Meta’s Make-A-Video system reached a significant milestone in September 2022 by enabling video creation from text prompts without requiring paired text-video training data. The text-to-video AI market grew from $0.24 billion in 2023 to $0.31 billion in 2024, representing a 30.7% compound annual growth rate. Meta announced $60-65 billion in planned capital expenditure for 2025 to expand AI infrastructure, supporting advanced video generation capabilities.
Make-A-Video Key Statistics
- Make-A-Video launched on September 29, 2022, developed by a team of 13 researchers at Meta AI
- The text-to-video AI market reached $0.31 billion in 2024 and is projected to grow to $1.18 billion by 2029
- Meta’s Movie Gen Video model uses 30 billion parameters to generate 16-second videos at 1080p HD resolution
- Marketing professionals using AI video generation tools increased from 18% in 2023 to 41% in 2025
- Companies implementing AI video tools achieve up to 80% savings in production time and budget compared to traditional methods
Make-A-Video Technical Architecture and Model Design
Make-A-Video employs a spatial-temporal pipeline that builds on existing text-to-image models while incorporating temporal learning capabilities. The system separates learning into two distinct phases: visual appearance understanding through text-image data and motion dynamics comprehension through video-only training.
The architecture includes a video decoder, interpolation model, and two super-resolution models that work together to produce high-quality output. This design eliminates the requirement for paired text-video training data, accelerating development while maintaining the diversity of image generation systems.
| Technical Component | Specification |
|---|---|
| Architecture Type | Spatial-Temporal U-Net with Diffusion |
| Training Approach | Text-Image pairs + Unsupervised video |
| Key Innovation | No paired text-video data required |
| Output Capabilities | Text-to-Video, Image-to-Video, Video Variations |
| Pipeline Components | Video decoder, interpolation, 2 SR models |
Make-A-Video Market Growth and Financial Projections
The text-to-video AI market demonstrated consistent expansion from 2023 through 2024, with projections indicating sustained growth through 2034. The market recorded $0.24 billion in 2023, increasing to $0.31 billion in 2024, marking a 30.7% annual growth rate.
Multiple research firms forecast the market will reach $1.18 billion by 2029, maintaining a compound annual growth rate of 30.9%. Long-term projections show the broader AI video market expanding to $246.03 billion by 2034, representing one of the fastest-growing segments in AI technology.
| Year | Market Value | CAGR |
|---|---|---|
| 2024 | $0.31 billion | 30.7% |
| 2025 | $0.40 billion | 30.9% |
| 2029 | $1.18 billion | 30.9% |
| 2030 | $14.8 billion | 35% |
| 2034 | $246.03 billion | 36.2% |
Regional Market Distribution for AI Video Generation
North America maintained dominance in the AI video generator market with a 40.61% share in 2024, valued at $249.7 million. The region benefits from rapid technology adoption and robust digital infrastructure that supports AI development.
Asia-Pacific secured the second-largest position with 31.40% market share and is projected to record the highest compound annual growth rate at 35.2%. The United States accounted for 36.9% of the global market, with an expected value of $155.3 million in 2025.
| Region | 2024 Market Share | 2024 Value |
|---|---|---|
| North America | 40.61% | $249.7 million |
| Asia-Pacific | 31.40% | Largest by revenue |
| United States | 36.9% | $3.1 billion |
| Europe | Substantial | $165.8 million |
| Germany | Growing | $36.2 million |
Professional Adoption and Usage Trends
Marketing professionals and content creators accelerated their adoption of AI video generation tools between 2023 and 2025. The percentage of professionals using AI for video creation jumped from 18% in 2023 to 41% in 2025, with an additional 19% planning to adopt these tools.
Video marketers reported 75% usage of AI tools in their workflows, while 49% of all marketers incorporated AI video generation into their content strategies. Media companies showed strong interest, with over 70% planning AI video integration by 2025.
Industry-Specific Implementation Rates
Small businesses adopted AI video tools at a 50% rate, demonstrating accessibility across different organization sizes. The daily active user base for generative AI platforms ranged between 115-180 million globally, indicating widespread consumer and professional engagement.
| User Category | Adoption Rate |
|---|---|
| Video Marketers | 75% |
| All Marketers | 49% |
| Professionals (2025) | 41% |
| Small Businesses | 50% |
| Media Companies (planned by 2025) | 70%+ |
Cost Efficiency and Production Impact
Organizations implementing AI video tools recorded significant reductions in production costs and timelines. Companies achieved up to 80% savings in time and budget compared to traditional video production methods.
