AI has been a fixture in the ad tech ecosystem for quite some time.
In ad tech, artificial intelligence is used for predictive modeling and real-time campaign optimization. DSPs use the technology to determine winning bids, and the sell-side uses AI to organize and tag inventory.
However, these days, there’s a new kid on the block that has the streets watching. Generative AI took the ad tech world by storm, but is the new IT girl a fad that will fade away in a few years, or does it have the staying power to impact the future of the advertising ecosystem? Experts have mixed opinions.
For example, DMEXCO conducted a study surveying 500 industry professionals, and nearly three-quarters of them stated they found generative AI useless for their business.
“Around one in ten of those surveyed in marketing, communication, and media may already be testing AI in terms of content production and search engine optimization, but most communicators are still cautious or even skeptical about actually using AI in their companies,” a DMEXCO representative said.
On the other hand, publishers and tech platforms are adopting generative AI tools such as ChatGPT into their systems. Microsoft, Meta, and Snapchat have integrated generative AI technology into their business on a large scale. Publishers like Insider and Bloomberg are using or testing generative AI in their newsrooms.
What Is Generative AI?
Generative AI does as its name suggests; it generates content. The content ranges from images, videos, text, audio, and other mediums. This subtype of artificial intelligence could disrupt every aspect of the advertising industry, from brainstorming to copyrighting to targeting ads.
ChatGPT is the most popular form of generative AI. The technology takes text-based input and generates text-based outputs. Users can ask questions such as “What is the largest monument in the world?” or “How can I start understanding the programmatic supply chain?” The tech even allows the user to be creative. For example, for a recent AdMonsters event, we asked ChatGPT to write a song about ad tech in the style of a Beyonce song. Here are the results:
I’m in the ad tech game, ain’t no shame
I’m the queen of targeting; it’s my claim to fame
I got the data; I got the insight
I know your interests; I know what you like
I’m in the ad tech game, making waves
Driving engagement, like it’s Beyoncé’s stage
I’m always innovating, never slowing down
I’m the queen of ad tech, wearing my crown
While the song won’t be winning any Grammys, the results show that the technology is intuitive enough to pick up some of Beyonce’s signature cadences while also sprinkling in knowledge about ad tech.
Critics have also accused the tech of stealing jobs. With its ability to generate content, some have tested its skills. Actor Ryan Reynolds recently created an ad for his company, MintMobile, where the script was entirely written by ChatGPT.
Despite some industry skepticism, publishers and agencies are integrating the tech. Microsoft is one of generative AI’s most prominent investors. They invested $1 billion into OpenAI, the company that created ChatGPT. ChatGPT is now incorporated into the Bing search engine, Office apps — Word, PowerPoint, and Outlook — and its Azure Cloud Service. Even ad tech vendor Admiral integrated ChatGPT into its Visitor Relationship Management platform to automate visitor engagement and generate publisher revenue.
What Are the Concerns With Generative AI?
The list of benefits for the tech is long, but industry experts have also raised concerns.
The most pressing issue is the spread of misinformation. OpenAI even posted a disclaimer on its website. It reads: “All the information on this website is published in good faith and for general information purposes only. ChatGpt Connect does not guarantee this information’s completeness, reliability, and accuracy.”
The tech is only as good as the information it receives. If the data aggregated into the tech is false, so will the output. Some businesses have already banned posting answers generated by ChatGPT. Stack Overflow, a platform used to generate questions, prohibits users from posting answers created by ChatGPT because of the model’s low accuracy rate. Often these answers have the appearance of truth, so it is hard to falsify them.
In addition, generative AI could open new questions about data privacy. Some technologists are worried about how AI will disrupt targeted advertising once companies can upload their data to a model’s neural network. From the consumer’s perspective, how will they feel about artificial intelligence having access to personal data?
There are efforts to regulate the tech. In March, The White House and the European Commission had their first official meeting to discuss the development of AI tech and how they can regulate it. The European Union created the Artificial Intelligence Act in 2021 for that very purpose; the emergence of generative AI upended those regulations.
“There are potentially real benefits from large language models and applying those to various social problems, specifically writing more secure code to make sense of a lot of the data and insights out there,” a senior administration official from the White House said. “We want our economy to get those benefits, but there’s also real worry about it. The same vulnerability analysis one might do for cybersecurity; one might also do to find vulnerabilities to hack.”
Generative AI and the Advertising Ecosystem
We’ve already heard claims about how generative AI will vastly change the advertising ecosystem, but how?
Customer Targeting: According to AI experts, generative AI can help brands target customers more directly and effectively. AI has the potential to sift through large data sets and find hidden patterns. Machine learning can highlight which customers are most likely to make a purchase based on their demographics and past purchase history.
Personalization: There are also claims that generative AI can help produce marketing assets and web content that is more personalized to the consumer. If brands aggregate the users’ data into the tech, it can help determine if an ad or web content aligns with a consumer’s interest.
Testing: The tech can also help you measure the performance of a marketing campaign. You can give generative AI access to your company’s ad performance data, and it will generate answers for what you can do next. Experts warn that human intervention is necessary for this part of the process.