This essay was originally given as a speech at Jasper’s Gen AI event on February 14th, 2023. You can see a clip of that speech below.
Years ago, I was in the back seat of a Lyft on my way to an appointment. It was a beautiful spring day and I was, as usual, immersed in my phone. The driver — he was a nice guy — turned back to me a couple of times and asked, “How are you?”
And I gave the universal modern response: “Good. Busy.”
And then he said something I’ll never forget:
“ For all our smarts, for all our wealth, the one thing we’ll never be able to do is add even one more second to the end of our days. Time has us beat.”
It’s an obvious statement, trite even, but it really caught me off guard that day. It made me think about powerlessness — tiny speck in the universe stuff.
Because of course he was right.
He was right except that during a handful of times in the course of human history we have had these moments, often triggered by technology, that have sparked major leaps forward in our collective capacity. Advancements that give us seconds, hours, days back in our lives. We’ll call them: Expansion events. And if we’re smart, we pour that time back into something meaningful.
One of my favorite technologists, Clay Shirky, wrote a lot about Web 2.0 and specifically Wikipedia. We take it for granted now, but the the ability to collaborate and contribute to the digital world around us brought on by the second phase of the internet created this incredible upwelling of knowledge. In 2010, right around the time that Wikipedia passed 100 million hours of human-contributed content, Shirky said:
[We are] living through the largest expansion in expressive capability in human history.
The largest expansion in human history. And HE was right.
Well, until now.
The internet removed barriers to start a business. Web 2.0 removed barriers to connection, mobile removed barriers to purchase, the cloud removed barriers to collaboration. And today, all around us, generative AI is removing barriers to creation.
Let me put it this way: Last year 15,103,281,925 words were written with AI assistance through Jasper. That’s just one year. And just one AI platform.
Do you know how many words are on all of the English language articles on Wikipedia? 4.2 billion.
Why does that matter? Is this some ego-driven word-counting project? Is the goal here to see how many words can be written in a year? No.
It matters because as Octavia Butler said: The goal in writing is to process, to think, to persist until we hit upon something great. And that takes words.
In fact, you could argue that the only thing that has ever held our ideas back has been our inability to convey them well and at scale. That and the time back to clear our heads and come up with new ones.
Well, AI is undoubtedly a transformational technology. But transformational technologies require a transformation in strategy to take hold and go well. This is important because too often in marketing we see a new technology or tactic and we rush to the lowest bar of its usage.
We see what it makes easier, not what it makes possible.
There are factors here, of course. Marketing teams are under immense pressure to do things faster and cheaper, or to gamify them to such an extent as to extract every last ounce of value out of them.
Let’s talk about social media. Social media meant that for the first time ever we didn’t have to be muted passengers in the flow of global information. We could contribute. We could connect. We could report.
Incredible things sprung out of that participatory culture: being a witness to underreported stories, connecting across boundaries, incredible things. And also some exceptional marketing.
Social paved the way for things like Adobe Creative Cloud’s perspective campaign. A campaign which encouraged countless creators to share their work from the platform over the course of years. Social created online communities of peer learning like Reddit, Linkedin groups, and more. Really, it ushered in the entire field of community management.
And then there was that time during the pandemic, when everyone was bored out of their minds. And this guy, Josh Swain, reached out to every other josh swain he could find on Facebook messenger and arranged to meet up in a field in Nebraska for a pool noodle fight.
Now, okay that last one wasn’t marketing. But you can’t know about something like that and not share it. So social media was an expansion event for marketing, but it also led us to do stuff like this:
Now, I don’t have anything against seizing opportunities, or galaxy lights for that matter. They look pretty cool. But this is not living out the promise of social media.
Let’s talk about another expansion event: mobile. Mobile phones led to the ability to access information anywhere and fundamentally changed buying behavior. It advanced the field of thinking when it came to accessibility, responsiveness, and UX design. It led to the growth of podcasts as people found new moments to consume content in their days.
But it also led to SMS spam marketing, which for the love of god, needs to stop.
The point here is not to bash on marketers. We are smart, lovely and adorable people. The point is to save as many marketers as possible from falling into the same traps and pressures that have gotten us in the past.
AI is going to make us faster. We need to make sure it also makes us better. Better because we could reinvest that time and money savings back into the work — to elevate it.
We’ve reached a turning point in marketing. I’d argue that this is bigger than social media, bigger than mobile. I believe this is on the scale of the internet itself. For AI to reach its potential — we need to change our approach.
Most marketers know what good content looks like. The creation process for truly quality stuff should look something like this:
But it cant look something like that because we have three blogs, a newsletter, a podcast episode and a zillion Twitter threads to create by Friday. So, more often it looks like this:When you short-change ideation you end up with a lot of posts like, “7 ways to use X.” When you short-change research, your content ends up shallow. When you skimp on editing, you let that shallow content through. And when you fail to optimize and repackage your content for distribution, you get less value out of each piece. Not to mention that maxed out bar in the middle is killing your teams.
AI brings that center bar down. It gives you that time back. Time that can be reinvested in the parts of the content strategy we’ve long abandoned.
But this doesn’t just happen. Odds are that if you don’t make any shifts in strategy, this time will just go back into making shallow content more quickly or dissipate in other ways. You need to be intentional about this shift. You need to invest in your team in new ways.
This is important because the big question out there right now is: will generative AI kill marketing jobs? And the answer from all of us marketing leaders out there should be “no”.
Because when you bring that composition bar down, the answer should be that you reinvest your team’s time in to the parts of the content strategy that will become more consequential. If we do this well… AI will shift the focus and expand the capacity of teams so they get off the content treadmill and into the editor’s seat. Or so they can spend more time developing story ideas, conducting first person interviews or original research, the stuff that actually makes writing good.
