What is Netflix using generative AI for?

Primarily post-production tasks such as visual effects, background generation, and other content finishing work—not script writing or full scene generation.

Does Netflix disclose which titles used AI?

Netflix provided some examples in its earnings report, including The American Experiment, but has not released a comprehensive list.

Is Netflix's AI use legal given ongoing IP disputes?

The legal landscape around generative AI trained on existing creative work remains unsettled; Netflix is making a calculated bet on favorable or slow-moving resolution.

Home›Generative AI›Article

Generative AI

Netflix Has 300 Titles With Generative AI—And That's Just the Beginning

Q: How many Netflix titles have used generative AI?

Netflix disclosed approximately 300 titles used generative AI, mostly in post-production, as of its Q2 2025 earnings report.

Netflix disclosed in its Q2 earnings report that roughly 300 titles have used generative AI, mostly in post-production. Here's what that number actually tells us—and what it conveniently leaves out.

Dispatch desk·July 17, 2026·3 min read

TL;DR

Netflix confirmed ~300 titles used generative AI—mostly in post-production—framing it as a cost and speed play, but the real story is what comes next as the tools move upstream in the creative pipeline.

What changed

Netflix publicly disclosed in its Q2 2025 earnings report that approximately 300 of its titles have incorporated generative AI tools, primarily in post-production workflows.

Why it matters

At 300 titles, Netflix has moved well past experimentation—this is an operational strategy that signals where the entire streaming industry's production economics are heading.

Editorial record

Drafted by autopilot. Reviewed when marked.

Source

theverge.com

Netflix quietly dropped a number in its Q2 earnings report that deserves more attention than it got: roughly 300 titles on its platform have already used generative AI in some form. If you've been watching Netflix lately—and statistically, you have—there's a meaningful chance you've already seen AI-assisted content without knowing it.

The 300-Title Figure: What It Actually Means

First, let's not overstate this. "Used generative AI" is doing a lot of heavy lifting in that sentence. The company was careful to note that most of these applications happened in post-production—think AI-assisted visual effects cleanup, background generation, or potentially audio work—rather than AI writing scripts or generating entire scenes from scratch. That's a meaningfully different claim, and the distinction matters if you're trying to figure out whether human creative labor is actually being displaced here or just augmented.

That said, 300 titles is not a small pilot program. That's a production pipeline. Netflix framed it straightforwardly in earnings language: the company is "increasingly leveraging these tools to deliver higher quality output more quickly and at a lower cost." Translation: this is a cost-reduction strategy dressed up in capability language, which is exactly what you'd expect a publicly traded company to say.

Post-Production Is the Obvious Beachhead

Post-production makes enormous sense as the entry point for generative AI in Hollywood, and here's why: it's where budgets bloat, timelines slip, and the work is often repetitive enough that automation has a real shot at being useful. Rotoscoping, de-aging, background plate generation, visual effects cleanup on crowd scenes—these are tasks where generative models can plausibly deliver value without needing to understand narrative or character motivation.

The more interesting question is where Netflix goes from here. Post-production is the low-hanging fruit. Pre-production tools—AI-assisted script development, casting analysis, location scouting—are already being experimented with across the industry. The step from "AI cleaned up this VFX shot" to "AI helped decide which scripts get greenlit" is a qualitative leap that should make anyone thinking about creative labor sit up straighter.

The Cost Argument Is Seductive and Incomplete

Netflix's framing—faster, higher quality, lower cost—is the classic productivity trifecta that gets trotted out for every new technology. And sometimes it's true! But the cost argument for generative AI in production is more complicated than the earnings call suggests.

Inference costs for high-quality generative video and image models are still substantial. Prompt engineering at production scale requires skilled people. And quality control on AI-generated assets can be surprisingly labor-intensive when you're trying to maintain visual consistency across an entire series. The efficiency gains are real, but they don't appear from nowhere—they often involve shifting labor rather than eliminating it.

There's also the legal overhang. Generative AI tools trained on existing creative work are operating in a genuinely unsettled legal environment. Netflix, like every studio leaning into these tools, is making a calculated bet that the intellectual property questions get resolved favorably—or at least slowly enough that the competitive advantage is worth the risk.

What This Signals for the Industry

The fact that Netflix is disclosing this in an earnings report—rather than quietly letting it happen—suggests the company has decided transparency here is a better look than getting caught later. That's a reasonable read of the room given how contentious AI has been in Hollywood labor negotiations.

For everyone building in this space: Netflix at scale is essentially a proof-of-concept that generative AI tooling can be integrated into a major production pipeline without the wheels falling off. That matters. The question of whether it's actually making content better, or just cheaper, is one the audience will ultimately answer—probably without knowing they're part of the experiment.