Manus probably isn’t China’s second ‘DeepSeek moment’

Manus AI, despite its hype, has performance issues but is in early access.

: Manus, an autonomous AI platform, garners unexpected hype despite performance limitations revealed by early adopters like Alexander Doria. The Butterfly Effect uses existing AI models, including Claude and Qwen, for Manus' functionalities but its capabilities have been questioned. Media and social platforms inaccurately compare Manus to DeepSeek, which maintains a different development and distribution strategy. Manus, while still in early testing, highlights potential in AI but also the gap between promise and current performance.

Manus, an 'agentic' AI platform developed by The Butterfly Effect, has ignited widespread excitement despite running into several performance issues. It combines existing AI models like Anthropic's Claude and Alibaba's Qwen to manage tasks such as report drafting and financial analysis. Yichao 'Peak' Ji, a research lead for Manus, boasted that Manus surpasses tools like OpenAI's deep research on GAIA, a benchmark for AI assistants. Despite this, early users like Alexander Doria reported errors and loops when navigating Manus, questioning if the hype matched the actual performance.

The initial attention towards Manus is partly attributed to its limited accessibility, causing invite codes to sell for high prices on platforms like Xianyu. The Chinese media has been enthusiastic, encouraging perceptions of Manus as a major technological breakthrough. Publications such as QQ News labeled it as a paragon of domestic innovation, contrasting it with DeepSeek, another Chinese AI company with different operational models. Despite initial confusion, Yichao 'Peak' Ji clarified that some viral content about Manus represents more aspirational capabilities than concrete functions.

Acknowledging the limitations, The Butterfly Effect emphasizes their intent to use the ongoing beta phase to 'stress-test' Manus and adjust the platform based on user feedback. Manus currently lacks capability in areas like booking flights or ordering food seamlessly, evoking user frustration, including failed tasks of creating a Naruto-themed game. Despite continuous performance refinements, the AI is prone to factual mistakes and incomplete task execution, underscoring limitations that accompany its early stage of development.

Fervor around Manus reveals a broader trend of AI interest propelling platforms forward on social media and among influencers, sometimes irrespective of actual technological readiness. The excitement generated is reminiscent of how public anticipation often outpaces technical developments, creating pressure around the AI initiatives that are still refining themselves. It's within this context that tech companies like The Butterfly Effect are navigating the fine line between maintaining engagement and setting realistic user expectations.

While Manus stands as an exemplar of potential future AI applications, questions remain about its readiness and legitimacy as a current technological feat. Comparisons to successful predecessors or contemporaries often involve inflated perceptions and skewed realities, amplifying challenges that come with disseminating cutting-edge AI products. The Butterfly Effect aims to address these through ongoing development and open communication with its evolving user base.

Sources: TechCrunch, QQ News, Xianyu

AItechnologyinnovation