Hook: The Dashboard That Connects Hollywood and Fortune 500
A Hollywood film production team and a Fortune 500 ad agency are both running their AI-generated video pipelines through the same dashboard. That dashboard just closed a $12 million funding round in two tranches – a $2 million pre-seed led by The General Partnership, followed six months later by a $10 million seed round led by Sequoia. The product is Preview, and Sequoia’s thesis is clear: AI video needs a 'video version of Cursor.'
But here’s the catch – Preview isn’t a new AI model. It’s a central control panel that brings scripts, storyboards, shot lists, AI generation, review, and feedback into one workspace. Teams can simultaneously use different models, manage characters, scenes, and props, and each frame records who generated it, what model was used, and the parameters applied. Over 100 studios are already using it, and 3,000 more are waiting in line.
Context: Why Now, and Why the Pipeline Gap Matters
We’ve seen this pattern before. In the 2017 ICO frenzy, every project had a whitepaper but no product. In the 2021 NFT mania, every artist had a collection but no minting workflow. Now, AI video is exploding – Runway, Pika, Sora, and dozens of open-source models are generating stunning clips. Yet the production process remains fragmented: teams juggle separate tools for scripting, storyboarding, generating, and reviewing. The result is chaos, not cinema.
From my experience covering the crypto-AI crossover, I’ve watched countless startups build better models while ignoring the integration layer. Sequoia’s bet on Preview signals a shift: the bottleneck isn’t generation quality – it’s workflow. Digital gold rushes turn pixels into portfolios, but only if the pixels can be assembled into a coherent story. Preview aims to be that assembly line.
Core: The Technical and Market Mechanics
Preview’s core value proposition is simple: it acts as a unified workspace for AI video production. Think of it as a digital production board where every asset – from character designs to camera angles – lives in one place. The platform integrates with multiple AI models (e.g., Runway, Pika, Stable Video Diffusion) and allows teams to switch between them mid-project. Each frame carries a metadata audit trail: who generated it, which model, what parameters, and when. This is critical for professional workflows where reproducibility and accountability matter.
- Funding Breakdown: $2M pre-seed (The General Partnership) + $10M seed (Sequoia) = $12M total. The speed of the rounds (six months apart) reflects the urgency of the market.
- Traction: 100+ active studios, including agencies producing ads for Fortune 500 companies and Hollywood film production teams. 3,000 studios on the waitlist – a demand signal that cannot be ignored.
- Technical Edge: Unified character/scene/prop management across models. This solves the 'uncanny valley' of inconsistent AI outputs – one model might generate a character’s face, while another generates the background. Preview keeps them consistent.
Sequoia’s comparison to Cursor is telling. Cursor became the standard for AI-assisted coding by providing a seamless interface on top of existing models. Preview is attempting the same for video. But unlike code, video production involves creative direction, emotional beats, and client feedback loops. The metadata tracking is a blockchain-like provenance system – every frame has a hash of its creation. This could be a goldmine for verifying authenticity in a world of deepfakes, but it’s centralized in Preview’s servers.
Contrarian: The Pipeline Isn’t the Problem – It’s the Monetization
Here’s the unreported angle: Preview solves the workflow problem, but it doesn’t solve the revenue problem. In a bear market, studios are cutting costs, not expanding toolchains. The 3,000 studios on the waitlist are likely testing the waters, not committing to a paid subscription. Pulse checks on the volatile heartbeat of exchange tell me that the real bottleneck is not the production pipeline – it’s the lack of a reliable revenue stream for AI-generated content.
Consider the NFT market. Dynamic NFTs and programmable royalties sound cool, but artists need stable buyers, not a more complex tech stack. Preview’s metadata tracking could enable on-chain royalties for every frame, but the platform itself is not blockchain-native. The smart money whispers that the real value lies in becoming the standard for provenance, but that requires decentralization – which Preview doesn’t offer.
Also, the 'video version of Cursor' analogy is flawed. Cursor succeeded because developers already had a clear workflow (coding) and needed a better interface. Video production is still in flux – the models themselves are rapidly evolving, and any workflow tool risks becoming obsolete when a new generation model emerges. Preview’s reliance on multiple models is a strength, but it also means they are dependent on third-party APIs that can change pricing or shut down.
Takeaway: Watch the Integration Layer, Not the Models
From frenzy to function: tracing the cycle – we are in the 'function' phase of AI video. The hype is cooling, and the survivors will be the ones that integrate into professional workflows. Preview has a strong start: Sequoia’s backing, real traction, and a clear pain point. But the 3,000-studio waitlist is a double-edged sword – it signals demand, but also immense pressure to scale.
My forward-looking judgment: Preview will either become the de facto standard for AI video production, or it will be disrupted by an open-source alternative that integrates blockchain-based provenance and decentralized storage. The next 12 months will tell us whether the pipeline is the gold mine, or just another picks-and-shovels story in a market that hasn’t yet found its footing.