Griffin, the first Human Interaction Model to pass video Turing Test it's already #1 on NVIDIA's benchmark for full-duplex AI video - 44% of people thought it was a real person while other systems are at ~3%
Summary
Griffin, described as the first Human Interaction Model to pass a video Turing Test, has already reached #1 on NVIDIA's full-duplex AI video benchmark, with 44% of people mistaking it for a real person compared to ~3% for other systems.
Similar Articles
@wesbos: This whole thing stinks Every quote tweet is paid for and every reply is from a bot. There is no chance 48% of people t…
Tavus announced Griffin, what it calls the first Human Interaction Model and the first real-time video system to pass a video Turing test, with a self-run study claiming 48% of one-minute call participants believed they spoke to a real person, plus a top score on NVIDIA's VideoFDB perception track.
We're cooked. This is the first model to pass the Video Turing Test. Half of people who talked to it thought it was a real human
A new video model reportedly passed the 'Video Turing Test,' with half of participants believing they were conversing with a real human — a milestone in realistic AI-generated video interaction.
Griffin by Tavus
Tavus 推出 Griffin——号称全球首个「人类交互模型」，可实时看、听、说并响应停顿、表情和手势，一次一分钟视频通话中 48% 的参与者误以为在和真人对话。Griffin-Lite 作为研究预览向早期测试者开放，支持全双工视频到视频交互及 720p 生成。
Tavus's Griffin avatar fooled 48% of job-interview subjects, and the fooled were as confident as the un-fooled
Tavus 初创公司发布新模型 Griffin，称其 AI 虚拟形象在求职面试测试中骗过了 48% 的参与者，被认作真人，声称已能通过图灵测试。
AI can finally pass the Turing Test better than a human, study warns
A new study published in PNAS shows that advanced LLMs like GPT-4.5 can pass the Turing Test, with participants finding them more human than actual humans, prompting a reevaluation of what the test measures.