Yet despite the promise of a revolution in how we interact with services and companies online, progress has been utterly miserable - the vast majority of chatbots are gimmicky, pointless or just flat out broken.
But this week I was given great cause for optimism, in the form of Alec Jones, a 14-year-old from Victoria, Canada.
For the past six months, Alec been working on Christopher Bot, a chatbot that helps students keep track of homework they've been given over the course of a week.
To set things up, a student shares his or her schedule with Christopher Bot, and from then on it will send a quick message at the end of each lesson asking if any homework had been set.
"Do you have homework for maths?" it asked 30-year-old me pretending to be a child for the sake of this piece.
"Yes!" I replied.
"Your teacher needs to chill out on the homework," came the auto-response, adding, "what homework do you have?"
"Ok, got it."
Through this interface, I'm able easily insert "algebra" - urgh - into a weekly schedule that I can then refer back to at any point to see what I need to get done.
Once I complete a piece of homework, I tell Christopher Bot, and it congratulates me, automatically removing the homework from my list of things to do. The best bit? The bot keeps quiet during the holidays.
What makes me so impressed by this is that, of all the experiments I've seen so far, it is the first time a chatbot has genuinely been the best way to tackle a problem.
Other chatbots are a lesser experience of something else. The CNN news chatbot, for example, is worse at giving you the news than any of CNN’s other products.
And popular weather bot Poncho, while cute and well-branded, has a habit of telling me it's about to rain five minutes after water started falling on my head.
But Christopher Bot shows the potential for producing a service that is completely at home within chat - removing the need for students to access some extra tool to keep track of what needs doing, and interacting in a way that (slightly) lessens the unavoidable chore of homework.
"I wanted it to not just sound like a robot," Alec told me.
"I wanted it to sound kind of like my friends would. If you get homework, everyone always just shakes their heads and says 'that sucks'."
And it does this within an app his friends are already likely using (though perhaps Snapchat would be a more useful place for it, one day).
In short, it's a product those companies banking on chatbots being a winner should seek to emulate.
It's extremely difficult, for now, to measure the success of chatbots. Ad industry magazine AdAge noted that: "Bot analytics and bot-building software companies all have shortcomings, largely because this technology is in its infancy.
"Few benchmarks exist, especially when trying to compare data across platforms."
So without data, we can't say what's working just yet - though there are some clues to what isn't.
Google's AI-powered messaging app Allo, since being launched to much fanfare last year, has failed to make even a minor dent in a messaging app market dominated by Whatsapp and Facebook Messenger.
And that's because there's no compelling reason to bother with Allo. None of its features - like asking it for directions - provide enough of a benefit beyond what you'd get from just tapping in your request the "old fashioned" way. Users have an incredibly short fuse for chatbots not working exactly as we expect.
Most big companies are missing the point, Alec told me. "There are a lot of chatbots made by these big companies that are supposed to help you interact with their service more and give you more functionality," he said.
"But it feels like they just saw this new platform, bots, and thought 'oh that's cool, people are looking at these now, let's build a bot'.
"It feels like they've just made a compromised version of what they're actually trying to build."
"You're solving a problem many students have," read one reply.
"Fellow 14 year old here," began another. "Great job man! That's sick that you’re my age and made such a cool and useful product. Awesome!"
Like any good developer, Alec has aspirations to build on the what he’s made - he wants to make it work for people in the working world, too.
But first he feels Facebook and others must do more to prove the usefuless of chatbots to people.
"I think that the real problem is that not enough people on Facebook who aren't 'techies' don't know what a bot is, and then they don't use it. More people need to know what a bot is," he said.
When Mark Zuckerberg took to the stage in front of his developers last year, he said he was opening up Messenger so that anybody could make great apps. I bet he didn’t think it would be a 14-year-old who would show him how it’s done.