Artificial intelligence has made it easier for brands to produce more social media content. That part is no longer surprising.
A marketing team can generate scripts, resize videos, create dozens of ad variations, and test new concepts without waiting weeks for production. The tempting response is to keep making more.
TikTok’s latest research suggests that this may be the wrong goal.
A new report produced with marketing intelligence company WARC argues that relevance matters more than sheer creative volume. Brands do not necessarily win by publishing the most AI-generated content. They win by understanding what their audience cares about while those interests are still taking shape.
TikTok and WARC Examine How Marketers Use AI
The TikTok AI marketing report draws on feedback from 400 marketers across the United States, the United Kingdom, Australia, and Brazil.
The research looked at how generative AI has changed advertising workflows, creative development, campaign planning, and content production. Most marketers now see AI as an important part of the creative process, particularly when they need to brainstorm ideas, analyze information, or produce campaign assets faster.
That does not mean marketers are ready to let an AI system control everything.
The report found weaker confidence in AI for tasks such as copywriting and script development. These areas still depend heavily on tone, timing, humor, emotion, and an understanding of how people actually speak.
AI can produce a usable first draft. It does not always know why one phrase feels natural and another sounds like an advertisement pushed into someone’s feed.
More Content Is Not Automatically Better Content
Generative AI has created a strange pressure inside marketing departments.
Because companies can now make more assets, teams may feel that they should make more assets. One campaign becomes 20 variations. A product image becomes several videos. One creative brief turns into an endless stream of captions, hooks, and edits.
Volume looks productive. It can also become noise.
TikTok Global Head of Creative and Brand Ads Andy Yang said the strongest brands are not simply generating the largest amount of content. They are learning more quickly from the communities they serve and responding in ways that feel natural to those audiences.
That distinction matters on TikTok, where trends can grow, mutate, and disappear before a traditional approval process reaches its final meeting.
A technically polished video may still fail because it arrived late, copied the surface of a trend, or completely missed why people found the original interesting.
AI Still Struggles With Originality and Cultural Context
The report points to one of the biggest weaknesses in AI-generated advertising: creative sameness.
Many generative AI systems rely on patterns found in existing material. They are good at producing something recognizable. Producing something genuinely surprising is harder.
This is why AI-created campaigns can start to look strangely familiar. The lighting is clean. The copy is acceptable. The format follows current platform habits. Yet nothing sticks.
Marketers surveyed for the report said that stronger cultural insights could improve the relevance of AI output. In practice, that means feeding AI more than basic demographic details such as age, location, or income.
WARC reported that 67% of marketers still rely on basic demographic data, even though 59% agreed that traditional demographic segmentation is becoming less effective.
Knowing that someone is 28 years old and lives in a major city does not explain which jokes they share, what they search for at midnight, or why a particular creator suddenly feels important to them.
AI needs better context. Otherwise, it produces an efficient version of a weak assumption.
TikTok Wants Community Intelligence to Guide AI Creativity
TikTok’s argument is that activity across its platform can give brands a more detailed picture of audience behavior.
People search, watch, comment, share, create, and shop inside the same ecosystem. Those actions can reveal more than static audience categories because they show what people are paying attention to in the moment.
The report describes this as an intelligence loop. Brands observe how communities behave, turn those signals into creative ideas, publish content, measure the response, and feed what they learn into the next campaign.
AI can speed up parts of that process. It can organize signals, identify patterns, generate variations, and summarize performance.
The machine still needs someone to decide which signals matter.
A comment section filled with the same joke may represent an opportunity. It may also mean the audience is making fun of the brand. An AI dashboard can count the mentions. A human marketer has to understand the mood.
Better Creative Briefs Could Produce Better AI Outputs
One practical lesson from the report is that AI performance often depends on the quality of the creative brief.
A vague instruction such as “make a viral TikTok ad for young consumers” gives the system very little to work with. The result will probably resemble thousands of other videos built from the same broad prompt.
A useful brief needs sharper information. It should explain the community being addressed, the behavior or conversation behind the campaign, the brand’s role in that conversation, and the reaction the creative should encourage.
WARC’s report introduces a five-step S·C·A·L·E framework designed to help marketers turn community insight into creative advantage, covering areas from briefing through measurement.
The broader point is fairly simple. AI does not rescue a lazy strategy.
It often exposes one.
TikTok Is Building More AI Tools for Advertisers
The research also fits TikTok’s larger advertising strategy.
At TikTok World 2026, the company introduced and expanded several AI-powered products for creative production, creator discovery, campaign optimization, and performance analysis.
These included Creator AI Search in TikTok One, the integration of ByteDance’s Dreamina Seedance 2.0 video model into TikTok Symphony, and a Reference to Video feature that gives advertisers greater control over the products and images appearing in generated videos.
TikTok also expanded Smart+, its automated performance advertising system. New features can collect different creative assets, select those most likely to perform, and offer AI-generated campaign summaries and optimization suggestions.
The platform clearly wants AI to sit across more of the advertising workflow.
Still, its own research makes an important admission: faster production is not the same as stronger creative work.
What the Report Means for Social Media Marketers
Brands should probably stop measuring AI success by how many assets a team can generate in a day.
A better question is whether those assets reflect something real happening inside the target community.
AI works well as a research assistant, production tool, testing engine, or editing partner. It becomes less convincing when brands treat it as a replacement for observation, taste, and human judgment.
Marketers still need to spend time inside the communities they want to reach. They need to read comments without immediately turning them into a campaign. They need to notice language, frustration, small jokes, creator behavior, and changes in how people talk about a category.
That part is slower.
It may also be where the actual advantage lives.
AI Can Scale Creativity, but It Cannot Invent Relevance on Its Own
TikTok’s message is not that brands should avoid generative AI. The company is building too many AI advertising products to make that argument.
Instead, the report frames AI as a system that becomes more valuable when brands give it better cultural and community information.
More prompts will not fix a weak understanding of the audience. More videos will not make a campaign feel timely. Automation cannot make a brand culturally aware simply because it can generate content quickly.
The brands that benefit most from AI may not be the ones using the largest number of tools.
They may be the ones that know what to tell those tools in the first place.
