When it comes to artificial intelligence, communication professionals are often filled with either excitement or apprehension. On the one hand, AI is hailed as a revolutionizing tool that can enhance efficiency by speeding up and simplifying analytical, strategy, and execution processes. On the other hand, it is viewed as a quagmire that can perpetuate or exacerbate existing biases and deceive audiences with fabricated content.
AI – Friend or Foe? Probably Both
Diving into the realm of artificial intelligence and social media, Zoetica founder Kami Huyse recently spoke with Jennifer Texada, marketer at AI fintech startup Vic.ai, about using AI responsibly and ethically.
The Importance of Determining Your Ethical Standards
Don’t look externally towards AI for ethical guidance. Rather, start by determining how you define ethics and ethical behavior, and move forward from there. “It’s organizations, it's companies that have the responsibility to make determinations on what their ethical standards should be,” says Jennifer.
Referencing the ongoing Hollywood actors’ strike and Hollywood studios' early proposal to scan and use performers’ likenesses in perpetuity without consent and compensation, she notes, “I think that what's happening in Hollywood is a great example of how not thinking about the ethical side of AI can really be the downfall of your organization.”
Understand How AI Learns
“Most of what everyone’s talking about right now is this generative AI base using a large language model,” she says. While impressive as a new technology, it’s still in its relative infancy, and using it effectively involves training it in a way most beneficial to your unique needs. “It's like when you're trying to train a toddler to speak.”
In other words, think of how a young child is taught a select number of words which they eventually string together on their own to form more complex sentences and ideas. “What the AI is doing is exactly that. It is taking all these little bits and pieces of information, and it's learning and trying to understand how words are put together,” Jennifer notes.
“With tools like ChatGPT by OpenAI, they use the internet and literature and art and music to train and then trying to put it all together in a sequence based on what it's learned.” Next, it makes predictions about “what you want it to know, so when you ask it a question, it's predicting what kind of answer you might need or want based on what it already knows.”
Kami explains that it’s also important to set parameters when using AI. “You need to tell it what the guardrails are, like ‘you can't do this,’ ‘you can't do that,’ ‘you can't have this.’” She continues, “whenever you're talking to an AI you have to remember that you can actually put limits on that AI. You can tell it what not to do.”
Adapt by Learning How to Use AI or Risk Getting Left Behind
Kami draws a parallel between the early days of social media and the current concerns around AI and the impact it will have on people’s careers.
“When social media first started we had some of the same conversations… the world was ending, all the jobs were going to be gone, there'd be nothing for communicators anymore, everybody is going to be an intern,” she says. “It sounds a lot like what we’re having today [around AI].”
”It feels like deja vu,” Jennifer adds. Back then, “there were people who weren't ready to learn [social media] and then there were people who jumped immediately into it.” She acknowledges that the communications landscape changed as a result of social media and that immense changes are happening again now, but notes that people should learn from the approach taken by those who successfully navigated adjusted in the past.
“Those people who lost jobs, they reinvented themselves, they learned the tools, they got really savvy at social media, they really learned how to do things very differently and adjusted their careers for the most part,” she says.”I think that the lesson there was the people who succeeded and survived were the people who understood the new technology and made the effort to embrace it.”
Let AI Help with Content Creation and Strategy
“The main thing I'm really focused on on the AI side of things is trying to learn how to use AI efficiently for marketing. I use it from not only the content perspective but the strategy and analytics perspectives,” Jennifer explains. “And what I'm trying to do is work within my team to understand, what are all the tools that are available and how can we use them effectively and efficiently to get as much work done as we can, and do it well, and do it accurately.”
She suggests starting by listing all the tasks you perform on a regular basis. “We've been starting by looking at all of the activities that we were doing on a day-to-day basis,” she says. “We're writing emails, we're creating social media graphics… maybe we take an e-guide and we're trying to create social media graphics from it and we're trying to write three or four emails to support it and we're trying to create a subset of it, maybe we're trying to write two or three blog posts off of it.”
After going through this process, “now we have a list of all the things we need to accomplish as a team and then we need to step back and say okay, what do we do… and what can we get [AI] to do.”
Choosing the right tools is also key. Kami uses both ChatGPT and Bard, depending on the specifics of what she needs to be done. “I'll often ask Bard, ‘What are people asking about x topic?’ and see what it comes up with. That helps me sometimes to shake loose things that I can write about on that particular topic.”
“To me, ChatGPT is a lot like a puzzle,” she says. “It's all the pieces and you're thinking like how do I put this together.” She recommends thinking about it in terms of how to assemble one piece after another: “What do I need to do first? I need to have my voice in there. Now what? Now I need to have a blog post about this topic. Okay. And here's some text from which you can pull from.”
Using the free version of ChatGPT will give you access to basic functionality, but Jennifer prefers the paid version because of the ability to use plugins. “If you've tried the free version you know that you can't get anything off the internet, you can't ingest your own data.” Although there is a cost involved, she feels it’s worth it. “When you start to use the plug-in tools, they do all kinds of little things. They do things like you can go and pull in a PDF, upload it, and have it read that PDF and summarize it for you.”
She recently discovered the Visla plugin tool and is impressed by what it can do. “If you're every week trying to create some social video so you have content [to share,] this thing will do that for you.” Kami, meanwhile, likes Voilà, the AI assistant that works on multiple browsers. “There's a free version of it, but there's also an eight-dollar-a-month version of it. And by the way, Voilà is using ChatGPT. ChatGPT is $20 a month…and so [Voilà] is cheaper than that and it's using [the GPT-4] version.”
Using AI Boosts Your Productivity
“If you think about it from the perspective of [making] my life easier, then that's where the value of AI is,” says Jennifer. Adds Kami, “The other thing that I think is interesting is that you can then spend your time creating one really rich, excellent piece of content, either a written content, or video, or whatever it is, and then take that one piece of content that you created yourself and turn it into all kinds of different pieces of content, blog posts, and social media posts.”
“Especially if you're a small team or an entrepreneur, and like I know the thing is you're always trying to create more content, like more information, more, more, more, and if you're a one-man shop or just a handful of people, that can be really daunting, but [AI] can be a solution if you really understand what you're trying to create,” says Jennifer.
Another powerful benefit of AI is its capability to analyze data. “We can now query data without [using] just keywords. Keywords are what we've been stuck with [for a long time]…. I love the idea that you can take a huge pile of data, your own data, and put it into an AI model. There are tools out there that are doing that right now,” Kami explains.
But Don’t Share Sensitive Information… and Always Check Content For Accuracy
While taking advantage of the full spectrum of AI tools currently available, remember not to share any intellectual property or other sensitive information with them. For example, Kami recommends adjusting the settings on ChatGPT to disable chat history and model training.
Additionally, review all content created by AI with a critical eye. Kami warns, “if [ChatGPT] doesn't really have the answer it'll sometimes create the answer,” producing inaccurate information about people, places, or facts that may sound reasonable but isn’t truthful.
Adds Jennifer, “That's why it's really important for human interaction to be a part of this. You can't just take something out of any of these GPT models and stick it up on your stuff. That's the ethical side of this. It needs human intervention. You need to read… what it's outputted.”
To learn more about using AI for your business and which additional tools Kami and Jennifer recommend, watch the full livestream.