Metricool, a social media management platform, surveyed more than 700 creators and marketers and found that 98% of respondents now use AI in their work. The survey, shared exclusively with CNET, also shows usage is deepening: nearly 40% of respondents rely on four or more AI tools.

The most common strategic tasks handed to AI are strategy design (62%), trend research (61%) and metrics analysis (59%) — all up significantly from 2025. The top two use cases remain unchanged: writing copy and captions (86%) and generating content ideas (84%).

One figure stands out. The share of social media professionals using AI for graphic design or image editing jumped from 38% in 2025 to 61% now. The likely driver is tool quality. Over the past year, Adobe, Canva and Figma have integrated AI in more purposeful ways — smaller, constrained edits rather than generic generation. In Adobe and Canva, users can now restrict AI output to content that matches brand guidelines, a priority for creative teams.

"I think, for a lack of better words, [AI is] becoming better and less AI-slop-esque," said Anniston Ward, US marketing manager at Metricool. "It's allowing for less friction for designers. They're able to build in a way that's a little bit more reflective of their work, instead of a generic generative design."

From chatbot to operating layer

The survey data points to a shift in how marketers deploy AI. "People are using it more strategically rather than just bounc[ing] ideas off ChatGPT or Claude," Ward said. "They're able to use it more strategically in a deeper way. … It's going to be a core skill moving forward, to use AI thoughtfully rather than just at a very fundamental, basic level."

Matt Navarra, social media expert and author of the Geekout newsletter, describes the same trend in architectural terms. "Good social teams are increasingly building an entire AI stack," he said. Instead of asking ChatGPT for 10 caption options, teams now connect AI tools directly to their data in Google Drive. "That moves AI from being a chatbot into something much closer to an operating layer for the whole social marketing team."

Digital marketing illustrates a broader workplace pattern: workers use AI for daily chores while organizations pursue deeper integrations — handing tasks to AI agents or wiring AI tools into existing databases for better insights.

Where to draw the line

Not everything should be delegated. Navarra's advice: keep the backstage work automated, keep the judgment human. "The smarter use of AI is often backstage," he said, pointing to social listening and data analytics. "Don't outsource your taste, your judgment or personality, because AI should remove the boring bits of social media work, but not the human bits."

Strategy carries particular risk. Using AI for strategic work can produce what Navarra calls "strategy-flavored autocomplete: polished, plausible, but completely average." His argument: "Strategy isn't just data. It's about judgment, culture, timing and instinct, and also taking risks — because AI doesn't often give you the riskier options. Knowing when the statistically obvious answer is the wrong one is something that you learn from years of experience rather than asking an AI."

No rulebook

For all this adoption, formal governance is nearly absent. Only 12% of social media professionals say they have clear AI guidelines, according to the Metricool survey. Nearly half — 45% — report no formal restrictions or rules around AI usage at all.

Those guardrails matter in practice. They set expectations for what's allowed, such as banning fully AI-generated images or requiring human review checklists for AI-produced content. Privacy is another open question: agencies working with multiple clients need data privacy procedures for AI-integrated datasets.

Without formal rules or training on best practices, individual workers decide for themselves what's appropriate. In every case, the survey's experts argue, human judgment should stay in charge.