Google's Veo 3 Can Now Create An 8-second Video From A Single Image - How To Try It

1 week ago

The latest loop of Google's video-generating AI model, Veo 3, continues to germinate astatine a accelerated clip.

As portion of its latest upgrade, nan exemplary now lets users make eight-second video clips, including AI-generated audio, from a azygous still image. According to Google Cloud documentation updated Monday, nan characteristic is now disposable arsenic a "preview offering." Josh Woodward, caput of Google Labs and nan Gemini App, primitively wrote successful an X post last week that nan institution was moving connected image-to-video capabilities for Veo 3.

What to usage Veo 3 for

An influencer, for example, could upload a azygous headshot of herself and punctual nan exemplary to make a short clip of her stepping down a runway wearing a merchandise from a marque she's collaborated with. Veo 3 would automatically see ambient noise, for illustration nan murmurings of nan crowd and her footfalls connected nan floor; nan personification could besides petition that her AI-generated likeness speak a fewer lines, like successful this example.

Also: This caller AI video editor is an all-in-one accumulation work for filmmakers - really to effort it

Brands could besides usage nan caller characteristic by feeding nan exemplary an image of a merchandise and asking for a clip that displays it from a scope of different perspectives. Amazon has developed an AI tool for advertisers pinch akin capabilities, while Meta has vowed to spell further, stating its plans to automate nan entirety of nan ad-production process.

Veo 3's caller image-to-video capacity could thief imaginative professionals successful various industries prevention clip and resources that would different beryllium spent organizing on-site video shoots. It tin besides supply much imaginative materials for usage crossed societal media and different channels.

Google revealed Veo 3 successful May astatine its annual I/O developer conference. The exemplary quickly attracted attraction from AI researchers and imaginative professionals for its expertise to seamlessly merge AI-generated video and audio, a highly technically analyzable feat that promises to unfastened caller doors for AI-assisted filmmaking. It besides excels astatine simulating real-world physics and isn't hindered by nan galore method glitches that plagued earlier AI-generated video tools.

Also: How AI companies are secretly collecting training information from nan web (and why it matters)

There's nary motion that Google's finance successful Veo 3 will slow down anytime soon. Last week, Google DeepMind CEO Demis Hassabis appeared to hint successful a X post that nan exemplary could soon beryllium utilized to make virtual worlds for video games. The timing of that prediction that is interesting, fixed Microsoft laid disconnected 9,000 people from its gaming section earlier this week.

How to effort it

Originally only disposable done Gemini Ultra and Flow, Veo 3 was generally released arsenic a nationalist preview past period -- each Google Cloud customers and partners tin entree it successful the Vertex AI Media Studio. The exemplary is now disposable crossed 159 countries.

Controversy and imaginable risks

Veo 3 has sparked concerns over AI's imaginable to supercharge nan spread of online misinformation and manipulate users connected societal media. There are besides questions astir nan sourcing of its training data, which Hassabis has said could see YouTube videos.

Also: This open-source bot blocker shields your tract from pesky AI scrapers - here's how

Since AI companies scraped overmuch of nan text, image, audio, and video contented they usage to train their models from nan unfastened internet, creators from crossed nan publishing, art, and movie industries person raised copyright issues pinch these generators. If you're looking for a much airtight AI-generated video tool, see checking retired Moonvalley's Marey, which claims to beryllium trained exclusively connected licensed data.