{"rewrite":{"id":"r_8752cc34ea663925c3ba2c2d","clusterId":"c_bd3d114dc520c189d8363abb","slug":"tavus-unveils-griffin-dialog-model-after-48-of-test-callers-took-it-for-human","model":"deepseek-v4-1-flash","headline":"Tavus Unveils Griffin Dialog Model After 48% of Test Callers Took It for Human","summary":"US company Tavus announced Griffin on October 1, a dialogue model that reads audio and video at the same time. In an internal test, 26 of 54 participants recruited in the US and Europe who held a one-minute video call with the research preview Griffin-Lite judged the other party to be a real human. Tavus is working on safety measures, including features that indicate it is AI, before general availability.","whyItMatters":"Tavus is holding back general availability of a model it says people could not reliably tell from a human, which puts the disclosure features, not the dialogue quality, at the center of when Griffin actually ships.","webCardHtml":"\u003cp\u003eThe test ran on Griffin-Lite, the research preview, not the announced Griffin model itself. Participants were told the other party was another participant, and 26 of 54 judged it to be human. Griffin is built to decide whether to speak, wait, or backchannel at intervals under one second while it listens to and watches the conversation, and it generates facial expressions, gaze, and gestures in step with the exchange. Only early testers can use Griffin-Lite now. Tavus says human-indistinguishable dialogue models require cautious release and will reach general availability after its safety features are in place.\u003c/p\u003e","blueskyPost":"Tavus says Griffin reads audio and video together, but Griffin-Lite convinced 26 of 54 callers it was human. Safety labels have to work on someone already treating the call as real.","twitterPost":"Tavus plans AI-disclosure features for Griffin, but Griffin-Lite already fooled 26 of 54 callers. A label only helps if the caller is still asking.","threadsPost":"Tavus is building disclosure features for Griffin before general availability, yet in its own test 26 of 54 callers judged Griffin-Lite human. The disclosure lands at the moment a convincing model is the product, which is the hardest moment to make a label land.","newsletterBlurb":"Tavus announced Griffin on October 1, a dialogue model that processes audio and video at the same time. In an internal test, 26 of 54 participants who held a one-minute video call with the Griffin-Lite preview judged the other party to be a real human. Tavus says general availability will come after it addresses safety measures, including features that indicate the model is AI.","attributionJson":"[{\"source\":\"ASCII.jp\",\"url\":\"https://ascii.jp/elem/000/004/439/4439453/?rss\",\"title\":\"AI Video Call for 1 Minute: 48% Judged It 'Human' - US AI Company Unveils New Model Aiming for 'Human-Like Dialogue'\"}]","lintFlagsJson":null,"lintHits":0,"costUsd":0,"inputTokens":4718,"outputTokens":610,"status":"published","repairAttempts":0,"nextRepairAt":null,"factsAttemptedAt":1791177264,"createdAt":"2026-10-05T05:09:49.000Z","publishedAt":"2026-10-05T05:13:22.000Z","updatedAt":"2026-10-05T05:13:22.000Z"},"cluster":{"id":"c_bd3d114dc520c189d8363abb","canonicalTitle":"AIと1分間ビデオ通話、48％が「人間」と判断　米AI企業が“人間らしい対話”目指す新モデル","representativeArticleId":"a_c74a18d23e116df112234d93","sourceCount":1,"writtenSourceCount":1,"writeAttempts":0,"isSolo":true,"entitiesJson":"{\"anime_titles\":[],\"manga_titles\":[],\"work_titles\":[],\"studios\":[],\"people\":[],\"type\":\"announcement\",\"domain\":\"other\",\"is_roundup\":false}","contentType":"news","status":"published","firstSeenAt":"2026-10-02T01:50:00.000Z","lastSeenAt":"2026-10-02T01:50:00.000Z","updatedAt":"2026-10-05T05:13:20.000Z"},"attribution":[{"source":"ASCII.jp","url":"https://ascii.jp/elem/000/004/439/4439453/?rss","title":"AIと1分間ビデオ通話、48％が「人間」と判断　米AI企業が“人間らしい対話”目指す新モデル"}],"entities":{"anime_titles":[],"manga_titles":[],"work_titles":[],"studios":[],"people":[],"type":"announcement","domain":"other","is_roundup":false},"keyFacts":null}