Mechanical finger draining fluid
Image generated with assistance from DALL-E.

Generative AI: from experimentation to integration

There was a palpable feeling of excitement, energy, and opportunity within PR and communications back in 2004 - 2005. Technology was changing at break-neck speed and it was making it possible for individuals and companies to communicate in brand-new ways. Blogs, podcasts, RSS, and wikis were like a ‘Wild West’ for communications professionals. They offered huge advantages for early adopters, but came with some risk as well.

Today, we find ourselves in a similar period of dramatic technological change and the implications for PR and communications are profound. I’m talking, of course, about the rise of large language models (LLMs), multimodal AI, and generative AI. 

When ChatGPT hit the scene in late 2022 it looked like an interesting experiment, but by March 2023 with GPT-4 it quickly became clear that the game had changed. Suddenly, an entire sea of possibilities was right at our fingertips, seemingly only limited by our imagination or the ability to craft a solid prompt.

Today, I do marketing in the healthcare industry, which is notorious for being slow to adopt new technology and generally risk-averse. As part of my job, I produce a podcast where we speak to bold thinkers and executives in the field of medicine. One C-level executive at an East Coast academic health system recently said, “Generative AI is the future. It is probably the biggest technology advancement since the birth of the Internet.”

Another health system vice president, who is not one for hyperbole, told us, “Generative AI, LLMs, all of that, is truly going to be one of the biggest inventions of our lifetimes.”

The reasons why are clear. 

First, the latest models have been trained and fine-tuned on a massive corpus of text and media. We’re talking billions of parameters, making them (in theory) more knowledgeable than any human could ever hope to be. On top of that, tools like ChatGPT, Google Bard, Midjourney, and others can generate responses to queries and produce content at insane speeds. Things that would take a human several minutes or even several hours can now be done in a matter of seconds.

When huge technology shifts like this take place, there are at least three ways to approach them. You can ignore them, after all, they say ignorance is bliss. You can fear them and dismiss them as something that will never measure up to the status quo. Or, you can embrace them, lean in, and learn everything you can about it and how it might help you do your job better.

In late March 2023, I identified nine different marketing-related use cases for generative AI that I wanted to test. My team and I set out to understand how generative AI could help with things like research and insights into our ideal customer profile (ICP), keyword research and analysis, content gap analysis, creating outlines, frameworks, and drafts of new content, generating thumb-stopping headlines, writing compelling captions for visuals and infographics, conducting on-page SEO reviews, doing key message analysis, and more.

The results speak for themselves. We’ve generated more content, in more formats, with fewer human resources doing the work, than ever before. But it goes beyond volume. Much of that content continues to perform better by a variety of measures (leads generated, organic search ranking, engagement) than most of our previous content. 

Among our biggest learnings? You have to have humans with creativity, knowledge of the business, a keen understanding of the challenge at hand, and the know-how to use the tools in order to achieve results. This will not change.

To be clear, it’s not all sunshine and roses. With generative AI there are obvious issues related to bias, ethics, plagiarism, and more that simply cannot be ignored. But, what also can’t be ignored is the change that is coming.

Over the next few years, businesses of all shapes and sizes will be able to do more work, of certain types, with far fewer people. What’s more, that work will be able to be completed faster. The notion of how long it should take to draft a press release, create organic social media copy, conduct an SEO audit, or build an end-to-end marketing campaign is going to change dramatically and the ripple effects will be significant. Entire business models will have to change, staffing models will too.

By all accounts, 2023 has been the year of experimentation with generative AI. 

2024 will be the year of integration. We will see the tools that are already deeply embedded in business, PR and communications integrate generative AI capabilities that will take human productivity to never-before-seen levels. 

Those who have experimented, tested, and learned will be far better positioned for the future than those who haven’t. At the same time, creativity and original thinking will rise in importance as the ability to produce a baseline of content continues to democratize. Personally, I’m having a blast exploring the ‘Wild West’ for the second time in my career.

After graduating from UC Berkeley, Trevor Jonas started his career in technology public relations in San Francisco during the original dot-com boom. Over the past 20+ years, he has worked with dozens of brands across multiple industries, providing PR, communications, editorial, content strategy, and social and digital marketing counsel. Currently, he leads content marketing for a Silicon Valley-based healthcare technology start-up and lives with his wife and two teenage children in Napa Valley.