Snapchat and Meta Boost Ad Automation with New AI Agents

Snapchat and Meta both moved to expand AI agent capabilities for advertisers this week, with Snapchat’s June 21 ad platform update and Meta’s broadened Business Agent Platform arriving within days of each other. Together they mark a practical shift in how brands can automate campaign execution across major social platforms. The tools on offer now go well beyond content generation: they handle targeting logic, sales conversations and creator matching with limited human input.

Snapchat’s Leap into AI-Powered Advertising

Snapchat has introduced AI-powered tools designed to streamline its ad platform and improve campaign performance. A chatbot integrated into Snapchat Ads Manager lets advertisers describe their campaign objectives and receive guided setup.

Snapchat also updated its Dynamic Product Ads with what it describes as agentic recommendation models. According to the company, these models synthesise user behaviour, product affinity, full-funnel signals and real-time intent to surface more relevant products to Snapchat users. Alongside this, Snapchat is opening its platform to third-party AI agents via a Model Content Protocol server, allowing marketers to connect their preferred AI tooling directly to the ad stack rather than working through Snapchat’s native interface alone.

Rounding out the update, Snapchat’s new Snap Creator Network uses AI to match advertisers with creators based on audience profile, tone and campaign goals, reducing the manual work typically involved in influencer sourcing and campaign activation.

Meta’s Expanding Multi-Platform Agent Capabilities

Meta’s Business Agent Platform now lets enterprises build and deploy AI agents across WhatsApp Business Platform, Messenger and Instagram. These agents connect directly to enterprise systems including Shopify, Zendesk and Shopee and are built to handle product recommendations, sales closes, appointment booking and lead qualification autonomously. Administrators can set intervention thresholds so that agents operate without constant oversight while still routing edge cases to human review.

For individual creators on Facebook, Meta has introduced a Creator Assistant that provides guidance on content performance, audience behaviour and growth opportunities, drawing on data from the creator’s own account. The dual focus, large-scale enterprise automation on one side, individual creator support on the other, reflects Meta’s broad surface area across both B2C commerce and content publishing.

AI Agents Streamline Social Media Workflows

Beyond the platform-level announcements, a range of third-party tools is already reshaping day-to-day social media operations. Canva‘s AI features cover image generation from text prompts, automatic resizing across platform formats and AI-assisted caption writing. Opus Clip repurposes long-form video into short clips with auto-generated captions and what the platform calls “viral scores,” handling editing work that previously required dedicated post-production time.

Hootsuite’s OwlyWriter AI generates platform-optimised captions, repurposes high-performing posts and recommends posting schedules based on audience engagement patterns. Ocoya takes a trigger-based approach: RSS updates or e-commerce events can automatically fire content generation and publishing, removing routine execution from the workflow entirely. For analytics and engagement, Sprout Social’s AI Assist runs real-time sentiment analysis, categorises audience tone, flags emerging issues and drafts contextually relevant replies to inbound messages.

These tools collectively shift social media teams away from manual execution. The operational risk is fragmentation: teams running five separate AI tools without a unified data layer often find the coordination overhead offsets the time saved.

The Agentic Advertising Model in Practice

What separates the current wave of tools from earlier social automation is goal-oriented autonomy. Rather than executing a fixed instruction, these agents observe the environment, make decisions against a defined objective and act without continuous prompting. In practice that means an agent configured to improve engagement with a specific audience segment can generate multiple content variations, test them, and adjust targeting logic without a human in the loop for each step.

This direction at the infrastructure level involves open frameworks that connect advertisers with AI agents to automate audience targeting, creative development and campaign measurement via open APIs and Model Context Protocols.

Hyper-personalisation is the practical upside. An agent can generate content variations tailored to different audience micro-segments simultaneously, for instance, emphasising a product’s environmental credentials for one segment while foregrounding price for another. How reliably current models execute that kind of nuanced variation at scale is harder to verify from public material alone.

Oversight, Brand Risk and Tool Selection

The market for AI social media tools spans a wide capability range. Some platforms, Blaze.ai and Predis.ai among them, are built AI-first, with native text, image and video generation. Others are established scheduling tools that have added AI writing features without rearchitecting the underlying platform. The distinction matters when evaluating what a tool can actually automate versus what still requires manual handling.

Brand risk is the persistent concern. Agents operating with high autonomy can drift from approved messaging, misread audience context or generate content that is technically on-brief but tonally wrong. Meta’s intervention threshold model addresses this directly: administrators define the conditions under which an agent must escalate to a human reviewer rather than act. That control layer is likely to become a standard procurement requirement as agent autonomy increases. For a detailed look at how multi-agent orchestration is being governed at enterprise scale, see how Cognizant and KPMG are approaching multi-agent deployment.

LinkedIn has indicated it is actively working to reduce the visibility of generic AI-generated content, reflecting a broader concern that high-volume agent output will degrade platform quality if left unchecked. For enterprises evaluating these tools, the practical question is not whether to adopt AI agents in social media operations, but how to configure the oversight layer so that speed gains do not come at the cost of brand consistency. For more analysis on enterprise AI strategy, visit our Enterprise AI section.