🚨 Salesforce is learning a hard lesson about AI hype vs. reality. After aggressively deploying AI agents and reducing its customer support headcount from ~9,000 to ~5,000 (with CEO Marc Benioff famously noting they "need less heads"), the company is now facing reliability and trust issues with its AI systems. Reports suggest executives have admitted they were "more confident than they should have been" about how ready large language models were for complex, customer-facing work. Experienced employees bring context, judgment, empathy, and accountability that pure AI still struggles to replicate consistently, especially at enterprise scale. Hallucinations, "drift," and limitations with multi-step instructions have apparently forced a rethink. This isn't an anti-AI take. AI is incredibly powerful for automation, pattern recognition, and scaling routine tasks. But the rush to replace humans entirely (rather than augment them) is exposing the limits of current tech in high-stakes environments. Key takeaway for leaders: ✅ Augment, don't just automate. The smartest companies are using AI to make their best people even more effective. ✅ Trust is everything. In customer service, sales, and operations, reliability and human oversight still win. ✅ Institutional knowledge matters. Cutting too deep into experienced talent can create hidden costs that show up later in CSAT scores, churn, and brand perception. What do you think? Is your company seeing similar friction with AI deployment in customer-facing roles? Or are you finding the right balance between human + AI? Would love to hear real experiences from the trenches. #AI #Salesforce #Agentforce #Leadership #CustomerExperience #FutureOfWork