
AI Agents Struggle with Freelance Work
News Summary
A recent study conducted by researchers at Scale AI and the Center for AI Safety (CAIS) reveals that artificial intelligence agents are currently ineffective at performing online freelance work. Despite significant advancements in AI capabilities, the study found that leading AI models could only complete less than 3 percent of simulated tasks, earning a fraction of the potential income. The research tested several AI tools, including Manus from a Chinese startup, Grok from xAI, Claude from Anthropic, ChatGPT from OpenAI, and Gemini from Google. These AI agents struggled with tasks such as graphic design, video editing, game development, and administrative chores due to their inability to use different tools, perform complex tasks, and learn from experiences. This study challenges the notion that AI will soon replace many human jobs, especially in freelance and creative fields.




