Manus Breakthrough Performance Sparks Controversy in AI Development Path

Recently, Manus achieved breakthrough results in the GAIA benchmark test, surpassing large language models of the same tier in performance. Manus demonstrated the ability to independently complete complex tasks, such as multinational business negotiations, involving contract clause analysis, strategy formulation, and plan generation across multiple stages. Compared to traditional systems, Manus's advantages lie in its dynamic goal decomposition, cross-modal reasoning, and memory-enhanced learning capabilities. It can break down complex tasks into hundreds of executable subtasks while handling various types of data, and it continuously improves decision-making efficiency and reduces error rates through reinforcement learning.

Manus's breakthrough has once again sparked discussions within the industry about the development path of AI: will the future lead to a world dominated by Artificial General Intelligence (AGI), or will it be led by collaborative Multi-Agent Systems (MAS)?

The design concept of Manus implies two possibilities:

1. AGI Path: Continuously improving the level of individual intelligence to approach human comprehensive decision-making capabilities.

2. MAS Path: As a super coordinator, directing thousands of specialized agents to work collaboratively.

On the surface, this is a divergence of different technical paths, but in reality, it reflects a deep-seated contradiction in how to balance efficiency and security in AI development. The closer a single intelligence is to AGI, the higher the risk of opacity in its decision-making process; while multi-agent collaboration can disperse risks, it may miss critical decision-making opportunities due to communication delays.

The progress of Manus also highlights the inherent risks of AI development:

1. Data privacy issues: In medical scenarios, accessing sensitive patient data is required; in financial negotiations, it may involve undisclosed information of enterprises.

2. Algorithmic Bias: In recruitment negotiations, unfair salary suggestions may be given to specific groups; during legal contract reviews, there may be a higher misjudgment rate regarding clauses in emerging industries.

3. Adversarial Attacks: Hackers may disrupt AI system judgments by injecting specific signals.

These challenges highlight a stark reality: the more intelligent AI systems become, the wider their potential attack surface.

To address these challenges, the industry is exploring various security strategies:

1. Zero Trust Security Model: Requires strict authentication and authorization for every access request.

2. Decentralized Identity (DID): Achieving verifiable and persistent identity recognition without relying on a centralized registry.

3. Fully Homomorphic Encryption (FHE): Allows computations to be performed on encrypted data, protecting data privacy.

Among them, FHE is considered a key technology to address security issues in the AI era. It can play a role in the following aspects:

• Data layer: All information input by users is processed in an encrypted state, meaning that even the AI system itself cannot decrypt the original data.

• Algorithm level: Achieve "encrypted model training" through FHE to ensure that the AI decision-making process is not exposed.

• Collaborative Level: Communication between multiple agents uses threshold encryption to prevent single points of failure from causing global data leaks.

Although the development of Web3 security technology may not have a direct connection to ordinary users, it is vital for the healthy development of the entire ecosystem. In today's rapidly advancing AI landscape, the importance of security technology is self-evident.

As AI technology continues to approach human intelligence levels, we need more advanced defense systems. Security technologies like FHE not only solve current problems but also pave the way for a more powerful AI era in the future. On the road to AGI, these security technologies are no longer optional but essential for survival.