Thinking Machines Just Announced More Human Like AI

2026-05-14 · Source: AI Supremacy · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Emerging Technologies & Innovation · Depth: Intermediate, quick

Summary

Thinking Machines Lab, an AI startup founded by former OpenAI CTO Mira Murati, has introduced "Interaction Models," a new category of AI systems designed to enable more human-like, real-time interactions. Unlike traditional AI models that process input sequentially, Interaction Models utilize a micro-turn architecture, processing data in 200ms chunks to allow for simultaneous input processing and response generation. The company's model, TML-Interaction-Small, reportedly responds in 0.40 seconds, matching the speed of natural human conversation and outperforming comparable models from OpenAI and Google. This approach aims to move beyond "turn-based" AI interactions, facilitating native collaboration across audio, video, and text without external scaffolding, thereby addressing the "collaboration bottleneck" in human-AI interfaces.

Key takeaway

For AI builders, developers, and power-users seeking more natural and efficient human-AI interfaces, Thinking Machines Lab's Interaction Models offer a significant advancement. Your current "turn-based" AI interactions can be replaced with full-duplex systems that process input and generate responses simultaneously, akin to human conversation. Consider integrating this "micro-turn architecture" to create more intuitive and real-time collaborative AI products, potentially enhancing user experience and expanding application possibilities.

Key insights

Thinking Machines Lab introduces "Interaction Models" for real-time, human-like AI collaboration via simultaneous input/output.

Principles

• Interactivity should scale with intelligence.
• AI interfaces should adapt to human interaction.

Method

Interaction Models employ a micro-turn architecture, processing data in 200ms chunks for full-duplex, simultaneous input and response generation across audio, video, and text.

In practice

• Achieve 0.40-second AI response times.
• Enable real-time audio/video/text collaboration.

Topics

• Thinking Machines Lab
• Interaction Models
• Full Duplex AI
• Real-time AI Interaction
• TML-Interaction-Small

Best for: Machine Learning Engineer, NLP Engineer, Entrepreneur, AI Engineer, AI Product Manager, Investor

Related on AIssential

• New Trends in AI Interaction Models — Thinking Machines Lab's Interaction Models enable full-duplex AI communication, allowing simultaneous listening and talking to eliminate conversational latency.
• [NOTICE] This feed is no longer maintained — Native interaction models enable real-time, multimodal human-AI collaboration by integrating interactivity directly into the AI's core architecture.
• Thinking Machines shows off preview of near-realtime AI voice and video conversation with new 'interaction models' — Thinking Machines' interaction models enable real-time, full-duplex human-AI communication by processing multimodal inputs and outputs simultaneously.
• Thinking Machines unveils AI that can interrupt users in real-time — Thinking Machines Lab developed a full-duplex AI model enabling real-time, interruptible human-AI conversation.
• Thinking Machines drops a new, highly responsive model designed for humanlike interactions in real time — New AI architecture enables full-duplex, real-time multimodal interaction by processing data in 200-millisecond micro-turns.
• Mira Murati steps back into the spotlight, carefully — Thinking Machines Lab is developing "interaction models" for real-time, continuous human-AI communication beyond turn-based prompts.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by AI Supremacy.