With AI ads, streaming platforms will watch you more than you watch them

A Raghavan is IE’s AI columnist, offering deep dives into the world of artificial intelligence and its transformative impact across industries. The bi-monthly AI Logs column explores the latest trends, breakthroughs, and ethical dilemmas in AI, delivering expert analysis and fresh insights. To stay informed, subscribe to our AI Logs newsletter for exclusive content.

When we were young, we had a single television set at home (mostly because we couldn’t afford more than one), and a family of 10 people would huddle together and watch a cricket match or a sappy movie. There were advertisements back then too, albeit the same 7-8 advertisements would play between the televised content, which we’d also sometimes know the dialogues of by heart. So, we would avoid watching them the second time they came round. No one, even back then, would watch advertisements with undivided attention.

In the last two decades, things have changed. Many of us have graduated from television to streaming platforms like YouTube, Amazon’s Prime Video, Netflix, etc. There was a short period when these streaming platforms were a sanctuary from a relentless barrage of commercials. In 2007, Hulu was the first to take the plunge and plug advertisements between content. 

YouTube followed soon after with overlay ads and pre-roll ads around 2007. In 2009, it launched the YouTube Partner Program, allowing creators to monetize videos with ads. It was the first platform where user-generated content became monetized through ads. If one wishes to opt out, they pay for a Premium subscription.

Then Netflix, a mail-order DVD rental service, launched its streaming platform in 2007. Once a haven for uninterrupted binge-watching, Netflix embraced advertisements with open arms in 2022. And in 2025, armed with artificial intelligence, these platforms don’t want to show us ads. They want to tailor the ads to match our reality.

Product placement on steroids

Netflix plans to introduce AI-generated interactive mid-roll and pause ads by 2026 into their lowest price tier. What are interactive mid-roll and pause ads, you ask? 

Interactive mid-roll ads will show up in the middle of a show or movie, but instead of just watching them like regular ads, you might be asked to click on something, answer a question, or explore a product right there. These ads will also be smart, using AI to determine what kind of ad you’re most likely to respond to based on what you watch or how you watch it. Whereas, pause ads will pop up when you pause your video. So instead of just freezing the screen, Netflix might show a banner or an ad while you’re away, turning even your bathroom breaks into marketing opportunities.

Netflix says its AI-powered ad system promises to make commercials feel more “relevant” by blending them into the shows we watch. Using generative AI, these ads will match the look and tone of the content, with interactive features like buttons and pop-ups to keep us engaged.

This is a clever way to make ads harder to ignore and turn every pause, click, or moment of distraction into a monetizable event. The announcement, which Media Play News first reported, has since had people up in arms, concerned about how far platforms will go to monetize our attention. Some have even threatened to cancel their subscriptions, longing for the days when “Netflix and chill” didn’t involve a sales pitch.

Amy Reinhard, president of advertising at Netflix, recently said at the Netflix Upfront 2025, “…members pay as much attention to midroll ads as they do to the shows and movies themselves.” But, do they, Amy? Are you sure they do not just tolerate watching the ads so they can get to the content?

It looks like that is exactly the case. Netflix’s ad-supported plan is growing rapidly, with 94 million monthly active users, more than double the 40 million it had a year ago, and up from 70 million just six months back.

At $7.99/month, it’s far cheaper than the $17.99/month ad-free tier, and clearly, price matters. Half of all new Netflix subscribers are now choosing the ad-backed option. With over 300 million total subscribers across all tiers globally, this surge shows users are increasingly willing to tolerate ads in exchange for savings. This incentivizes Netflix to double down on ads and roll out even more targeted, AI-powered formats.

Other platforms using AI

Not to be outdone, YouTube is also going the AI way regarding advertising. The platform will be leveraging AI to identify “Peak Points” in videos using Google’s Gemini AI. The AI will identify emotionally charged moments where viewers are most engaged while watching a video. Exactly after these moments, YouTube will slyly and strategically place ads. It’s the platform’s equivalent of saying, “We noticed you’re emotionally vulnerable, here’s an ad for tissues.”

The AI scans videos frame by frame and dissects transcripts to find these high-engagement moments. While it’s unclear if YouTube is also factoring in user behaviour like rewinds or pauses, the overall strategy reeks of emotional manipulation. It’s not just targeted advertising anymore; it could very possibly be jarring to suddenly see an advertisement when you’re expecting a climax to the scene. And don’t forget it’s that annoying.

“You wanna know a good way for me to not buy your product? Interrupt what I’m doing to advertise it to me,” wrote a user on YouTube.

“Subscription-based streaming services need to understand that their entire business model is built on being slightly more convenient than piracy,” wrote another.

AI is also helping Chinese tech giants boost ad performance. According to a CNBC report, Tencent, JD.com, and Alibaba are leaning heavily on generative AI to sharpen targeting, raise click-through rates, and simplify ad campaign management.

Tencent, for instance, saw a dramatic jump in ad effectiveness, from a mere 0.1 percent to nearly three percent click-through rates, which it credited to AI. JD and Alibaba also report double-digit growth in marketing revenues thanks to AI tools that automate and optimize ad campaigns.

These AI-powered ads might seem smarter and more interesting, but they raise big privacy and control issues. How do these platforms know what level of attention a user is paying to an advertisement vs the programme?

We know that the platforms collect and store data of our every move to send us targeted ads. What we watched, what we paused on, how long we hovered over a thumbnail, how often we rewound that one scene in Bridgerton. They take notes. Our preferences, our impulses, and even our guilty pleasures are quietly being logged and fed into the hungry mouth of the algorithm. And they are not collecting this data from people who don’t pay; they are collecting it from people who do pay.

This brings to mind the first episode of Black Mirror Season 7, “Common People,” which ironically aired on Netflix. The episode shows Amanda surviving through a tech service that streams her consciousness back into her body for a subscription fee. As subscription costs soar, she’s forced to speak out ads involuntarily, even in her sleep. To stop the involuntary ads, she must pay for a higher tier plan, only to realise later that the higher tier plan has now become the basic tier plan. Her partner, Mike, ends up doing degrading livestreams to pay the bills. The episode is a biting critique of a world where even staying alive is paywalled.

These AI-powered ads will be built on constant surveillance. We’re paying to be watched more closely, not less. And the more “personalised” things get, the more we lose control over what’s truly ours: our time, our choices, and our privacy.

At some point, we have to ask: how much of ourselves are we willing to trade just to save a few bucks on a subscription?