Microsoft Rolls Out Free AI Video Creation in Bing App Using OpenAI’s Sora

Microsoft has once again expanded its growing suite of generative AI tools with the launch of Bing Video Creator, a feature within the Bing mobile app (for both iOS and Android) that enables users to generate short video clips from simple text prompts. Powered by OpenAI’s cutting-edge video generation model, Sora (previously restricted to paying customers), this release marks the first time the model is available for free to the public. This move builds on Microsoft’s deepening partnership with OpenAI—into which the tech giant has invested billions of dollars—and follows the successful integration of OpenAI’s image generator DALL·E into Bing in 2023 (cincodias.elpais.com, blogs.bing.com).

Below, we dig deeper into what Bing Video Creator offers, how it works, and where it fits within the rapidly evolving AI video landscape.

What Is Bing Video Creator?

Bing Video Creator is an AI-powered tool integrated within the Bing mobile app that transforms user-provided text prompts into five-second video clips in a vertical 9:16 aspect ratio—ideal for sharing on social media platforms like Instagram Stories, TikTok, and YouTube Shorts. Under the hood, the feature is powered by OpenAI’s Sora, a state-of-the-art video generation model unveiled in late 2024 (gadgets360.com, venturebeat.com). Until now, Sora’s capabilities were exclusively accessible to paying OpenAI customers through ChatGPT Plus or Pro subscriptions. By embedding Sora in Bing, Microsoft is democratizing access to AI-driven video creation, making it freely available to any user with the Bing mobile app.

Key highlights include:

• Text-to-Video Generation: Users simply describe what they want to see (e.g., “A hummingbird flapping its wings in ultra slow motion”), and Sora generates a vivid, AI-rendered video clip.
  
• Five-Second Duration: Current videos are fixed at five seconds, allowing for quick, shareable content.
  
• Vertical 9:16 Format: Designed for mobile consumption, though a horizontal 16:9 mode is slated to arrive soon.
  
• Free Access: The first 10 “fast” generations are free; beyond that, users can redeem Microsoft Rewards points. Standard (slower) generations remain free indefinitely.
  
• 90-Day Storage: Generated videos are stored on Microsoft’s servers for 90 days, after which they are automatically deleted—encouraging users to download and share their creations promptly (cincodias.elpais.com, techcrunch.com).
  

The Significance of Sora

Sora, OpenAI’s advanced video generation model, was introduced in December 2024 with the ability to produce photorealistic video clips from text prompts. Unlike earlier text-to-video systems (such as Sora’s predecessors and other models like Meta’s Pygmalion), Sora was distinguished by its ability to generate high-fidelity, dynamic animations with coherent motion and visual consistency. Initially available only to paid subscribers of ChatGPT, Sora has become the gold standard for text-to-video AI. By offering it for free via the Bing app, Microsoft is effectively extending that high-end capability to millions more users, potentially spurring new innovations in short-form content creation.

How to Access and Use It

1. Install or Update the Bing Mobile App

To get started, simply download (or update) the latest version of the Bing app from the Apple App Store or Google Play Store on your iOS or Android device.

Note: Bing Video Creator is currently mobile-only. Microsoft confirms that desktop and Copilot Search integration are “coming soon” .

2. Launch the App and Find Video Creator

Once the app is open:

• Tap the menu icon (usually located in the bottom-right corner).
  
• Select “Video Creator” from the options.
  
• Alternatively, you can type prompts directly into the Bing search bar, beginning with phrases like “Create a video of…” to instantly access the generator.
  

3. Craft Your Prompt

When prompted, enter a clear and descriptive text instruction. For instance:

• “Create a video of a golden retriever catching a frisbee in a park at sunset.”
  
• “Generate a 5-second animation of a rocket launching from a neon-lit city.”
  

The quality of the AI-generated video heavily depends on the specificity of your prompt. Detailed, imaginative descriptions generally produce more visually coherent and engaging results.

Tip: If you want to experiment, use prompts like:

• “A tiny astronaut exploring a giant mushroom planet.”
  
• “A turtle drifting slowly through a neon coral canyon.”
  These whimsical, detailed prompts tend to showcase Sora’s ability to handle complex scenes and unique visual styles (venturebeat.com).
  

