| Mobile Web

Search results for Thinking

AI & Enterprise

Mira Murati says Thinking Machines differs from other LLMs

Thinking Machines, an AI startup founded by former OpenAI CTO Mira Murati (미라 무라티), is seeking to differentiate itself from other large language models with a human-centered interaction model. Murati described a shift away from turn-based AI systems that stop perceiving context while generating answers. She outlined a time-based approach that continuously processes audio, text and video in 20-millisecond slices to capture nuances in human conversation. She also stressed “tools for thought” and human-in-the-loop collaboration.

AI & Enterprise

Context before models as agentic OS race intensifies

Competition is intensifying over platforms that optimise multiple AI models with internal corporate data so AI can better understand organisations and execute tasks. Enterprise software companies are treating such platforms, which help AI grasp corporate context and play an operating system role, as strategically important. Microsoft, Snowflake, Adobe and SAP have highlighted products and strategies focused on context layers, governance and services for AI agents, while interest also rises in PCs and chips running AI locally.

AI & Enterprise

ChatGPT, Claude stumble in attention test, raising questions for AGI

ChatGPT and Claude, among the latest large language models, performed worse than expected in the Stroop test, a psychology experiment used to gauge human attention and executive control, a study found. Researchers tested OpenAI\'s GPT-4o and Anthropic\'s Claude 3.5 Sonnet and saw performance fall sharply when word meaning conflicted with ink colour, with accuracy declining further as tasks increased. Follow-up tests on newer models showed limited improvement, the study said.