding assistants, allowing them to perform multiple tasks with reduced human intervention.
According to a paper published by the company's DeepSeek Harness team and reported by Singapore's South China Morning Post, the initiative aims to minimize the human effort required to coordinate and monitor multiple AI agents.
The research team noted that managing multiple active agents currently demands significant human attention, stating that while single-task limits depend on model capabilities, multitask limits depend on human attention. The paper highlighted that an individual can typically oversee only three to five active AI agent sessions simultaneously.
The newly proposed system aims to distribute tasks among various agents and coordinate their work autonomously. This setup would allow human users to focus primarily on reviewing final results and making decisions.
This development unfolds amid intense competition among global artificial intelligence developers, which includes prominent US-based companies such as OpenAI and Anthropic.
Previously, in October, data from blockchain analytics platform CoinGlass indicated that certain Chinese AI models had outperformed US counterparts in cryptocurrency trading experiments. During that test, DeepSeek led the experiment as the only AI model to generate a positive unrealized return of 9.1%, achieving the results at a fraction of the cost of its US rivals.