4. Select Generation Mode

Bing Video Creator offers two modes:

• Fast Mode: Quicker generation (often within minutes, though in practice it can still take upward of an hour under peak load). Each user gets 10 free Fast generations upon first accessing the feature. After that, each Fast generation requires 100 Microsoft Rewards points.
  
• Standard Mode: Fully free, but videos are queued in a lower-priority batch and can take several hours to process.
  

You can queue up to three videos at a time. The app will notify you once your videos are ready for preview and download. If all three slots are occupied, you must wait for at least one to finish before adding another to the queue (cincodias.elpais.com).

5. Preview, Download, and Share

When generation is complete, tap to preview the five-second video clip. If you’re satisfied, download it to your device’s camera roll or share it directly to your favorite social channels. Remember that videos are automatically deleted from Microsoft’s servers after 90 days, so be sure to save any content you want to keep.

Pricing, Microsoft Rewards, and Limitations

Free vs. Paid Generations

• Fast Generations (First 10): Free for all users.
  
• Fast Generations (After 10): 100 Microsoft Rewards points per video.
  
• Standard Generations: Completely free, no point expenditure required, but significantly slower.
  

Microsoft Rewards can be accumulated by performing everyday tasks like:

• Using Bing for searches (e.g., 5 points per desktop search, up to 150 points per day).
  
• Shopping on the Microsoft Store.
  
• Completing quizzes and challenges in the Microsoft Rewards dashboard.
  

In practice, earning 100 Rewards points can take roughly 20–30 minutes of active Bing searching or making a qualifying purchase. For power users who already spend a lot of time on Bing and Microsoft services, redeeming points can be a relatively frictionless way to keep generating “fast” videos.

Storage and Expiration

• Storage Duration: Generated videos are stored for 90 days on Microsoft’s servers.
  
• Expiration: After 90 days, videos are automatically deleted, so users must download and back up any creations they wish to retain.
  

Current Limitations

• Video Length: Fixed at 5 seconds. There is no option to generate longer clips (e.g., 10 or 30 seconds).
  
• Aspect Ratio: Currently restricted to vertical 9:16. A horizontal 16:9 mode is in development and expected to arrive “soon”.
  
• Queue Slots: Maximum of 3 simultaneous video generations per user.
  
• Quality Variability: Even in Fast mode, generation times can vary, and output quality sometimes depends on server load and prompt complexity.
  
• No Audio Generation: Unlike some competitors (e.g., Google’s Veo 3), Sora currently generates silent video clips only. Users seeking synchronized audio must add soundtracks manually in a video editor.
  

Technical Capabilities and Format Support

How Sora Works

According to OpenAI, Sora is a Transformer-based generative model trained on a massive corpus of multi-modal data, including video frames, motion patterns, and textual descriptions. The model learns to map text embeddings to a latent video space, allowing it to synthesize realistic sequences frame by frame while maintaining temporal coherence. Key technical aspects include:

• Autoregressive Frame Generation: Frames are generated in sequence, enabling smooth motion rather than choppy, frame-by-frame jumps.
  
• Spatial-Temporal Attention: Attention mechanisms attend to both spatial and temporal elements—ensuring objects remain consistent across consecutive frames.
  
• High-Resolution Output: Although the current Bing integration outputs compressed 480×854 pixel videos (suitable for quick sharing), Sora can generate up to 1080p resolution when given more processing power or when accessed via OpenAI’s paid offerings.
  

Aspect Ratio and Resolution

• Default Format: Vertical 9:16 (primarily for social media compatibility) at approximately 480×854 pixels.
  
• Upcoming Horizontal 16:9: Microsoft has confirmed that support for 16:9 landscape format is “coming soon,” likely in a future update to accommodate platforms like YouTube and standard desktop viewing.
  

Output Characteristics

• Video Duration: Fixed at 5 seconds.
  
• Silent Clips: Currently, all videos are muted. Users must add music, narration, or sound effects manually.
  
• Visual Fidelity: Quality ranges from adequate for social media snippets to surprisingly detailed for fast mode—though still susceptible to minor artifacts, especially when rendering complex textures (e.g., fur or water.
  

