“Artificial Intelligence” & “Big Data” Are Newest Entries to My 2017 Dictionary of Uselessness

“Artificial Intelligence” & “Big Data” Are Newest Entries to My 2017 Dictionary of Uselessness

Cross-posted on Medium

Sometimes you just can’t take it anymore.

Over the past few years, I have seen an absolutely delirious spike in the number of startups quoting “artificial intelligence” and “big data” in their pitches. This would be okay if these startups actually did something with artificial intelligence or big data, but unsurprisingly for early-stage companies, they often have neither the data nor the technology to fully capitalize on these “trends.”

Much like how the words “innovation” and “Silicon Valley” have become meaningless, AI and big data no longer say anything about a startup, but instead represent a completely vacuous description of the otherwise exciting features of a new business. These terms are no longer distinctive, and my (first) advice to founders in 2017 is to not bother touching them from here on out.

This isn’t a rant against buzzwords, per se, which in specific contexts can be quite useful. Rather, it’s a criticism of a facile thought process of what differentiates a technology-based startup. Saying you use artificial intelligence is like saying you use a networking library to build the company. These days, some level of artificial intelligence is built into every single product built with code.

Likewise with big data: every startup today is tracking their data and using it as part of the feedback loop. Some do it better than others of course, just as some teams push the AI boundaries a bit further than others. But it’s not an interesting point to start a conversation.

We are witnessing an absolutely incredible period of innovation where some of the most frontier work in artificial intelligence, data processing, computer vision, and more are available as open source libraries available with a quick pip install. It’s incredibly exciting what a little bit of coding can gain you in capabilities. But remember that this ability is not differentiating – it requires access to GitHub. Welcome to the democratization of engineering.

Instead, what I would encourage founders, marketers, and others who write the messaging of their startups is this: focus your complete and utter attention on the challenge and solution of your problem space. If you are building a product for salespeople, don’t say “we are using AI and big data to tackle sales efficiency.” Instead, make the case that the product is the best on the market at solving a problem that would encourage adoption among salespeople. Great solutions sell products, not flashy and popular words.

But don’t take this as just a criticism of founders – venture capitalists have done more than just about anyone to popularize and hype AI and big data. The number of thought leaders who have built entire careers around buzzwords is frankly mesmerizing. I can’t stop this train from leaving the station, but I can certainly try to convince people to stop buying tickets to board at the next platform.

Instead, we as an industry need to do more to read in-depth about the specific technical issues and capabilities that come with these new technologies. Big data is great, but will it actually solve a problem that an organization has? Does more data actually lead to a better result? Because the irony of statistics is that they can be remarkably effective with existing data if implemented properly. The question is often more about avoiding the answers than solving data acquisition problems.

Venture is in a tough pickle as we [struggle to avoid desperate waves]. But inventing them doesn’t do anyone any favors. Instead, it’s time to think broader and deeper and find the diamonds in the rough. That’s my goal for 2017; hopefully, it will be yours as well.

Photo by Michael Shaheen used under Creative Commons