AI has no problem when it comes to hype and bluster. Implementation is a wholly different animal. Though adoption of AI is trending in the right direction, Implementing the technology is costly and may require significant cultural shifts to work. These aren't the only hurdles to AI's widespread adoption within the advertising space, either.
These industry professionals shared their thoughts on the greatest challenges to AI adoption in the advertising industry. Here's what they said:
1. Laura Mingail, SXSW Advisory Board Member in Advertising & Brand Experience
“The concept of AI can be daunting to marketers, because it powers so many solutions and many may not be immediately relevant to them. The #1 challenge to overcome is the limited awareness about the range of ways AI will help with significant improvements of their relevant KPIs. AI on its own will not solve the problems that they need solved – the growing number of solutions powered by AI will.”
2. Herve Utheza, Head of Media, Advertising & Network Operators at HERE Technologies
“The overhaul of a computing infrastructure, and a data science infrastructure and ecosystem takes time.
Upgrading a compute engine from the past methodologies to a new methodology requires a complicated process, as none of the systems supporting an AdTech system live in isolation.This is why the large publishers lead the pack, while the rest of the industry will take a more slow, systematic and reasoned approach to changing systems.
Also, this is a very new system culturally. We even see campaign managers question AI because, sometimes, AI produces campaign strategy results, which are counter-intuitive to what humans used to do. It’s hard for a campaign manager to admit — to themselves and their bosses — that they may have been wrong all along when the machine explains you should target, segment and plan differently.”
3. R.J. Talyor, founder and CEO of Pattern89
“In an eMarketer survey, 32.9% of marketers have reported that applying AI to their current roles and workflows is their biggest challenge. 30.6% say they are unclear about what AI is and how it works, and 28.5% say their budgets can’t afford it.
For those marketing AI products, it is necessary to explain that AI won’t threaten jobs, but it will change them. Incorporating AI into advertising will allow advertisers to focus on creative and strategic work, while their software completes smaller tasks and analyzes data. Educating people on what AI is and how it really works is the best way to combat these challenges.”
4. Matthew Fanelli, Senior Vice President of Digital at MNI Targeted Media
“The #1 challenge, in my opinion, is how and the pace at which, AI will be adopted by businesses and marketers across the board. Another real concern is the data that is being collected and how it will be used. With data and privacy concerns at an all-time high, being compliant and having clear rules, regulations and guidelines will be of the utmost importance. As we hear more about GDPR, CCPA and PII compliance, AI capabilities will need to comply on all fronts.”
5. 7P.K. Kannan, Dean’s Chair in Marketing Science in the Robert H. Smith School of Business at the University of Maryland
“Using AI for targeting purposes, creating personalized ad copy and communication messages and understanding their effectiveness requires analyzing a lot of individual level data that tracks customers across websites, devices and platforms using cookies, pixels and other invasive technologies. Given the privacy concerns of using such data for targeting and communicating, there is going to be pressure on using less of such data which might dampen the use of such technologies. The challenge in these times is to come up with privacy-preserving analytics to aid advertising targeting. There is work on-going in this area but still such time there will be increased scrutiny of such data usage.”
6. Luke Taylor, Founder & COO of TrafficGuard
“Complete and substantial data is the foundation of AI. The benefit of AI is in its ability to process highly dimensional data for fast, accurate and actionable insight. To collect the volumes of data required to train ML models and drive the success of AI applications takes time. The time it takes to collect this data is the limiting factor that will determine when an organization is ready to leverage AI, and the effectiveness of their AI solutions once implemented.
An additional and related challenge is data privacy due to the nature and volume of data being processed. The regulatory environment around privacy is very active and with increasing AI/ privacy literacy, consumers are also concerned. A survey conducted by the World Economic Forum found 41% of global internet users had a degree of concern around the use of AI.
Businesses leveraging AI for digital advertising need to diligently adhere to privacy regulations and ensure data is stored and processed securely.”
7. Peter Bordes, CEO of Kubient
“Training Deep Neural Networks requires performing computationally intensive operations parallelly at high speed for which graphics processing unit (GPU) architecture is most well suited. In real-time programmatic advertising, there is a very small window of 300 milliseconds wherein the response has to be generated, ensuring that the predictions are made within that small time frame. Thus it is very crucial to have the correct infrastructure to ensure accurate and quick responses.”