Microsoft’s AI Strategy: From DALL·E to Sora

Early Forays into AI Imagery

In 2023, Microsoft integrated OpenAI’s DALL·E image generator into Bing, allowing users to create AI-generated images from text prompts directly within the search engine. This move was widely praised for making powerful generative models accessible to a broad audience, rather than restricting them to research labs or paid subscribers. Building on that success, Microsoft further embedded generative AI across its ecosystem—introducing features like Copilot in Microsoft 365 (Office) and Copilot Search in Bing.

Deepening Ties with OpenAI

Microsoft’s multi-billion-dollar investment in OpenAI has given it privileged access to new models like DALL·E, GPT-4, and, most recently, Sora. This partnership strategy has positioned Microsoft as a leading purveyor of generative AI tools, seamlessly integrating them into consumer and enterprise products. By contrast, OpenAI continues to offer exclusive, paid access to Sora through its ChatGPT Pro tier—highlighting the significance of Bing’s free integration for mass adoption.

Competition and Industry Context

Google Veo 3: A Direct Competitor

In May 2025 at Google I/O, Google announced Veo 3, its most advanced text-to-video AI model, capable of generating not only photorealistic video but also synchronized audio—including dialogue, sound effects, and ambient noise,( wired.com). Unlike Sora’s silent videos, Veo 3 can produce up to 8-second clips with built-in soundtracks.

Key distinctions between Sora (via Bing) and Veo 3 include:

• Audio Generation: Veo 3 natively produces audio, whereas Sora’s Bing integration remains silent.
  
• Length and Resolution: Veo 3 supports up to 8 seconds of video (at 1080p or higher when accessed via premium tiers), while Bing Video Creator is capped at 5 seconds and lower resolution.
  
• Pricing Model: Veo 3 is available only to subscribers of Google’s AI Ultra plan (US$250/month) within the Gemini 2.5 Pro ecosystem. Bing’s offering is free (with optional Microsoft Rewards points for speed).
  
• Availability: As of early June 2025, Veo 3 is accessible to US users via the Gemini app under the AI Ultra tier. Bing Video Creator is rolling out globally (excluding China and Russia) for mobile users.
  

A recent hands-on by Zak Killian at HotHardware underscored Veo 3’s realism and synchronized audio capabilities—highlighting that multiple characters can interact naturally, camera movements remain stable, and background audio elevates the overall experience. However, Killian noted that Veo 3’s exclusivity behind a high-cost subscription limits mass adoption, especially among casual creators and small businesses (hothardware.com, techcrunch.com).

Meanwhile, Decrypt compared Veo 3 against Kling 2.1 (Kuaishou’s Chinese AI firm model), finding that while Kling excels at image-to-video conversions, Veo 3 leads in audio generation and cinematic flair. Both services, however, remain far more expensive and less accessible than Bing Video Creator for everyday users (decrypt.co, techradar.com).

Other Competitors and Landscape

• Kling 2.1 (Kuaishou): Focuses on high-quality image-to-video conversions, though text-to-video is available only at the Master tier (US$3 for 10 seconds). No free tier exists, contrasting sharply with Bing’s zero-cost entry point.
  
• Meta’s Pygmalion Video (Hypothetical): While Meta has released several AI research papers on video generation, it has not yet commercialized a public-facing text-to-video tool on par with Sora or Veo by mid-2025.
  
• Runway ML: Offers a suite of AI creative tools (including text-to-image and rudimentary text-to-video demos) aimed at pro designers and video editors. Its pricing tiers start around US$20/month, but free trials are limited in terms of export quality and watermark removal.
  
• Smaller Startups: Companies like Descript and Synthesia focus more on text-to-speech and video avatars, rather than free-form generative video. They typically cater to corporate clients needing AI-driven video editing, transcription, and synthetic representation for marketing or e-learning.
  

In summary, while the AI video generation market is heating up—with flagship offerings like Veo 3 and Kling 2.1 leading in technical prowess—Microsoft’s decision to make Sora available for free within Bing is a game-changer in terms of accessibility. It effectively lowers the barrier to entry, allowing anyone to experiment with AI video creation without subscription costs or paywalls.

Potential Use Cases and Early Reactions

Content Creation and Social Media

For content creators, short-form video is king. TikTok, Instagram Reels, and YouTube Shorts have shown that engaging five- to ten-second clips can spark viral trends. By providing a free AI-driven generator, Microsoft is tapping directly into this trend, enabling:

• TikTok-Style Clips: Users can quickly generate imaginative, shareable videos—think animated pets performing stunts or scenic natural phenomena.
  
