Generative AI video is perfect for social media, but could disrupt social media companies

Some independent content creators are being empowered by generative tools

Generative video models can create short clips of high-quality video and audio that are nearly indistinguishable from “real” content.⁶ The relative ease of use and cost-effectiveness is empowering some creatives to run with their creativity, try things with much lower risk of failure, and even rapidly test creative ideas in the hyper-competitive marketplace of social video.

Though they may not be able to deliver 30-minute TV shows or full two-hour movies, generative video tools are very capable of producing compelling, made-for-social video content—like high-profile ads bringing internet memes to life.⁷ In fact, the perceived limitations of generative video that may slow Hollywood adoption seem to be empowering some creators and social video platforms, where short-form, fast cuts, and selfies are common to virality, and where audiences may be less discerning about the free, open-source entertainment they receive.

For independent creators—and maybe soon for all media production—generative tools may be less about replacing the entire production stack to render fully synthetic content, and more about eliminating costly micro-tasks, compressing the time to create, and empowering smaller outfits to do more.

Generative AI and video tools are powering cheaper and faster content creation, eliminating more of the micro-tasks in production, distribution, and measurement.⁸ This can amplify outputs to help keep up with a fast cadence of publishing, often necessary to engage followers and stand out in the algorithmic feeds of social platforms.

Many tools focus on time- and money-saving shortcuts, like quickly generating videos from scripts and “one-click” clip generation.⁹ This can help enable creators to rapidly test variants to determine which approaches work better with specific audiences and trending algorithms. Other tools are enabling creators to generate AI avatars of themselves that can reduce fatigue while still engaging audiences and even enabling greater personalization at scale.¹⁰ The same features are expanding into generative ads.¹¹

Generative AI tools can also support non-generative content with faster editing, like removing “ums,” silences, and bad takes, fixing shaky cameras, and automating the removal of dead space.¹² Multilingual dubbing tools can open access to foreign language audiences, expanding engagement and ad revenue potential.¹³

With these capabilities, creator studios can more quickly ideate, generate content, target audiences, measure results, and repeat. This could not only disrupt the economics of content but also lead to exponentially more content. A greater supply of content could create more competitive pressure among creators, which could inspire even more creative content.

Generative video could threaten Hollywood and social media platforms alike

As Deloitte showed in its “2025 Media and Entertainment Outlook,” some major studios and publishers have been exploring generative video but have been hesitant to integrate it into productions. This caution may come, in part, from a fear of undermining their premium content offerings with synthetic media, but also due to challenges from talent. The SAG/AFTRA strike of 2023 included demands for limiting the use of generative AI in productions.¹⁴ Yet, Hollywood studios are often overburdened by high production costs and may wish for generative AI to eventually reduce this burden.¹⁵

At the same time, traditional studios and streamers face greater competition for advertising dollars.¹⁶ While some Hollywood studios work to stem ad losses from a declining linear TV business and migrate their advertising businesses to connected TVs and streaming video services, many social platforms have been taking more digital advertising dollars. Advertiser spending on these platforms is showing significantly greater growth than other digital media, like streaming video services.¹⁷

Generative AI’s ability to both quickly generate content and predict which segments and individuals will engage with it appears to be transforming digital advertising.¹⁸ With simple prompts, social platforms can automatically generate thousands of ads with small variations, and then instantly test which variants perform the best.¹⁹This is enabling ad buyers to spend less on creating ads and more on testing variants that return the highest success rates, reinforcing the competitive advantage of social platforms.²⁰

Yet, social platforms will also likely see greater risks from their own generative video efforts. They may confront a further boom in the amount of video content they should deliver and manage, some of it likely treading into copyright violations or worse categories. They may risk audiences being overwhelmed by “AI slop” and a rapid devaluation of the principal currency that has fed much of social media’s growth: authenticity.

A year ago, we asked people in the United States how they felt about generative media.²¹ Sixty-four percent agreed that generative AI on social media is dangerous; 76% agree that online content creators should be transparent about when and where they use generative AI in their content; and 53% agreed that online content creators who use generative AI are not authentic.

Now, a year later, the capabilities of generative video and the amount of it on platforms have both grown considerably. Generative video appears to be quickly approaching parity with reality and could soon tread into dangerous territory, potentially to both attract regulators and empower bad actors. There are concerns of potential fraud as models enable bad actors to use AI to impersonate people.²² Along with sasquatch selfie videos come the likely influence campaigns, scams, political disinformation, and conspiratorial rantings. This could even impact legal proceedings if video evidence becomes untrustworthy. Yet, without regulatory oversight, audience exodus, or punitive damages, platforms are typically unincentivized to rein it in.