8. Tod Loofbourrow, Chairman and CEO of ViralGains
“Disappointingly, the biggest challenge for many seems to be finding the right line defining the ethical use of data and AI in advertising and staying on the right side of that line. Both Facebook with Cambridge Analytica and Google with YouTube’s radicalizing recommendation engine and dubious monetized content have stepped over that line a few too many times. Just because you can do something doesn’t mean you should.”
9. Steven Rothberg, founder of Job Search Site College Recruiter
“Fear. I was thinking cost, but College Recruiter's use of AI costs thousands of dollars a month and some spend even less than we do when they license AI technology from companies like Google, Microsoft, IBM, and Amazon. What's really holding a lot of companies back is the fear of the unknown. They seem to know that AI is in their future but they're afraid to investigate its value today, perhaps because they intrinsically know that if they investigate, they'll see tremendous value and if they see tremendous value, they'll have to do something about it. But deliberate ignorance was never a good business strategy.”
10. Mahi de Silva, co-founder and CEO of Amplify.ai
“Until now, advertising has been a largely one-way communication medium, with marketers pushing out their message and consumers, well, consuming it. Conversational ads, driven by Conversational AI, opens to door to two-way communication between brand and prospect, brand and customer. The makers of digital ads need to master the design of effective conversational flows, and need to find the right technology platform/partner in the process.”
11. Brandon Gains, Vice President of Marketing at MonetizeMore
“Publishers have always had editorial and sales as their key focus when it comes to staffing. The engineering department inside these publishers have typically been strapped for resources and don’t have the skills in-house to build advanced algorithmic strategies. Publishers need to shift more resources towards their engineering departments to unlock the massive revenue potential of AI monetization.”
12. Michael Harrison, Managing Partner at Winterberry Group
“Inadequate data or data quality is the number one challenge to successful AI adoption. Marketers need to have their data strategy and infrastructure in place before implementing AI. Technology is not a solution for poor data or data quality. Sparse data could cause a lack of meaningful insight or even worse, incorrect insight. Marketers are still struggling with the customer identity issue, which if AI is applied, will lead to false customer experience recommendations.”
13. Ciprian Borodescu, CEO at MorphL.io
“The accuracy of the data fed into the AI platform is an important impediment. Moreover, ad buyers fear being replaced and some are concerned by the variable cost structure of the technology. I believe increased familiarity with AI workflows will improve clarity and confidence around the process.
Leaders and laggards face different challenges to AI adoption:
Leaders strive to attract, acquire, and develop AI talent
Laggards struggle with lack of leadership support for AI initiatives or unclear business cases.
However, AI adoption has tripled in 12 months: 14% of enterprises already use AI and 23% intend to deploy it within the next year.”
14. Sean Byrnes, CEO of Outlier
“Surprisingly, the biggest challenge to AI adoption continues to be human resistance. Despite evidence that shows that AI can improve human behavior and results, making them more valuable to their organizations, many analytics teams still fear being replaced by automated systems.
In this case, some choose to do nothing and continue attempting to sift through mountains of data, ultimately missing important trends. This resistance to adoption is real and leaders need to identify and overcome it to drive the efficiencies they need to survive in the market.”
15. Kevin Groome, founder of CampaignDrive
“There’s more to adopting an AI solution for ad creation, buying, and optimization than just the technology alone. Major advertisers (or their agencies, need to build an operating model that takes the speed – and the risks – of AI implementation into account. This is one reason why, in recent surveys, AI capabilities barely made it onto the radar screen for most adtech buyers.”
16. Stephen Swift, Creative Technologist at Allen & Gerritsen
“Any discussion of ethics aside, the challenge is similar to the one faced by blockchain technology advocates: Proposing a technology that is also buzzword is difficult. As Kodak and Overstock have learned, namedropping buzzworthy tech can open doors and wallets quickly, but actually delivering value thereafter generally requires investment and wisdom far beyond what proven technology might.
Take our current-moment notion of AI as a tool to abet creative output. Off-the-shelf solutions which provide this utility are few and expensive, and getting an entire agency on board to pot-commit to an experimental tweak to process is, at best, an intimidating proposition. Developing a bespoke AI tool, on the other hand, or leveraging AI as a core creative function (for an activation involving deep learning, for example) requires having or sourcing developers experienced in cutting-edge technology—and they don't exactly grow on trees.
Some clients already know this; those who have learned this the hard way may hear the term “AI” and reflexively scowl. Even if you're capable of executing on world-class AI innovation — this, too, will be a challenge.”
Share your feedback and we'll consider adding it to the piece!
Have expert insights to add to this article?
Share your feedback and we'll consider adding it to the piece!ADD YOUR INSIGHTS