AI agents ate the jobs of 4000 experienced employees, this company laid off its staff

Salesforce Layoffs: Last September, Salesforce (software company) laid off about 4,000 experienced employees as part of a major strategic shift to integrate agentic AI into core operations. Marc Benioff, CEO of Salesforce, told during a podcast that through the use of AI, he has reduced his support staff from 9000 to 5000 employees (about 4,000 jobs).

Read :- Amazon Created Stir: More than a thousand employees of the company wrote an open letter to the CEO, saying- AI will destroy our future.

According to reports, the company is now regretting its decision to replace its staff with AI agents. It is reducing its reliance on large language models after facing reliability issues. Salesforce is said to have faced several serious technical challenges with large language models during real-world applications, such as models skipping instructions if given more than 8 instructions. Meanwhile, home security company Vivint, which uses AgentForce to handle customer support for 2.5 million customers, experienced these reliability problems firsthand. It was also observed that when AI agents were asked useless questions, they deviated from their main goals.

It seems that Salesforce CEO Benioff’s AI dreams are now clashing with market reality, as the CEO previously suggested that the company might change its name to “Agentforce.” The company is placing great confidence in the AI ​​system’s ability to completely replace human decision-making, especially in difficult customer service situations.

According to the Times of India, a spokesperson for Salesforce said in a statement, “While LLMs are amazing, they alone can’t run your business. Companies need to connect AI to the right data, business logic, and governance to transform the raw information they get from LLMs into reliable, predictable outcomes. That’s why we built Agentforce: trusted AI infrastructure that delivers real business value. We’re passionate about AI.” The guidelines and framework are optimized to deliver enterprise-grade reliability.”

Comments are closed.