True AI is great. You can use it to predict system errors, drive cars across the country, and stay healthy, wealthy, and wise. But please keep it out of my inbox.
AI assistants are all the rage and we all have seen them by now — “Hi, I’m Sally, Joe’s AI assistant. Let’s find some times when you two can meet!”
These chipper bots are fun, but if you’re using them daily, then you might be missing out on some important opportunities.
The bottom line: AI assistants are too dumb to trust right now and they’re too dangerous to ignore. I’ve tried a few of them and here’s my take.
Unexpected Consequence 1: AI assistants may save you time, but they waste mine
It’s that simple. These tools aren’t as smart as they claim. They don’t have the human ability to interpret subtleties in our conversation. For example, if you and I already agreed on having a Skype call for our meeting, your “secretary” will continue to insist on your preferred coffee shop.
I’ve seen this again and again and I’ve missed meetings this way simply because the AI didn’t get my phone number and instead told me to go to Blue Bottle in New York, when I live in Bay Area.
The takeaway? AI needs context quickly or it will fail. And when you use an AI without understanding its limitations, you will also fail.
Unexpected Consequence 2: You miss a golden opportunity to connect
A week ago someone introduced me to a business partner because they felt it would be mutually beneficial. When I emailed the prospect, a bot answered. Why miss out on an opportunity to connect?
I replied that AI destroys a lot of interaction, and we started chatting on Skype. If a robot had continued to interfere, we would’ve never spoken and she wouldn’t be working with Bugsee right now.
Direct interactions with humans can tell you a great deal. While scheduling the meeting, we get to see whether each of us is quick or slow with their email. We get to see who replies over the weekend. We get to figure out who is quirky, weird, or funny. We also get to see if we are brief emailers or story tellers.
In short, we get to understand each other as humans. These are some lines from our back and forth:
— Oh, sorry, I must reschedule, my kid is sick.
— No problem, mine came back from school with a nasty bug last week and shut down the entire family for a few days :) How old is your kid?
— I can’t do 9am, interviewing a digital agency then. How about 10am?
— That works. Would love to hear your experience interviewing there. We’re looking into that too these days.
These touch points are invaluable in building lasting business and human relationships. Don’t skip them!
Unexpected Consequence 3: You’re making me talk to a box, and it’s unpleasant.
Remember the feeling last time you got an email that said:
“Dear Michael from Acme, Inc.”
Remember how you looked at that “Inc” part? Without reading any further, you knew it was a sales automation email that got your company name from a database. No one likes receiving emails from machines.
Similarly, no human would ever include all the junk in this line:
“Alvin likes Voyageur du Temps, 288 1st St, Los Altos, CA 94022, United States”
Why do you make me talk to a fucking robot? I get enough of that when I call Comcast, where it’s often not even obvious when you get a human. The lack of real intelligence (or compassion) on the other end of the line is usually consistent throughout the call, making it a smooth transition from an IVR to an agent. But that deserves a separate post…
Ever talked to a chat bot? Remember that feeling of banging a wall with anything that is remotely non-standard and wondering — am I not worthy of a human time of that pretentiously luxurious brand?
Oh yeah, and its pretentiously fake textbook politeness. That’s always a treat. Maybe I’m grumpy, but come on!
Unexpected Consequence 4: Implied confidentiality breach
After our initial email exchange, we might’ve been talking about something sensitive or confidential. And now, with this AI, you’ve added a 3rd party to our thread.
You and I, both know that AI bots are 90% human and 10% automatic. They have to be, as the technology just isn’t there yet.
And most importantly, I’ve never given you a consent for any of this. But it’s too late, because you’ve violated the inherit trust of private communication.
Bottom Line: What message are you sending?
When you use a dumb robot, you’re telling me that your time is more valuable than mine. If you’re indeed a busy individual, hire a professional assistant that costs real money. Otherwise, don’t fake it with a $50/month pseudo-secretary.
Also, if a bot is the one that scheduled our meeting time, there is an increased chance you will actually miss it. Again, wasting my time. Nothing beats a personal commitment.
When Hotmail was born, the founders followed Tim Draper’s suggestion to add a footnote to every email “Get your free email at Hotmail”. It worked beautifully then, but it isn’t cool anymore.
Yet, you’ve decided that it’s ok to send me emails that end with a similarly spammy message, promoting a 3rd party service and tracking my clicks. I don’t need to see “Sent by Stella.ai, my digital assistant” at the bottom of every email.
Don’t get me wrong, I understand the need to improve productivity. However, when forcing a tool on your business partners and customers, think about how it may affect them and consider the message you’re sending by doing so.
Still unconvinced? I want you to try Boomerang Calendar, which is a system that I used and is much more suitable for the scheduling function. It shares your calendar availability with your prospects.
You can set a time slot with a single click and if your schedule changes, it automatically updates your availability. There’s no chirpy AI, no weird back and forth, and your data is secure. It’s the least you can do for the people you email.