Netflix can now place a different product in the same scene depending on who's watching

Product placement just shifted from being a shoot service to a post-production service with AI. Most production companies haven't caught up yet.

Netflix can now show Coca-Cola to one viewer and Pepsi to another in the same scene of the same show. Without shooting anything new. Without changing the script. Without either of them noticing they're watching something different.

This isn't a concept, a pilot or a conference slide. It's the explicit direction Netflix presented at its May 2025 Upfront, with its own infrastructure already deployed and use cases proven in production.

Most articles published about this analyze it from the advertiser's or the viewer's perspective. Almost no one has looked at it from where it matters: what it means for the audiovisual production chain. For the production companies, agencies and studios that until now sold brand integration as a shoot service.

Dynamic visual contextualization, personalized per viewer

In May 2025, Amy Reinhard — President of Advertising at Netflix — presented a modular framework of ad formats with generative AI. The platform uses AI to visually integrate brand advertising within the aesthetic universe of shows like Stranger Things, Wednesday or Bridgerton.

It's not a banner overlaid on the image. It's visual contextualization: the product appears integrated into the show's color palette, lighting and tone. Formats include mid-roll ads and pause ads with interactive overlays, second-screen buttons and direct calls to action. Available in 2026 in the 12 countries where the ad-supported tier operates.

• Netflix tested the concept with Domino's during Squid Game: characters visually integrated into the "Emergency Pizza" campaign with measurable brand awareness, favorability and purchase intent metrics
• Advertising on Netflix shows 8x more brand favorability than the Connected TV average
• 162% more sales per impression compared to standard connected TV
• In March 2026, Netflix acquired the AI production startup co-founded by Ben Affleck, focused on post-production

94 million monthly active users on Netflix's ad-supported tier. With an average of 41 hours per month of consumption per user in the US. That's the market that makes dynamic product placement viable at scale. Source: Netflix Upfront 2025.

Product placement has stopped being a shoot service

The global product placement market moved $33 billion in 2024, with four consecutive years of double-digit growth (12–14% annually). On television alone, the segment concentrates over $20.6 billion. The US holds 56% of the global market.

Until now, product placement works like this: a brand negotiates script presence before shooting, the product is shot in the scene, a single version is delivered to all viewers. High cost, long timelines, zero personalization.

What Netflix is doing is product placement in post-production: the footage already exists, AI generates the visual integration after the shoot, and the result can be personalized per viewer, region or demographic. Digital was the highest-growth channel in 2023, with a 15.1% increase. The direction is clear: money is moving toward digital, personalizable and measurable formats.

If you're a production company that bills for integrating brands during the shoot, the question you should ask yourself is: what happens when that integration can be done afterward, faster, and personalized per viewer?

Merging AI with real footage: the skill that splits the market

What Netflix needs to do this isn't "having AI". Anyone with a Runway subscription or access to Seedance 2.0 has that.

What it needs is to know how to merge AI-generated elements with real footage in a way that's imperceptible. That's a specific technical competence very few production companies have today:

• Lighting match: the inserted product must respond to the same light source as the original scene
• Correct perspective: the angle and distortion must be coherent with the camera that filmed the scene
• Color integration: the product must appear inside the color grade applied to the footage, not on top of it
• Movement coherence: if there's handheld camera or travelling, the inserted object must follow that movement believably

Production companies that master this integration will sit at negotiation tables previously reserved for those with set access. Those that don't will lose contracts without understanding exactly why.

Our core is exactly this competence

Our core service is merging real footage with AI-generated elements for branded content, product content and hybrid production. We don't do dynamic product placement for streaming — that's Netflix's business at platform scale. But the underlying technical competence is the same: imperceptible visual integration. And from that operational perspective is where we analyze what's coming.

What your production company should be doing right now

Not "exploring AI" or "staying on top of trends". Three concrete actions:

Audit whether your team knows how to merge AI with real footage

Not "using AI" in general. Specifically: can your team take a shot scene and insert an AI-generated element that's undetectable? Does it know how to match lighting, perspective and color grade? If the answer is no, you have a training priority that can't wait until next quarter.

Review your brand-integration value proposition

If your pitch to brands is still "we shoot with your product in scene", you're selling the analog version of a service that's being digitized. Can you also offer integration in post? Can you produce variants per market or segment? If not, someone who can will win those briefings.

Understand that your competitor isn't just another production company anymore

Netflix doesn't need an external production company to do AI product placement. It has the content, the technology and the viewer data. The only reason a brand would still need a production company is if it offers something the platform can't scale: creative direction with judgment, visual quality that meets premium brand standards, and adaptation to contexts that aren't streaming.

What this trend doesn't mean

• It doesn't mean traditional product placement disappears tomorrow. High-budget productions will keep integrating brands on shoot when the script justifies it. A James Bond drinking his Heineken on screen has a narrative value no post-AI can replicate with the same naturalness.
• It doesn't mean the technology is perfect today. Generative AI in 2026 still produces visible artifacts in complex scenes. The improvement trajectory is rapid, but we're not at the point where any insertion passes professional scrutiny. Saying otherwise would be selling smoke.
• It doesn't mean there are no open regulatory questions. The April 2026 WGA-AMPTP agreement begins to define the framework for AI use in production with transparency clauses, but there's still much ground unlegislated, especially in Europe and Latin America.

What it does mean is that the direction is set. Disney invested $1 billion in OpenAI in December 2025, with a license to use over 200 characters in video generation tools. The major studios aren't resisting AI. They're allying with it.

Three platforms, same direction

There's $33 billion a year in product placement. Netflix has just shown that a growing part of that spend can be executed in post-production, with AI, personalized per viewer. YouTube presented "Peak Points" at the same Upfront, with AI-optimized ad placement. Amazon Prime has been experimenting for years with virtual product placement where the viewer can buy what they see on screen.

The convergence isn't a prediction. It's a fact happening in parallel at the world's three largest content platforms. For a production company, the question is no longer whether this affects you. It's whether you'll position yourself as someone who masters AI integration in audiovisual production, or as someone who finds out when they've already lost the contract.

AI Audiovisual Production

Merging real footage with AI-generated elements for branded content, product integrations and hybrid production. Historical reconstructions, VFX and delivery up to 4K. First result in 48 hours.