tavus

Infobae

Una IA convenci a la mitad de quienes hablaron con ella por video: a ninguno le avisaron que poda ser una mquina

The Rundown AI

Tavus previews Griffin, a lifelike AI that reacts on video calls

Tavus

Griffin: The First Human Interaction Model

Business Wire

Tavus Introduces Griffin The First Face To Face Human Interaction Model Unlocking The Future Of Conversational Ai

XenoSpectrum

Tavus Griffin Video Conversation Test

SlopTV News

Half of Tavus's testers mistook its AI for a real person

CellCog

Tavus Griffin and the 48% Video Turing Test, Explained

Protos

Tavus call bot sparks AI scam, psychosis fears

MattWalshBlog

Metaculus

Byron Donalds

David Jolly

“As Governor I'd enact a moratorium on data center construction.”

“When it comes to technology, AI and everything else, Florida should lead.”

“I’m actually fighting in Congress to make sure data centers pay their own way and use their own water.”

Where do candidates in the Florida governor's race stand on AI data centers?

Tavus, a San Francisco-based AI research company, unveiled its system called Griffin on Oct. 1, describing it as the first Human Interaction Model, a full-duplex video-to-video system that sees, hears, speaks and reacts during live face-to-face conversations.

Tavus, a San Francisco-based AI research company, unveiled its system called Griffin on Oct. 1, describing it as the first Human Interaction Model, a full-duplex video-to-video system that sees, hears, speaks and reacts during live face-to-face conversations.

In a company study, 26 of 54 participants, or 48%, believed Griffin-Lite was a real person after a one-minute video call. Tavus' earlier system, combining Phoenix-4.5, Sparrow-2 and Raven-1, convinced 1 of 41 participants, or 2.4%.

In a company study, 26 of 54 participants, or 48%, believed Griffin-Lite was a real person after a one-minute video call. Tavus' earlier system, combining Phoenix-4.5, Sparrow-2 and Raven-1, convinced 1 of 41 participants, or 2.4%.

Participants were recruited through an independent research platform and told they would be matched with another participant to discuss what they were looking forward to this year. Only at the survey's end were they asked whether they suspected an AI.

Participants were recruited through an independent research platform and told they would be matched with another participant to discuss what they were looking forward to this year. Only at the survey's end were they asked whether they suspected an AI.

On NVIDIA's VideoFDB benchmark, Griffin-Lite scored 3.83 out of 5 on the generation track, against 3.92 for human reference recordings and 2.80 for the next listed system, Gemini 2.5 paired with Anam.

On NVIDIA's VideoFDB benchmark, Griffin-Lite scored 3.83 out of 5 on the generation track, against 3.92 for human reference recordings and 2.80 for the next listed system, Gemini 2.5 paired with Anam.

On the benchmark's perception track, Griffin-Lite scored 3.73 compared with 4.20 for humans. Its speaking-timing match rate was 62.8%, versus 78% for the human reference, and its median latency was 1,892 milliseconds against 900 milliseconds.

On the benchmark's perception track, Griffin-Lite scored 3.73 compared with 4.20 for humans. Its speaking-timing match rate was 62.8%, versus 78% for the human reference, and its median latency was 1,892 milliseconds against 900 milliseconds.

Tavus is limiting Griffin-Lite to select trusted testers as a research preview while it develops disclosure and safety features, and has not given a public launch date. The company says more than 150,000 developers and businesses use its earlier models.

Tavus is limiting Griffin-Lite to select trusted testers as a research preview while it develops disclosure and safety features, and has not given a public launch date. The company says more than 150,000 developers and businesses use its earlier models.

Infobae #

The Rundown AI #

Tavus #

Business Wire #

XenoSpectrum #

SlopTV News #

CellCog #

Protos #

Metaculus #

Tavus, a San Francisco-based AI research company, unveiled Griffin on Oct. 1, describing it as the first Human Interaction Model, a full-duplex video-to-video system that sees, hears, speaks and reacts during live face-to-face conversations.

Tavus, a San Francisco-based AI research company, unveiled Griffin on Oct. 1, describing it as the first Human Interaction Model, a full-duplex video-to-video system that sees, hears, speaks and reacts during live face-to-face conversations.

Rogue AI Agents

AI: Education

The AI Doc: Or How I Became an Apocaloptimist

AI: Deepfakes

AI: Chatbots

AI: Safety and Regulations

If Anyone Builds It, Everyone Dies

Artificial Intelligence

Tavus AI Says it Passed the 'Video Turing Test'

Half of Testers Thought Tavus' AI Video Caller Was Human

Half of Testers Thought Tavus' AI Video Caller Was Human

ai-security-institute

aibusiness

anthropic

arxiv

Axios

businessinsider

businessstandard

CNBC

covertswarm

csa-lab-space

csis

cybernews

cyberscoop

dataconomy

esecurityplanet

folketinget

Forbes

github

google

google-deepmind

Guardian

Independent

infoq

Intercept

malwarebytes

nature

notus

nypost

openai

reason.com

Reuters

reversinglabs

socket

substack

sysdig

TechCrunch

theassociatedpress

theregister

Time

topaithreats

venturebeat

Vice

Wired

youtube

Ukraine

troops

berkeleynews

bostonuniversity

cell-press

conversation

dailymail

dylan-wiliam

europeanunion

financialpost

gallup.com

grand-view-research

grand-view-research-1-public figure

ministry-of-education-china

mitnews

stanford

the-next-30-years

the-next-30-years-1-public figure

the-white-house

thenewyorktimes

thewashingtontimes

unesco

university-canada-west

walton-family-foundation

av-club

azfamily

cnet

engadget

exclaim

flickering-myth

focus-features

insession-film

los-angeles-times

mashable

rogerebert

screen-anarchy

screen-daily

sfchronicle

sundance-institute

the-ai-doc

the-joy-of-movies

the-nerds-of-color

variety

Verge

we-live-entertainment

frontiers

gov.uk

ofcom

onlinelibrary

research-gate

sagejournals

unherd

amazon-web-services

artificial-analysis

arxiv.org

brookings

elon-university

garymarcus

harvardbusinessreview

medium

mistral-ai

mit-press-direct

mittechnologyreview

nvidiablog

reutersinstitute

sciencedirect

sciencedirect-

technologymagazine

theofficialmicrosoftblog

universityofoxford

ai-safety-newsletter

apnews

asia-times

asiatimes

cato

civitas

committee-on-oversight-and-government-reform

common-sense-media

coursera

digital-watch-observatory

european-comission

europeancommission

futureoflifeinstitute

international-ai-safety-report

latham-watkins

lemonde.fr

national-conference-of-state-legislatures

ncbi

restofworld

shapingeurope’sdigitalfuture

stanford-medicine

the-daily-economy

thomsonreuters

trump-white-house

us-senate-committee-on-finance

whitehouse

Bloomberg

futurism

Half of Testers Thought Tavus' AI Video Caller Was Human

TavusHalf AIof SaysTesters itThought PassedTavus' theAI 'Video TuringCaller Test'Was Human

The Controversies

Jolly

Donalds

Go Deeper

Sign Up for Our Free Newsletters
Sign Up for Our Free Newsletters

Half of Testers Thought Tavus' AI Video Caller Was Human

TavusHalf AIof SaysTesters itThought PassedTavus' theAI 'Video TuringCaller Test'Was Human

The Controversies

Jolly

Donalds

Go Deeper

Sign Up for Our Free NewslettersSign Up for Our Free Newsletters

Sign Up for Our Free Newsletters
Sign Up for Our Free Newsletters