Social media marketing is big business. In addition to the intangible value offered by earned and owned media, advertising revenue across social networks worldwide amounted to $25.14 billion in 2015 and is projected to grow to $41 billion this year.

But bringing everything together into one social media management has always been a challenge.

Today Sysomos has announced the Sysomos Platform, which combines its full suite of social marketing and analytics capabilities, along with artificial intelligence to help marketers make the right moves.

The new platform unifies the company’s tools for search, discovery, listening, publishing, engagement, and analysis. Using AI, it lets marketers take action in real time and measure the effect of those actions in a single user interface. Think of it as a “social cloud,” if you will. The platform also offers role-based customization so that users can design an experience to match their needs.

So what is Sysomos proposing to do with AI that isn’t already handled by regular analytics and statistics?

“Sysomos’ AI uses intelligence to proactively deliver actionable insights,” Peter Heffring, CEO of Sysomos, told VentureBeat. “Instead of marketers creating rules to find what they want and get results, Sysomos’ AI will spot any anomaly, affinity, or association in the data, but then put in the proper context for the marketer.”

That’s important because it helps the marketer or social media manager see messages and interactions they might otherwise miss. And that’s just the beginning for Sysomos as it looks to enhance the platform further with additional AI capabilities.

“In the future, our AI will integrate and process multiple data sources beyond social and campaign data and use that to help identify cause-and-effect relationships that deliver automated actions based on a company’s business objectives,” Heffring said.

I’ve seen the application at work. In addition to using AI to identify engagements, it has image recognition that can help you determine secondary trends. For example, it might be that users are generating content that includes fashion, which they’re pairing with photos of your products or services. That helps you to identify affinities and assists with future campaigns.

The system isn’t perfect yet, like most image identification technologies I’ve seen. So how does Sysomos deal with the currently inevitable identification issues?

“This is Sysomos’ proprietary technology, and we constantly tune the model,” Heffring said. “When we have a specific request from a client, such as to focus on cars or shoes, we will train [the system] with more objects, and more varieties of those objects. For logo recognition, we are typically able to achieve an accuracy of up to 85 percent. For object recognition, depending on the object, accuracy typically is in the 60-80 percent range. The accuracy can be improved through further training by boxing or masking the objects to pinpoint what’s being detected.”

While the AI is helping at the moment, and the company’s plans sound appealing, Heffring doesn’t believe that AI will completely take over, at least not yet. After all, that would somewhat negate the “social” in “social media.”

“Marketers will always want humans involved, but for many of the common product, brand, and service questions, AI will take over those responses,” Heffring said. “However, there will be limits in the near term. When a customer is expressing emotions and speaking in nuanced language — like through emoji, GIFs, and similar media —  AI won’t cut it right now. That will change, though, as automated responses from marketers will soon be able to pass the Turing test.”

And AI has benefits beyond what the company’s customers see.

“AI also will be invaluable to aggregate insights from human interactions to funnel back to product development, customer service, marketing, and other departments,” Heffring said.

Ultimately, the combination of a unified platform for social media management, along with AI-powered advice and analytics, can help businesses interact at scale without losing the human touch.

“As AI takes over, these corporate interactions won’t be social in any meaningful respect, and that’s fine,” Heffring said. “Social media has always been about people interacting with each other. Consumers won’t care how automated the interactions from businesses are, as long as those interactions address consumers’ needs.”

The Sysomos Platform is available from today.