Zingtree reposted this
CEO at Zingtree: Talks about #automation #aiagents #customerservice #ai, #cx, #contactcenter, #digitaltransformation, and #startups
Marc Benioff is warning us that #AI is overhyped and even pointing fingers at Microsoft. Talk about a plot twist! 🤔 #AIagents are here, and they’re here to stay. I’m against people who think, “Oh, AI is never going to work, screw this.” It’s all in the expectation setting and approach to implementing AI. I agree with Marc: AI isn’t a magic wand that will solve every problem overnight. But rather than assigning blame, let’s address this real issue. When AI vendors claim that AI will revolutionize everything instantly, they risk helping no one at all. Implementing AI isn’t like installing a universal app; it’s more akin to tailoring a bespoke suit. You need deep #customization based on the problem, use case, complexity, industry specifics, company size, legacy systems, and other factors. It’s like giving every tennis player the same training regimen. You wouldn’t have a marathon runner and a weightlifter follow identical workouts, right? They have different goals, body types, and needs. Consider a century-old insurance company with complex regulations and on-premise systems with limited APIs. Plugging in an isolated AI agent could be disastrous. The risk of non-compliance and data mishandling is enormous. Now contrast that with a modern tech startup like a food delivery app, like DoorDash. Issues or mistakes are not detrimental to their business. The real breakthrough? Pinpointing specific areas where AI can genuinely make an impact and tailoring solutions to fit. Start small. For instance, in customer support, use AI to handle routine queries like password resets or account balances, freeing up human agents for more complex issues. So yes, I can agree with Marc that AI is overhyped and like him I'm a believer in this technology. I do see that AI solutions are advanced now and will continue to advance. But making AI tech solutions work for large established enterprises is very complicated; need to take a custom approach, prioritize challenges, and implement AI solutions that give guaranteed wins without risks. — What’s your take? Are we setting ourselves up for disappointment by overhyping AI? 💬