TL;DR
Thinking Machines, led by Mira Murati, is developing ‘interaction models’ that allow AI to respond in real time across multiple modalities. The company plans a limited research preview soon, with broader release later this year.
Thinking Machines, the AI startup founded by former OpenAI CTO Mira Murati, announced it is developing ‘interaction models’ that enable AI systems to respond to users in real time across audio, video, and text modalities. This marks a significant step toward more natural and seamless human-AI collaboration, with a limited research preview planned in the coming months.
Thinking Machines explained that current AI models process inputs in a single thread and wait for users to finish speaking or typing before responding, limiting real-time interaction. Their new interaction models aim to address this bottleneck by allowing AI to continuously perceive and respond as inputs occur, enabling more fluid and dynamic exchanges. The company demonstrated examples such as real-time speech translation, monitoring for mentions of animals in stories, and providing feedback on physical posture.
Mira Murati founded Thinking Machines in February 2025 after leaving OpenAI. The company has experienced some key personnel departures, with staff moving to Meta and returning to OpenAI, but remains focused on advancing AI interaction capabilities. A limited research preview of these models is planned for the coming months, with a wider rollout expected later this year.
Why It Matters
This development could significantly impact how humans interact with AI, making these systems more intuitive and effective across various applications, from communication to real-time translation and health monitoring. If successful, it could also influence the future design of AI interfaces, emphasizing continuous, multi-modal perception and response.
Language Translator Device No WiFi Needed, 150+ Languages Translation in Real Time for Voice/Text/Photo/Offline, Upgraded AI Translator Support ChatGPT, HD Touch Screen for Business Learning Travel
【150+ Languages Online Translation】: Our language translator device supports more than 150 online voice translations with an accuracy…
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Background
Mira Murati, known for her work at OpenAI, launched Thinking Machines in early 2025 with a focus on creating more interactive AI systems. The company’s announcement comes amid broader industry efforts to improve AI responsiveness and multi-modal capabilities. Previous AI models have been limited by their sequential processing, which hampers real-time collaboration. Thinking Machines’ approach aims to overcome these limitations by enabling AI to process multiple streams of input simultaneously.
“Our interaction models will allow AI to engage with humans more naturally and effectively, breaking the bandwidth bottleneck that currently limits real-time collaboration.”
— Mira Murati, CEO of Thinking Machines
“We plan to open a limited research preview in the coming months to gather feedback and refine our interaction models before a wider release later this year.”
— Thinking Machines spokesperson
Multi-Modal User Interactions in Controlled Environments (Multimedia Systems and Applications, 34)
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
What Remains Unclear
It is not yet clear how widely adopted these interaction models will become or how they will perform in real-world, large-scale applications. Details about specific use cases, technical capabilities, and potential limitations are still emerging, and the company has not disclosed detailed technical specifications.
Straight Plus AI Powered Posture Corrector Trainer Device (Necklace Included) – Back Brace Smart Straightener Office Work Support – Back Postur Corrector Device – Posture Corrector for Women & Men
FIRST-USE NOTICE: The battery may be low upon delivery. Please FULLY CHARGE the device before first use.
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
What’s Next
Thinking Machines will launch a limited research preview in the coming months, inviting select partners to test the models. The company plans a broader release later this year, with ongoing development and refinement based on early feedback.
VEKJ Ai Translation Earbuds Real Time with Audio and Video Calls, 3-in-1 Buletooth 5.4 Translator Headphones with 165 Languages/8 Translation Modes, 42H Audifonos Traductores Inglés Español, Black
Breaking Language Barriers, Conversation in Sync: Powered by next-generation real-time simultaneous interpretation technology, this AI translation earbud delivers…
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
What are ‘interaction models’ in AI?
They are AI systems designed to process and respond to inputs in real time across multiple modalities, such as audio, video, and text, enabling more natural and dynamic interactions.
When will the public be able to try these models?
A limited research preview is planned in the coming months, with a wider release expected later this year.
How do these models differ from existing AI systems?
Current models process inputs sequentially and wait for completion before responding, whereas these new interaction models aim to process multiple streams simultaneously for real-time interaction.
What applications could benefit from this technology?
Potential applications include real-time translation, interactive storytelling, health monitoring, and more intuitive human-AI collaboration across various industries.