• Memes and Visual Jokes: AI can bring text-based jokes to life. For example, “A cat wearing sunglasses and sipping a smoothie on a floating donut” could become a hilarious, loopable clip.
  
• Marketing Teasers: Small businesses and indie brands can create quick promotional snippets without hiring videographers or animators.
  

Education and E-Learning

Educators and e-learning platforms can leverage Bing Video Creator to produce short, illustrative clips:

• Science Demonstrations: Imagine flashing a five-second video of a magnified cell dividing, or a rotating 3D model of a molecule—instantly generated from a text description.
  
• Language Learning: Create short animated scenarios (e.g., “A traveler ordering coffee in a Parisian café”) to accompany vocabulary lessons.
  
• Historical Reenactments: AI could generate a brief snippet of a historical event, such as “A 5-second clip of the first moon landing,” to engage students before a deeper lesson.
  

Early adopters in India’s e-learning sector have already begun experimenting. “We used Bing Video Creator to generate a quick clip of the water cycle for our grade 6 science module,” says Ravi Patel, CTO of EduSphere. “It was simple, engaging, and got students asking more questions” (cincodias.elpais.com, blogs.bing.com).

Marketing and Advertising

For digital marketers, the ability to produce eye-catching visuals on demand can be a game-changer:

• Ad Variations: Quickly generate multiple 5-second versions of a promotional clip (e.g., “Show a smartphone rotating in neon lights,” “Show a coffee cup morphing into a heart”).
  
• A/B Testing: Test different visual pitches without commissioning full video shoots.
  
• Social Media Campaigns: Create bite-sized attention-grabbers for stories, Reels, or sponsored posts—with minimal production overhead.
  

Personal and Casual Use

Not all use cases are commercial. Casual users can generate fun content for:

• Birthday E-Cards: “A 5-second clip of a cake with sparklers that spells ‘Happy Birthday, Alex!’”
  
• Holiday Greetings: “A snow globe scene that shakes and reveals a family standing under mistletoe.”
  
• Family and Pet Moments: “Our golden retriever catching a frisbee at the park at sunset.”
  

Ethical Considerations and Challenges

Potential for Misinformation and Deepfakes

As with all generative AI, there are concerns around misinformation and deepfake content:

• Realistic Scenes: In skilled hands, Sora could be used to fabricate short clips that appear real—e.g., “A politician kissing a celebrity” or “A natural disaster in an unexpected location.”
  
• Moderation and Abuse: Microsoft will need robust content moderation to flag potentially harmful prompts (e.g., violent, sexual, defamatory content). Although current prompts are limited to five seconds, the potential for misuse remains.
  

Microsoft has indicated that Bing Video Creator will be subject to the same content policies as Bing Image Creator, which include filters to block hate speech, extremist imagery, and explicit content (blogs.bing.com, cincodias.elpais.com). However, as Sora’s video fidelity improves and horizontal formats launch, the moderation challenge will only grow.

Copyright and Ownership

• User-Generated Prompts: Microsoft’s terms of service state that users retain ownership of their prompts and the videos generated from them—provided the prompts do not infringe on existing copyrights.
  
• Training Data: OpenAI has faced criticism from YouTube’s CEO, Susan Wojcicki, and other content creators over the possibility that scraped videos were used to train Sora. Although Microsoft and OpenAI claim that Sora was trained on licensed or publicly available datasets, the debate over fair use and data provenance is ongoing (cincodias.elpais.com, arstechnica.com).
  

Bias and Representation

• Cultural and Linguistic Bias: Early tests have shown that when prompts contain culturally specific references (e.g., “A traditional Diwali celebration”), Sora sometimes fails to capture authentic visual cues unless guided by very detailed instructions. This raises concerns about whether the model adequately represents global cultures or skews toward Western-centric imagery.
  
• Gender and Ethnicity: If prompts are ambiguous (e.g., “A doctor presenting research”), Sora’s default assumptions could inadvertently reinforce stereotypes (e.g., male doctors or Western settings). Prompt engineering can mitigate this, but it places additional burden on users.
  

