Superhuman Improves Auto-Drafting with Mixed-Model AI, Targeting Inboxes
Large Language Model
AI assistant
Email drafting
Superhuman
Anthropic
OpenAI
4
Refinement Over Revelation
Media Hype
5/10
Real Impact
4/10
What is the Viqus Verdict?
We evaluate each news story based on its real impact versus its media hype to offer a clear and objective perspective.
AI Analysis:
Moderate hype focused on the 'next big thing' of AI assistants, but the actual innovation is an iterative improvement in model blending and tone capture, making it a useful, but not paradigm-shifting, workflow enhancement.
Article Summary
Superhuman has unveiled an advanced version of its auto-draft feature, addressing previous criticisms that AI-generated emails sounded overly robotic. The new system leverages a mixture of advanced models, including those from Anthropic and OpenAI, to create drafts that better capture the user's authentic tone and context. During beta testing, the service successfully drafted complex responses—ranging from confirming meeting times to politely declining pitches—with minimal manual editing. While the feature is not perfect, its ability to learn from usage patterns and provide multiple tonal variations makes it a compelling productivity tool. Superhuman plans to expand this capability into a broader, cross-platform assistant called Superhuman Go.Key Points
- The upgraded auto-draft feature uses a mixture of frontier models from OpenAI and Anthropic, significantly enhancing the quality and nuance of generated emails.
- The tool learns from user behavior and context, providing draft variations that match the user's typical tone and professional style.
- Superhuman plans to evolve this capability into a comprehensive, cross-platform assistant, Superhuman Go, extending AI assistance beyond the email inbox.