Mixus.ai is tackling one of AI’s most persistent problems—hallucinations—by reintroducing a critical component that modern AI systems often lack: human judgment. The startup’s approach of combining artificial intelligence with human expertise provides a safeguard against the embarrassing and potentially harmful errors that even advanced AI models frequently produce, offering a practical solution to a problem that has plagued enterprise AI adoption.
The big picture: Mixus.ai has created a hybrid AI system that routes AI-generated content through human experts before delivery, addressing the accuracy problems that continue to plague even the most advanced AI models.
- The platform allows users to not only receive AI-generated responses but also connects them with subject matter experts who can verify information and join the conversation.
- The company has expanded its offerings with “colleague in the loop” AI agents that can perform various tasks while requiring human review before any content is finalized or distributed.
How it works: Users can create AI agents on the mixus.ai platform without any coding knowledge, but these agents require human approval before completing actions.
- When a user makes an AI query, the system provides both an AI response and recommendations for people with relevant expertise who can be added to the conversation.
- The new “colleague in the loop” agents can generate social media posts, assist with recruitment, draft emails, and perform other tasks—but nothing is sent until trusted human colleagues review and approve the content.
By the numbers: Major AI models continue to struggle with accuracy issues, highlighting the need for human oversight.
- OpenAI’s o3 model hallucinated more than 50% of the time according to the company’s own reporting.
- Its o4-mini model performed even worse, with nearly 80% of responses containing incorrect information.
The business side: Mixus.ai has secured initial funding and appears to be pivoting toward enterprise customers.
- The startup raised $2.6 million in pre-seed funding to develop its human-in-the-loop AI solution.
- The company offers a free 14-day trial for individuals but has shifted its business model from B2C to targeting businesses.
What they’re saying: Mixus.ai co-founder Elliot Katz emphasizes that their approach maintains AI’s benefits while eliminating its greatest weakness.
- “By bringing colleagues into the loop, you get the full power of AI agents, the efficiency and the time savings, etc, without any of that downside risk of AI mistakes going undetected,” Katz explained.
- “The beauty of this is someone who’s never used AI, someone who doesn’t even know what an AI agent is, you can create and use agents on mixus,” he added, highlighting the platform’s accessibility.
Mixus AI Agent Tool Includes Human Intervention