Agentic AI Upends Contact Centers as Salesforce Halves Support

Salesforce has sent shockwaves through the global customer experience (CX) industry after Co-founder Marc Benioff revealed that the company had cut its customer service workforce by nearly half.

Speaking on “The Logan Bartlett Show,” Benioff disclosed that Salesforce now employs only 5,000 customer service agents — down from 9,000 just months ago.

The drastic cut followed Salesforce’s deployment of an Omni Channel Supervisor that splits customer service equally between humans and AI bots. Agentforce, the company’s AI platform, handled about 1.5 million conversations. Human agents managed roughly the same number.

Benioff said he found it “stunning” that customer satisfaction scores were nearly identical for AI-driven and human-handled conversations. Today, he added, half of all Salesforce customer interactions are managed by AI systems, while humans take care of the rest.

This move has made it clear for the first time that Agentic AI is directly replacing jobs in contact centers.

There is nothing surprising about this, according to Eddie Irvin, an AI solutions builder based in Nashville, Tennessee.

“It is 100% true that agentic is reducing roles in customer service,” Irvin said.

The pressure is visible across the entire CX industry, according to Wayne Butterfield, partner and global lead for contact center transformation at ISG.

“We are already seeing pricing decreases across most departments that BPOs are prevalent in of between 20% to 50%,” he noted.

“Although we are yet to see many of these capabilities in production, the BPOs are betting big on the role of AI Agents in current pricing negotiations with old and new clients.”

Asked if contact centers are indeed reducing headcount, Butterfield replied, “both YES and NO.”

The Klarna’s Story

Butterfield said that correctly deployed AI would require fewer human workers, but if implemented incorrectly, any reduction in staff would be offset by worsening KPIs and possible rehiring, as had been seen with Klarna and other companies in recent months.

The Nordic fintech Klarna, which had cut jobs after bringing in AI tools, eventually reversed its decision.

But today Salesforce isn’t alone in this shift. A day after its announcement, Australian collaboration software provider Atlassian laid off over 350 customer support staff, mainly in Europe, after deploying its own Agentic AI platform.

Back in March, broadcaster Sky replaced 2,000 call center workers with chatbots, arguing that customers were ‘tired of speaking with human’ agents.

At Salesforce, the Agentic AI even makes calls on its own. For decades, Benioff said, the company failed to respond to more than 100 million leads due to staffing limits.

“We now have an agentic sales that is calling back every person that contacts us,” Benioff said.

He went so far as to compare the AI tool to Tesla’s self-driving technology.

Long Term Future of AI CX Sector

“Yes, the immediate effect may require less BPO people to do the same job,” said Arie Brish, former professor at St. Edward’s University in Austin, Texas.

But Brish argued that AI adoption could ultimately create more jobs over time. He pointed to Amazon as proof.

“The online retailer adopted an AI tool to predict customers’ preferences. This led to more jobs with the suppliers that make all these products, more jobs at the Amazon warehouses, and more jobs for Amazon drivers,” he said.

Brish, author of Lay an Egg and Make Chicken Soup, noted that automation has replaced human labor since the Industrial Revolution.

However, new jobs in customer service may not keep pace with the roles being automated, Butterfield warned.

“There are certainly new areas of opportunity as a result of the new AI workforce being deployed; however, the roles needed to build out AI Agents are significantly less than those impacted by AI Agents,” he said.

Butterfield noted that the deployment of an AI Agent was more IT-focused than the no-code or low-code rollout of conversational chatbots, which also limited the redeployment of displaced contact center workers into those potentially complex IT roles.