AI is a hot topic, especially in AdTech. But how can AI be effectively leveraged in our industry? What does AI truly mean, and why is it generating so much excitement? We asked our ML engineer Richard Sundqvist some of these questions - read the full Q&A exploring how AI is transforming the advertising industry below: 

Why do you think AI is such a hot topic at the moment? ‍

AI has been evolving for years, with multiple 'mini revolutions’ pushing it forward. The latest leap, driven by OpenAI and ChatGPT, has taken AI to a new level with human-like interactions. This opens up new opportunities to integrate AI more deeply into organizations, enhancing workflows and enables a broad spectrum of roles to integrate AI in their daily tasks.

Even if AI has long been a crucial tool in Adtech, users now expect platforms to leverage it in more direct, user-friendly ways. What was once a background feature is now at the forefront, transforming not just technology but the entire AdTech landscape.

In the Livewrapped platform for example there is so much happening in a short amount of time, where AI is the only force that can make detailed decisions in that scale in real time - directly having an impact on performance and revenue in the platform for our clients.

For example, AI powers key features like S2S Optimization, which filters out traffic unlikely to generate bids, and our floor price optimization, which predicts and sets the optimal floor price in real-time. These optimizations are impossible to achieve at this scale, detail, and in real time manually, but AI makes it all possible.

What is your focus in 2024 at Livewrapped when it comes to AI? 

We focus a lot on the smartness on the platform and the value it brings to our clients and their business. The more we can automate the better. Currently we are working on optimizing lazy load settings and timeouts. Even if this is currently possible to do manually it can only be done on a higher level and settings can quickly become outdated. With AI we can finetune these settings and consider more factors like geo, device and current internet connection, plus make sure the settings stay relevant and current.  

What is AI? 

Artificial intelligence is a broad concept, basically anything that is not done by a human can be considered AI. Nowadays, we even consider statistical methods as part of AI. But in order to effectively harness AI’s potential we need to understand different sub groups of AI and how they can be applied to solve different challenges. 

So, what are the different subgroups in AI? 

There’s Machine Learning (ML), Deep Learning (DL), Large Language Models (LLM), and Generative AI—each with unique applications in adtech.

At Livewrapped, we focus on predictive AI, a subset of ML that forecasts outcomes, like predicting if a buyer will bid on a request. ML in general involves complex statistical optimizations that require significant computing power impossible for a human to do manually. DL, a subset of ML, uses neural networks to recognize patterns. In contrast, Generative AI, with tools like ChatGPT or Midjourney, creates content that’s open to interpretation, showcasing AI's creative potential.

About Richard Sundqvist‍

Richard is a problem-solver, specializing in challenges that can’t be tackled by conventional methods. He started in software engineering at Chalmers University, where he discovered his passion for AI, earning his degree in 2018. Originally from Stockholm, Richard recalls losing a bike per year during his first three years in Gothenburg. Despite these setbacks, he still precedes on the west coast with a continued passion for AI and he has been an important part of the Livewrapped team since the beginning of 2023. 

‍