And the reason I say this is because unless you do this… you will lose. In a world were everyone has endless capacity to make content, you will lose without a stand-out approach.
The pendulum of marketing often swings between limitations and breakthrough years. When I first got into marketing, advertising was the name of the game and in that world the teams with the biggest budgets won. But that left out smaller businesses so content marketing emerged as a counter to that limitation. With content marketing, any team with the capacity to create content could organically pull people into their websites without budget. But the proliferation of formats and channels started to burn-out teams. Today, generative AI is emerging as a foil to that.
Hopefully, in this age, with capacity and budget no longer a limitation, the best ideas will win.
Again though, this won’t simply happen. Good ideas aren’t stumbled upon. They are worked for and earned. They are invested in. We’ve spent the last decade putting all our focus and resources into content composition.
Let’s spend a little time talking about what we mean by investing in the rest of the writing process.
When you rush idea development you end up with content like this:
When you invest in idea development, your content gets better.
This in turn makes AI better because it infuses AI-assisted content with original ideas. AI lives on patterns. It’s not great at finding new territory. Humans are.
When you rush research, your team uses stats from 2020. They search for “Quote on [topic]”. They pull from Brainyquote without reading the greater context of the quote. And they end up with similar content and commentary that is in a score of other posts across the industry.
When you invest in research. Your team has time to do original interviews and studies. And this research counters another critical flaw of AI. That of bias. Bias in AI is pervasive and not the sort of thing that can be easily turned off.
“Sometimes [bias] is used to mean “the training data was poorly sampled and therefore does not reflect the real world accurately”, and sometimes it means “the training data does reflect the real world, but the world itself is biased in ways we don’t endorse, and we don’t want AI systems to amplify these biases.”
So original research matters. Finding human stories or conducting outside research will enhance the substance of your content and get AI out of the echo-chamber. And yes, we need to invest in it.
I want to talk about the editing process. For too long, we’ve rushed editing. Time squeezed it. In most companies today, the content team has time to check for typos and broken links and little more.
When you invest in editing your team can actually fact check. It can catch shallow content, and root out biases — assuming you educate them on how to do so.
It’s trendy to talk about “prompt engineering” as the new hot job created by AI. Personally, I think that’s misguided. The role I’m most excited about having a renaissance in this era is the role of the managing editor. Editors are going to be essential in an age of AI. If you are underinvesting in it, it’s time to re-examine that. To develop your team to grow into this role.
All of this deepens the quality of AI. If left alone, without human guidance, AI outputs are just words on a page. Editors will ensure that we’re not just putting out content, we’re putting out something worth consuming.
Finally, let’s address distribution. AI can be helpful in formatting your content to be optimized for search, writing up markup schema to enable it to be pulled into snippets or Q&A boxes, but the bigger discussion point here is the ways in which search as a channel will change as a result of Google and Microsoft integrating chat into their user interface.
Search is one of the biggest levers in marketing today. The ramifications of this change are going to hit every single channel. It will make some sources more of a driver and reduce the influence of others. As I write this we’re just a couple of weeks into this announcement, and time will make us smarter.
So rather than make a bunch of predictions today, I’m going to say this: Reinvest the time your brand has been spending on the content treadmill into exploring the ways that distribution is going to be changing. If you’re heavily reliant on search, think about how to diversify your acquisition channels. Have your team run experiments and talk with customers about how their search behavior is changing. Dive into this work.
Do not rest on your laurels. Each of these areas: ideas, research, editing and distribution are about to become far more consequential to your business.
Lest any of this sounds daunting, remember that this is what marketers are good at. Everything changed when the internet came about. Everything changed after social and mobile. And guess what, we found a way forward through that change. We found growth in it with new skills and new purpose. This will be no different.
You have heard a lot of grandiose terms when it comes to AI. "Magic...transformational." And while it does have the potential to be those things. It’s also important to remember that this is just a piece of technology; a tool.
It doesn’t come to life until it is in our hands. It doesn’t become transformational until we do.
AI is going to make things easier. It’s also going to require that we step up. Before I close, I want to play something for you…
This is a composition by David Cope and his computer EMI — or Experiments in Musical Intelligence. David Cope was a celebrated composer for 30 years before he ever discovered AI. Then, in the 1980s he hit a creative wall. He felt starved for new ideas. So he started playing around with programming a computer, training it with the likes of Bach, Stravinsky and Mozart. The initial outputs were somewhat formulaic. Over time though, Cope started to infuse the results with his own human perspective, experience and judgement. The hybrid of human creativity plus AI-assistance created an outpouring of new ideas and music.
He was asked once in an interview with The Guardian about his experience making AI-assisted music and he said:
“I have spent nearly 60 years of my life composing, half of it in traditional ways and half of it using technology. To go back would be like trying to dig a hole with your fingers after the shovel has been made, or walking to Phoenix when you can use a car.”
David Cope was never worried about his computer being better than him. He saw it as the tool that it was. Nothing more. Because that was enough. When combined with human ingenuity, judgment, experience, AI as a tool is enough.
He knew how central people are to this equation. Asked whether AI would make music soul-less, he said.
“The question isn’t whether computers possess a soul. But if we possess one.”
AI has dissolved one of the longest standing human barriers. AI has opened that door. The question is: will the absence of that weight make us better or lazier?Intellectually stronger or complacent?
AI has done the impossible and given us seconds, hours, even days back.
What we do with it is on us.
That is not a technical choice. That is a human one. As human as it gets.
Shopping Jasper for a large business? Update your input to 100+ employees to meet with our team. Please use business email to meet with our team.
Shopping Jasper for a large business? Update your input to 100+ employees to meet with our team. Please use business email to meet with our team.