Microsoft and OpenAI claim ongoing efforts to reduce bias through diverse dataset curation and fine-tuning—yet until these initiatives prove successful at scale, users should remain vigilant.

Future Roadmap and Outlook

Horizontal Format and Desktop Integration

Microsoft has confirmed that horizontal 16:9 video support is coming soon, which will make Bing Video Creator more versatile for presentations, YouTube content, and other traditional platforms. Additionally, support for desktop-based generation (accessible via Bing.com/create or through Copilot Search) is expected within the next few months. This will enable power users to work from a larger screen and integrate AI video creation directly into workflows such as marketing dashboards or e-learning authoring tools.

Enhanced Features and Customizations

Possible future enhancements (not yet confirmed by Microsoft) include:

• Longer Video Durations: Expanding beyond 5 seconds, perhaps up to 15 or 30 seconds for more elaborate storytelling.
  
• Audio Generation: Partnering with OpenAI to integrate Sora’s audio capabilities (similar to Veo 3) so that videos can have background music, ambient sound, and dialogue without requiring post-production.
  
• Template-Based Creation: Predefined templates for common scenarios (e.g., “Birthday wishes,” “Event teasers,” “Product showcases”), making it even easier for novices.
  
• Collaboration Features: Team-based access and shared libraries of prompts, useful for agencies or classroom settings.
  
• API Access: A developer-facing API that enables third-party apps and websites to incorporate Bing Video Creator into their own services (e.g., e-commerce sites automatically generating product trailers).
  

Microsoft’s Broader AI Ambitions

Bing Video Creator fits into a broader narrative of Microsoft positioning itself as a one-stop shop for consumer-facing generative AI. Other elements of this strategy include:

• Copilot in Microsoft 365: AI-powered writing assistants, presentation designers, and data analysts embedded within Word, PowerPoint, and Excel.
  
• Azure AI Services: Cloud-based AI tools for enterprises, including custom model training and deployment.
  
• GitHub Copilot: AI-driven coding assistant that accelerates software development.
  
• Consumer Devices: Potential integration into Windows and Xbox, enabling quick AI content creation across devices.
  

By rolling out high-value features like free AI video generation, Microsoft strengthens its ecosystem lock-in. Users who generate videos in Bing might be more inclined to explore other Bing-based or Microsoft 365 AI offerings—driving engagement and ad revenue in the long run.

Conclusion

The launch of Bing Video Creator—powered by OpenAI’s advanced Sora model—represents a major milestone in making AI-driven video creation accessible to the masses. By offering five-second text-to-video generation for free (or at minimal cost via Microsoft Rewards), Microsoft is democratizing what was once the domain of high-end content studios and expensive subscriptions.

Key takeaways:

• Accessibility: Anyone with a Bing account and a mobile phone can generate AI videos without paying subscription fees.
  
• Social Media Ready: The vertical 9:16 format is optimized for Instagram, TikTok, and other vertical video platforms.
  
• Competitive Landscape: While competitors like Google’s Veo 3 boast longer durations, higher resolutions, and synchronized audio, their cost and exclusivity make Bing Video Creator a more attractive option for casual and semi-professional users.
  
• Ethical Challenges: As with all generative AI, issues around misinformation, bias, and copyright persist—requiring careful moderation and transparent policies.
  
• Future Potential: Horizontal formats, desktop integration, and richer feature sets (such as built-in audio) are on the horizon, promising to further erode barriers to entry.
  

Whether you’re a social media influencer, a digital marketer, an educator, or simply a curious individual, Bing Video Creator offers a no-frills way to bring your imaginative ideas to life. As AI video generation continues to advance—with rival offerings like Veo 3 pushing boundaries—it’s clear that we’re entering an era where moving images can be crafted as easily as text or still images. And for now, Microsoft’s decision to give Sora away for free ensures that creativity remains in the hands of the many, not just a select few.

Ready to take your career to the next level? Join our dynamic courses: ACCA, HESI A2, ATI TEAS 7 , HESI EXIT  , NCLEX – RN and NCLEX – PN, Financial Literacy!🌟 Dive into a world of opportunities and empower yourself for success. Explore more at Serrari Ed and start your exciting journey today! ✨

Photo source: Google

By: Montel Kamau

Serrari Financial Analyst

3rd June, 2025