The AI Video revolution is here
Prompters,
It’s been a game-changing few months in the AI video generator world.
One major model after another has been released with mind blowing results.
Here's one of my tests of the Luma Labs "Dream Machine", which is free to use at the moment and taking the internet by storm.
PROMPT: An eight year old boy experiencing the wonder of his first plane flight
REPORT: (Results below)
This animated short was generated in a few minutes. Suffice it to say, this five-second clip would have cost a significant amount of money, time, and creative talent to create prior to the release of these models.
Currently, most are only allowing free generations of very short clips, typically five seconds in length - but that's enough for experienced video producers to stitch clips together to create entire trailers and soon films.
Here another few tests of Luma's Dream Machine, the first using only the still image on the right and a prompt.
AI advancements now allow anyone to create virtually any moving image with a simple text prompt. This is, of course, both exciting and concerning for anyone who makes a living creating content.
I asked ChatGPT to give me a list of all the job types that will be affected by this new technology. Here's the short version:
- Video Production and Editing
- Broadcast and Film
- Advertising and Marketing
- Animation and Visual Effects
- Journalism and Reporting
- Education and Training
- Corporate Communications
- Social Media and Influencers
- Event Coverage
- Entertainment
- Gaming and VR
- Customer Service and Sales
- Legal and Compliance
- Real Estate
- Healthcare
- Public Relations
- Arts and Culture
- Retail and E-commerce
- Freelance Creatives
The AI video landscape is still in its infancy, but models are getting better every day and people in these industries will need to adjust as the playing field changes.
I encourage everyone to try some of these tools, it truly is incredible technology and fun to use.
Here is a list of the current leaders in this important area of AI:
Leading AI Video Generation Models
Runway Gen - 3 Alpha
- Initial Launch Date: Early 2023
- Latest Model: Gen-2 (Mid-2023)
- Unique Features: Video-to-video and text-to-video generation, transforming existing videos into animated renders using text prompts, images, or style presets.
- Testing: Users can test Runway models here.
OpenAI - Sora 2.0
- Initial Launch Date: Late 2022
- Latest Model: Sora 2.0 (2023)
- Unique Features: High-quality text-to-video generation capabilities, creating realistic video content from textual descriptions.
- Testing: Available through OpenAI's platform, details here.
Kling Pro 3.0 (Chinese Model)
- Initial Launch Date: Mid-2022
- Latest Model: Kling Pro 3.0 (2023)
- Unique Features: Multilingual video generation catering to the Chinese market, offering a range of video styles.
- Testing: Limited public access, more information here.
Luma Labs - Dream Machine
- Initial Launch Date: Early 2023
- Latest Model: Lumalabs 2.0 (2023)
- Unique Features: Free-to-try AI video generation, supporting various input methods including text, images, and existing video clips.
- Testing: Free trial available here.
Google - VEO
- Initial Launch Date: May 2024
- Latest Model: Veo
- Unique Features: Generates high-quality 1080p videos in various cinematic styles, supports long video generation with detailed rendering.
- Testing: Available to select creators in private preview, join the waitlist here.
META - Make-A-Video
- Initial Launch Date: Mid-2022
- Latest Model: Make-A-Video (2023)
- Unique Features: Easy-to-use video generation from text, creating short clips with high realism.
- Testing: Available in beta, details here.
Hedra Labs
- Initial Launch Date: June 18, 2024
- Latest Model: Character-1 (Beta)
- Unique Features: Expressive character video generation, including talking, singing, and rapping characters, with control over emotional dialogue, movement, and entire worlds.
- Testing: Available for testing here.
Member discussion