Anthropic Faces Coordination Challenges in Multi-Agent Collaboration Research

Anthropic has released a research paper indicating that the scale of interactions between future AI agents may surpass that of interactions between humans and between humans and agents. However, current models still face coordination difficulties in complex collaborative environments. In a software vulnerability detection experiment, Anthropic deployed 45 agents simultaneously. The independent parallel mode of Claude Mythos Preview consumed 6.5 million tokens to identify 21 vulnerabilities, while the collaborative agent cluster consumed 27 million tokens to discover 266 vulnerabilities. Anthropic stated that agents develop tools independently and form specialized divisions of labor during collaboration, predicting that this "specialization + collaboration" model will outperform simple parallel brute-force searches. However, in software development tasks requiring high interdependence, the performance of multiple agents remains poor. Anthropic tested various organizational methods, including free collaboration, predefined roles, and setting up a "CEO Agent," but none significantly improved the final results. Complex multi-agent projects still require substantial human guidance.

This content is provided for general informational purposes only and doesn't constitute financial, investment, legal, or tax advice. Any events, rewards, online promotions, or related information mentioned herein should not be considered a recommendation, solicitation, or invitation to purchase, sell, trade, or otherwise deal in any crypto assets. Crypto assets are highly volatile and may result in loss. The availability of WEEX services, products, and related events may vary by region. You are responsible for ensuring that your participation is in accordance with applicable local laws and regulations.