AI May Not Take Your Job, But Someone Using AI Likely Will — Here's Why. Artificial intelligence is becoming ubiquitous across marketing and public relations agencies. These tools can increase productivity, but there are risks to consider.

Artificial intelligence is changing the way marketing and PR agencies are run.

Research shows that 68% of public relations professionals use AI, 68% of SEO experts are exploring it, and 61.4% of marketers are using it. It's easy to see why.

AI usage in social media is expected to grow at a compound annual growth rate (CAGR) of 28.04% through 2028, and large language models increasingly write online search results. We even have AI influencers taking over, with a reported 34% of American consumers putting their trust in them, and the highest-paid virtual influencer, Lu Do Magalu, is earning an estimated $16.2 million per year with sponsorships from brands like Adidas and Mac Cosmetics.

It's clear that AI is here to stay, and it's making a huge difference for companies in various ways, especially when creating online content. This is rapidly changing the tasks that marketers, SEOs, and PR professionals perform while augmenting their skill sets with powerful (albeit somewhat inaccurate) tools.

If you want to build a successful agency, the traditional advice is to hyper-focus on one niche and be the best at it to generate revenue and grow your team. Using AI, however, a single person can do the work of an entire team, and many solopreneurs are using it to diversify their offerings as the technology creates dramatic shifts in legacy agency models.

Related: How to Own Your Online Narrative – Even When the Internet Owns You

Exploring AI tools in marketing

There's no shortage of AI tools built for agencies–transcription tools like Otter AI and Descript, for example, make it easy to record and transcribe client meetings. These transcriptions make it easier to assign work to freelancers and ensure they have the client's exact wording and specifications.

AI image-generation tools like Midjourney and Adobe Firefly can reduce operational costs while ensuring you have high-quality images for client social media, marketing, ads, and other channels. Runway ML takes visuals a step further by transforming text and images into video, and ElevenLabs provides AI-generated voiceovers.

Finally, large language models like Perplexity make online research much faster and more efficient. Of course, there are also risks inherent in using AI in your agency.

Related: 10 AI Tools That You Should Be Using In Your Business This Year

Overcoming risks of using AI

Expect clients to have strict AI requirements, as a recent Cisco survey found that 27% of companies fully ban AI tools, while a majority limit employee use. This is because generative AI sparked a lot of controversy over the past two years, and there are various ethical considerations, including amplified biases and potential copyright infringement.

Last month, a study from Gary Marcus and Reid Southen showed Midjourney v6 reproduces almost exact movie stills, and both Wacom and Magic: The Gathering drew heat online for using generative AI images in their advertising, and the latter fired their agency after the snafu.

Related: What Will It Take to Build a Truly Ethical AI? These 3 Tips Can Help.

Making matters worse, AI image detectors like Hive and text detectors like Originality AI produce false positives, blurring the lines between human and machine-generated content. Even if you pledge not to use AI, the freelancers you hire may still introduce it into your workflow, causing unexpected problems.

If you do use AI, there are four steps you can take to ensure it's being used safely and correctly.

1. Don't publish AI outputs as your own

Raw AI outputs should never be used as the final published product, whether images or text. They aren't legally protectable, per the US Copyright Office's AI guidelines, and claiming AI outputs are your own is plagiarism. Since it's impossible to know what content most models were trained on, it's best to use AI either in the ideation phase or as one small piece of a finished product.

2. Be transparent about using AI

If you use AI in any way, be upfront with clients and the public. Having AI usage laid out plainly in your contracts and any content using it is best. Being caught using AI without disclosing it can lead to drastic negative consequences, like this indie book cover contest that was canceled after the artist lied about the use of AI against the rules of the contest. If a publication doesn't want AI pitches or editorials, be sure to honor those requests.

3. Be careful with proprietary data

Understand that any information input into an AI may potentially be trained on or viewed by an outside party. Several instances were reported in the past year of ChatGPT leaking sensitive information to other users, and it's important to be mindful of what information is entered into any AI tools. Don't input anything into a prompt that you don't want anyone to see.

4. Keep a human in the loop

It's easy to get lazy and submit mistakes you normally wouldn't when using AI. Channel Nine in Australia recently learned this the hard way when an image of politician Georgie Purcell was edited with Adobe Firefly and drew accusations of sexism. This technology isn't a replacement for humans–it's an augmentation. Be sure to thoroughly check any AI work before turning it in and making mistakes that ruin the reputation of your agency or client.

Related: 3 Smart Things to Know Before Getting Started with AI

Getting started with AI in your agency

AI can be a game-changer, and staying on top of technological advancements is important. It can give you a competitive edge over larger firms with a longer staff list. However, it requires purpose and care to implement correctly, and the consequences for misusing AI can be dire.

More than anything, honesty is the best policy. Companies are continuing to implement internal rules about the use of AI, and a subsection of consumers are always looking to catch a company using it without disclosure. As long as you properly label where AI is used in your workflow, the benefits and the efficiency that come out of it can be worth it.