Chinese artificial intelligence company DeepSeek is testing a new AI agent that can perform multiple tasks with limited human involvement, marking another development in the global race to build more independent AI systems.
The new system aims to reduce the need for users to supervise several AI agents simultaneously. It is designed to coordinate different activities, track progress and manage complex workflows more independently.
DeepSeek said recent advances in artificial intelligence have made it possible to develop agents capable of handling a broad range of everyday and professional tasks.
These include managing calendars, drafting professional emails, preparing technical reports and developing software.
DeepSeek Enters Growing AI Agent Competition
The project comes as Chinese AI developers compete with major US technology companies to build more advanced coding assistants and autonomous AI tools.
Anthropic’s Claude Code and OpenAI’s Codex have demonstrated capabilities in handling complex software development tasks, increasing competition in the rapidly evolving AI market.
Pakistan Extend Unbeaten Home Run to 10 Consecutive Cricket Series
DeepSeek’s new agent is intended to go beyond individual tasks by coordinating multiple activities and managing workflows with less direct human involvement.
However, some industry experts believe the company is moving in the right direction but question whether the new system offers significant improvements over existing AI tools.
DeepSeek Releases Open-Source Prototype
DeepSeek has also acknowledged concerns surrounding data privacy and confidentiality, issues that remain important as AI agents gain access to personal information and professional workflows.
The company recently released an open-source prototype of its AI agent on GitHub, allowing developers to examine and work with the early version.
However, DeepSeek clarified that the release represents an initial proof of concept rather than a fully developed commercial product.
The project highlights the growing shift towards AI systems that can plan, coordinate and execute multiple tasks with limited supervision. Whether DeepSeek can turn its prototype into a reliable product capable of competing with established tools remains to be seen.