Specific cost categories showed substantial improvements: voice talent expenses decreased by nearly 61%, translation costs fell by approximately 52%, and animation production time reduced by 68%. Corporate training departments saved up to 49% of video budgets through AI solutions.
| Efficiency Metric | Impact |
|---|---|
| Time and Budget Savings | Up to 80% |
| Production Cost Reduction | Up to 60% |
| Animation Production Time | 68% reduction |
| Voice Talent Costs | 61% reduction |
| Translation Expenses | 52% reduction |
| Corporate Training Budgets | 49% savings |
Meta AI Infrastructure Investment and Scale
Meta committed record capital expenditure of $39.2 billion in 2024 for AI infrastructure development. The company announced plans to invest $60-65 billion in 2025, representing a potential 130% increase over two years.
The planned infrastructure expansion includes deployment of over 1.3 million GPUs in 2025, with a compute capacity target of approximately 2 gigawatts. Meta AI reached over 700 million monthly active users in early 2025, with projections suggesting 1 billion users within the year.
| Investment Metric | Value | Year |
|---|---|---|
| Total Capital Expenditure | $39.2 billion | 2024 |
| Projected Capital Expenditure | $60-65 billion | 2025 |
| Q4 Capital Expenditure | $14.8 billion | Q4 2024 |
| Planned GPU Deployment | 1.3+ million | 2025 |
| Compute Capacity Target | ~2GW | 2025 |
| Meta AI Monthly Active Users | 700+ million | Early 2025 |
Competitive Landscape and Model Evolution
Meta progressed from Make-A-Video through successive models, each introducing improvements in output quality and duration. Movie Gen Video operates with 30 billion parameters, generating 16-second videos at 1080p HD resolution and 16 frames per second.
AI video startups raised over $500 million since January 2025, surpassing previous years’ funding totals. Major funding rounds included Synthesia at $180 million and Runway at $308 million, demonstrating investor confidence in text-to-video AI technology.
Model Progression Timeline
Make-A-Video launched in 2022 with short-clip generation capabilities and high resolution through super-resolution models. Emu Video followed in 2023, producing 4-second videos with direct high-resolution output. Movie Gen Video debuted in 2024, extending video duration to 16 seconds with synchronized audio through the Movie Gen Audio model.
| Model/Company | Year | Key Specification |
|---|---|---|
| Make-A-Video | 2022 | Short clips, high resolution via SR |
| Emu Video | 2023 | 4 seconds, high resolution direct |
| Movie Gen Video | 2024 | 30B parameters, 16 seconds, 1080p |
| Runway | – | $308 million funding |
| Synthesia | – | $180 million, 160+ AI avatars |
| Pika | – | $55 million funding |
FAQ
When did Meta release Make-A-Video?
Meta released Make-A-Video on September 29, 2022. The system was developed by a team of 13 researchers at Meta AI and represents the company’s first major text-to-video AI model.
How large is the text-to-video AI market?
The text-to-video AI market reached $0.31 billion in 2024, growing from $0.24 billion in 2023. The market is projected to expand to $1.18 billion by 2029 at a 30.9% compound annual growth rate.
What percentage of marketers use AI video generation tools?
49% of marketers use AI video generation tools as of 2024. Among video marketers specifically, the adoption rate reaches 75%, while 41% of all professionals use AI for video creation in 2025.
How much can companies save using AI video tools?
Companies achieve up to 80% savings in time and budget compared to traditional video production methods. Specific categories show 61% reduction in voice talent costs, 52% reduction in translation expenses, and 68% reduction in animation production time.
What are Movie Gen Video’s specifications?
Movie Gen Video operates with 30 billion parameters and generates 16-second videos at 1080p HD resolution. The system runs at 16 frames per second and includes synchronized audio capabilities through the 13 billion parameter Movie Gen Audio model.
Citations:
Make-A-Video: Text-to-Video Generation without Text-Video Data (arXiv)
Meta Movie Gen AI Video and Audio Generation (Neowin)
