Every day brings a new story (or three) about the promise and pitfalls of AI usage in business. We’ve written previously about both (promises and pitfalls), including:
How AI engine Perplexity can help you with research
Why productivity gains with AI are hit or miss at this point
Why and how your business should be using MS 365 Copilot
If you’ve read us before on this topic, you know we always try to bring reality to the situation: you SHOULD be using AI for certain tasks in order to save you time and effort, BUT you should not believe the over-hype. Right now, the AI prophets are making promises that don’t appear to be deliverable…yet. Moreover, blindly using and incorporating AI can actually cause harm to your business. Incorporating any tool into your business should be done with clear purpose, solid planning, and effective execution.
Robot Vacuums and Pizza Hut Delivery
Some recent stories illustrate two “pitfalls” of AI usage. The first of these involves a large Pizza Hut franchisee suing Pizza Hut for financial losses related (the franchisee says) to implementation of a new AI tool. There are several aspects to this case, but the important thing for us is this: an AI agent altered the Door Dash delivery process. This was supposed to increase efficiency, as it cut out the manual process of a manager choosing the best DoorDash driver. But what ended up happening is that DoorDash drivers themselves manipulated the system so that they could wait around in the store for multiple orders, thereby increasing delivery times for Pizza Hut customers. The Pizza Hut franchisee says this simple change brought significant reputational and financial damage.
The second story is funny and frightening. A man attempting to connect his DJI robot vacuum to a PlayStation controller discovered that he could see and control nearly 7,000 robot vacuums in 24 countries across the world! This included access to floor plans and to audio and video feeds.
A Few Takeaways
Alright, so what do these two seemingly unconnected stories show?
Most importantly, AI integration is not exclusively positive or neutral. Any tool implementation has real consequences, and you need to be clear-headed about this. If what is alleged in the lawsuit above is true, then a tool chosen to improve business efficiency ended up harming it.
Second, cybersecurity threats will only continue to increase as (a) more devices are internet-connected, (b) AI enables more efficient cybercrime, and (c) business owners are swamped with demands such that cybersecurity seems way down the list. You would never think you’d need to worry about the robot vacuum in your home or office. But that wide-open vulnerability could allow a modestly capable cybercrook to penetrate the rest of your network from that simple foothold.
Removing a responsible human from the AI loop has consequences. The seemingly simple decision to remove a Pizza Hut employee from the process allowed the system to be “manipulated” by Door Dash delivery folks. Of course, these delivery folks used the system to advance their own business interests, which ultimately came into conflict with Pizza Hut’s. With that human manager back in the loop, the Pizza Hut franchiser could have chosen its own preferred Door Dashers, which it had been doing for some time with success.
Broad Tips for CSRA Businesses
- Prefer people over tools. Maintain social trust by leaving humans in any AI loop and making sure that your customers see and experience your human presence.
- Integrate AI into your business with forethought and planning. You should have a clear purpose, a training process for personnel, and “sandbox” time before full implementation.
- Be paranoid about your cybersecurity. Cybercrime will be easier to conduct in the AI age. Crooks will have access to better tools, they will discover more devices that allow access to your networks, and they will continue to target poorly secured, “small fish” networks (rather than, say, large enterprise business with many IT resources).
Conclusion
AI is not magic. It’s a tool. And like every other tool in business, it can either help you gain an advantage or create entirely new headaches depending on how you use it. The businesses that benefit most from AI over the next several years will not necessarily be the ones that adopt the most tools the fastest. They’ll be the ones that implement technology thoughtfully, keep skilled humans involved, and stay realistic about cybersecurity risks.
That’s the real lesson from both the Pizza Hut lawsuit and the robot vacuum story. Efficiency gains can create unintended consequences. Small devices can create large vulnerabilities. And removing people entirely from the process can sometimes damage the very thing the technology was supposed to improve.
For CSRA business owners, the goal shouldn’t be to avoid AI. It should be to use AI intentionally, cautiously, and strategically. AI should genuinely help your employees, protect your customers, and strengthen your business rather than creating new forms of “Drama IT.”