Artificial intelligence (AI) is suddenly everywhere, promising self driving cars, medical breakthroughs, and entirely new ways of working. But how do you separate the hype from reality, and how can your business apply AI to solve real-world business problems in the near-term?
In the short term, AI does the math faster, saving money by automating normally complex processes. It makes your life easier even now, behind the scenes, like your Netflix recommendation engine or Salesforce Einstein’s natural language processing that scans customer emails.
In the longer term, though, AI will transform industries. For example, algorithms help healthcare professionals more accurately recognize anomalies or patterns in medical images. AI will help redesign the entire shopping experience, optimizing everything with more and better data.
Where do you start in your business?
Five ways to look past the shiny-object phase and into practical AI planning:
- Don’t fear the robots. The idea is to augment, not replace, work. AI can absorb cognitive drudgery, like turning data points into visual charts, calculating complex math formulas, or summarizing the financial news of the day into a single report. This frees up people to focus on acting on the insights.
- Start with the problem, not the solution. Before launching an AI program, identify concrete business problems, then consider if AI can help. For example, rather than ask, “What can we use AI for?”, think, “Where could we make our operations more efficient?” or “What decisions are we making without data?”
- Emphasize empathy. The more machines we employ, the more people skills we need. Leaders must build empathy across the organization to help employees see impact. Focus on how AI can help workers add more human value, rather than replace them. For example, McDonald’s added robots to their franchises, but doesn’t plan to cut human jobs.
- Engage the skeptics. Understand what they fear and start there. Fast Forward Labs’ Hilary Mason shared an example of winning buy-in by demonstrating how machine learning could solve a problem for an overburdened regulatory team.
- Remember: It’s not magic. If a vendor can’t explain their AI product or service in terms you understand, don’t buy it. Much of what’s called AI today (“AI personal assistants,” anyone?) is actually humans wrangling a trove of data behind the scenes. If it doesn’t make sense, it might not be real.
