How Generative AI Is Changing Social Media Marketing
Four years ago, Van Esch and Stewart Black (2021) posed a question that is now more urgent than ever for every marketing practitioner and business owner managing a social media presence: will AI free marketing professionals from mundane tasks to focus on higher-value activities, or will it threaten the very skills that make social media marketing effective? The research evidence that has accumulated since then provides a nuanced but actionable answer, and it is one that every social media marketer needs to understand.
This week's blog draws on four peer-reviewed studies to examine what generative AI can genuinely do for social media marketing, where it falls short, and how practitioners can use it strategically without undermining the brand authenticity that social media audiences demand.
The AI Productivity Promise
The most directly relevant study for social media practitioners this week comes from Gastmann and Bastos (2025), examining whether generative AI can produce genuine productivity gains specifically in social media marketing strategy and not just content generation, but the strategic thinking that underpins it.
Their study used a two-phase approach. First, they conducted in-depth interviews with 20 experienced social media professionals across multiple industries and company sizes, mapping out what high-quality social media strategy actually looks like in practice, covering target audience identification, platform selection, content ideation, and benchmarking. Second, they ran a follow-up experiment using ChatGPT-4 to generate complete social media marketing strategies for two real companies (a sustainable fashion brand and a chemical corporation) and evaluated the quality of the AI-generated output against the professional standards identified in the interviews.
The findings support what the researchers call the productivity gain hypothesis. The AI-generated strategies were rated as high quality in terms of brand fit, their ability to address specific business challenges, and strategic flexibility. Importantly, the study found that AI is particularly effective at rapidly synthesising contextual information and generating structured strategic frameworks and tasks that would typically require considerable time investment from a human strategist.
However, the study also identifies clear and important limitations. Off-the-shelf AI tools such as ChatGPT present significant concerns around intellectual property and data protection. When practitioners input sensitive company information and proprietary customer data, confidential business strategy, unreleased campaign plans into commercially available AI platforms, that data may be processed and retained in ways that create legal and competitive risk. The researchers flag this as one of the most pressing practical concerns for organisations adopting generative AI in their marketing workflows, and note a growing trend toward companies developing in-house AI tools that operate within controlled data environments as a direct response.
The Authenticity Problem on Social Media
Productivity gains mean little if the content that AI helps produce damages your brand. And the evidence from Brüns and Meißner (2024) establishes this risk with striking clarity. Across three experimental studies, they found that when social media followers are informed that a brand's content was created using generative AI, their attitudes toward the brand and their engagement behaviours deteriorate significantly. The mechanism is a loss of perceived brand authenticity followers feel that AI-generated content lacks the genuine human intention and creative effort that makes social media communication meaningful.
What makes this finding particularly important for practitioners is that consumers cannot reliably detect AI-generated content on their own. The negative response is triggered by disclosure, not by the quality of the content itself. As social media platforms increasingly mandate the labelling of AI-generated material, a regulatory direction that is accelerating globally, and this is not a risk that brands can manage by simply producing better AI content. It requires a strategic approach to how AI is positioned in the creative process.
The Solution: Human-AI Collaboration as a Brand Signal
The experimental research by Kim and Cho (2026) provides the most direct evidence of how to resolve this tension. Their research, involving two controlled experiments with over 400 participants, demonstrates that social media content produced through human and AI collaboration generates significantly higher engagement than fully AI-generated content. The psychological mechanism is curiosity: collaborative content creates what the researchers describe as a productive information gap, which is novel and interesting enough to engage the audience, while the human element provides the trust and credibility that sustains that engagement.
This finding has immediate practical implications for how social media teams structure their creative workflows. The goal should not be to use AI as a replacement for human creative judgement, but as an amplifier of it. A social media manager who uses AI to generate first drafts, brainstorm content angles, and rapidly prototype campaign frameworks, then applies their own creative direction, brand voice, and audience knowledge to refine and finalise is producing collaborative content. That collaborative process, when communicated transparently to the audience, becomes a positive brand signal rather than a liability.
The Bigger AI and Social Media Strategic Picture
Van Esch and Stewart Black (2021), writing at the dawn of the AI marketing era, argued that AI would fundamentally transform how organisations create content, generate leads, reduce customer acquisition costs, manage customer experiences, and convert consumers via social media. What they could not have anticipated fully was the speed at which these changes would arrive, nor the authenticity challenge that would accompany them. But their framing of the core question ‘automation versus augmentation’ has proven prescient. The evidence now consistently points toward augmentation as the more sustainable and effective model for social media marketing.
The productivity evidence from Gastmann and Bastos (2025) confirms that AI can meaningfully augment strategic work. The authenticity evidence from Brüns and Meißner (2024) and Kim and Cho (2026) confirms that full automation of creative work carries real brand risk. Together, they define the productive space for AI in social media marketing: use it to work faster and smarter at the strategic and operational level, while keeping human creativity, judgement, and voice at the centre of what your audience actually sees.
Practical AI Recommendations for Social Media Practitioners and Business Owners
The evidence from these four studies points to five concrete actions.
• Use AI for strategy and structure, not just content. Research demonstrates that generative AI can produce high-quality social media strategies, covering audience targeting, platform selection, content frameworks, and business challenge responses. Use AI to build your strategic scaffolding faster, freeing your team to focus on creative execution and audience engagement.
• Protect your data before you use AI tools. Do not input sensitive business information, proprietary customer data, or confidential campaign plans into off-the-shelf AI platforms. Either use tools with robust data protection guarantees or develop internal processes that keep sensitive information out of commercial AI systems.
• Build a human-AI collaborative workflow and make it visible. Structure your creative process so that human judgement, brand voice, and creative direction are always present in the final output. When you disclose AI involvement which you increasingly must frame it as a statement of how AI enhances your team's creativity, not a confession that humans were not involved.
• Develop a proactive AI disclosure strategy. Mandatory disclosure of AI-generated content is becoming the norm across platforms and regulatory environments. Brands that get ahead of this by building transparency into their content culture will be better positioned than those who disclose reluctantly under pressure. A short, consistent disclosure statement that positions AI as a creative tool not a replacement for human expertise is better than silence.
• Segment your social media audience by AI familiarity. Research shows that the collaboration advantage is strongest for audiences with lower AI competency. If your primary social media audience includes consumers who are less familiar with AI technology, the human element in your content and your disclosure communications matters even more. Know your audience before deciding how prominently to feature AI in your brand narrative.
References
Brüns, J. D., & Meißner, M. (2024). Do you create your content yourself? Using generative artificial intelligence for social media content creation diminishes perceived brand authenticity. Journal of Retailing and Consumer Services, 79, 103790.
Gastmann, J., & Bastos, M. (2025). Strategising with generative AI: Productivity gains in social media marketing. Journal of Marketing Communications, 1–20.
Kim, S., & Cho, E. (2026). How AI involvement affects curiosity and content engagement in AI-generated advertisements. Journal of Retailing and Consumer Services, 92, 104756.
Van Esch, P., & Stewart Black, J. (2021). Artificial intelligence (AI): Revolutionizing digital marketing. Australasian Marketing Journal, 29(3), 199–203.
