AI: Beyond the Hype

In a world that feels dominated by artificial intelligence, we find ourselves asking questions: What can't be automated? Will AI ever feel emotions? How will AI change the way agencies work? While we don't expect to answer all of these questions in one attempt, we aim to guide you toward some of the innovations and insights at dentsu that could help shed light on these thoughts. Welcome to the second edition of Noted.

We're collecting Infinity Stones

The buzz of AI has never been louder. Popping up on our LinkedIn feeds and becoming a staple in our strategy and creative meetings, the dramatic increase of AI is leading businesses to massive investments, in both the technology itself and the people harnessing it. AI's ability to streamline workflows, analyse data at speed, and create engaging content is already transforming the way we work, as well as how we think about what is possible in the future.

But let’s look beyond the hype for second. What grounds us at dentsu is that innovation is always in pursuit of the outcome, not just the output — we innovate where it makes sense and where it drives impact for the benefit of clients, our people and society at large. While AI brings many opportunities, it can also bring risk, which is why we have been busy creating a secure, private and responsible place within which our teams and clients can safely conceptualize, test and develop new solutions. And it’s already seeing incredible results...did we mention our Grand Prix winning Scrolling Therapy campaign?!

Last year marked a significant milestone with dentsu's introduction of AI tools and services from industry giants such as Microsoft, Google, Salesforce, and Adobe. Building upon this foundation, in 2024, we've extended our relationship with Amazon Web Services (AWS) by adopting two key services, Amazon Bedrock and Amazon SageMaker, into our GenAI stack. Will we see the addition of more “infinity stones” this year? You’ll have to wait and see, but in the meantime, these recent technology additions have driven new levels of innovation and opportunity for our clients, as well as helping our teams around the world to innovate quickly, think differently and scale to new heights. Read the press release.

Want to know more about dentsu’s AI strategy? You’re in luck! The latest episode on The Short Story is, features none other than our very own Brian Klochkoff and Kate Slade to talk you through it:

The sidekick to brands

Imagine welcoming new team members with a comprehensive reservoir of brand insights readily available at their fingertips. No more scouring the web for scattered information; instead, they can access a centralized platform that offers in-depth responses to their queries. From understanding company values to navigating internal processes.

Recently, we had the opportunity to speak with Thomas Walter, Chief Services Officer at Merkle Northern Europe and a key player in a team developing and utilizing AI solutions at a market level. Thomas revealed Merkle's unique approach to generative AI, which led them to build "Merkle Gen-AI", an experiment turned sales-support tool.

Merkle Gen-AI is a conversational interface that acts as a sidekick for Merkle. It's part of their online platforms and is trained with data from the past ten years of Merkle Imperatives, case studies, client references, and, notably, unique insights from 20 global offices — giving it know-how beyond what is available on the web.

Find out more about how AI is changing the way dentsu agencies like Merkle operate.

Our question to you: As more agencies harness the power of AI, what is going to set them apart? Share your thoughts in the comments.

Let AI be the judge

The rise of AI in creative agencies has sparked interesting discussions about whether it could replace human interactions between clients and teams. While people often value human intuition and judgment in client relationships, we might see changes over time, thanks to projects like "The Artificial Client" by our Dentsu Creative team in Amsterdam.

This experiment introduces three made-up client personas — the "Ruthless Critic", "Idealistic Dreamer," and "Confused Creative" — each with their own viewpoints and feedback styles. Designers can submit brand ideas, website designs, or TV scripts and get instant feedback from these AI-generated personas.

Although the Amsterdam team says the feedback is more fun than serious, the experiment shows how AI could affect client interactions in our industry. In essence, rather than simply replacing clients, AI has the potential to challenge client-agency relationships in exciting ways. Try AI Client now!

TUNA SCOPE: AI modernizing traditions

Let’s take a deep dive into the seas of innovation with a one-of-a-kind application by dentsu Japan . TUNA SCOPE is an AI-based system to determine the quality of a tuna by image analysis — a fundamental part of the valuation of catches. Through deep learning, bulks of cross-sectional images, and data on quality assessments conducted by master artisans, the TUNA SCOPE team successfully passed on this sophisticated, but rapidly disappearing skill to AI, with remarkable accuracy. 

In a world where nothing is exempt from the touch of AI, this tool is not just improving efficiency; it's redefining industry standards.

The knowledge accumulated by the AI-based technology through the development of TUNA SCOPE has untapped potential for applications across a wide variety of industries and academic disciplines. By teaching AI the secrets and wisdoms required to perform the tasks humans have passed down over the years from eye to eye, we’ve found a new ally in AI to sustain and modernize traditions. Read more.

So, what can’t we automate?

AI can execute, but the ideas still belong to us

When it comes to AI-generated imagery, a simple prompt is enough to draw a masterpiece! It's both an exciting and daunting time for creatives, as AI continues to astonish us with its ability to rapidly translate our words into visuals.

With 60% of marketers already embracing AI, brands have already started testing the waters with AI generated creative. Heinz engaged DALL·E 2 to produce images of ketchup, resulting in a variety of bottle shapes reminiscent of the iconic Heinz ketchup bottle. Since generative AI is trained on large datasets of existing images, this campaign cleverly illustrated the dominance of Heinz across the category.

As amazing as it is, this advancement in AI has led to many people becoming concerned about the blurring of lines between AI images and intellectual property.

Denis van Leeuwen, Chief Strategy Partner at dentsu Benelux, cleverly explains why this has been the consequence of AI training: "Instead of providing AI with rules to identify a cat, we show it millions of cat pictures and let it discern patterns on its own. As a result, AI lacks a structural understanding of a cat. AI matches or recreates patterns resembling the concept of a cat”.

This training model is why artists and image libraries argue that these AI models must have been trained on their data without permission to generate these new pictures. For further insights into the future of advanced AI capabilities and challenges, download our 'The Pace of Progress' dentsu's 2024 Media Trends report.

Our question to you: What's your stance on the intersection of AI image generation and copyright? Leave a comment below.

You can’t automate heart

Recent insights from Dentsu Creative's CMO Report reveal a striking reality: 

• 77% of marketers agree that AI can't create deeply moving content
• 81% of consumers are willing to pay a premium for human-created content
• 82% of clients believe in the irreplaceable value of human emotion in brand communication
• 75% assert that Generative AI can't fully replicate human creativity

These statistics collected from CMOs hightlight a key challenge for generative AI in our industry: the inability to create content that resonates with our emotions. Or, as Pats McDonald, Dentsu Creative 's Chief Strategy Officer says, "You can't automate heart!"

While AI has made creativity more efficient, the special qualities that make storytelling and brand communication powerful — like empathy and connection— are deeply tied to human experiences, rendering the human touch indispensable.

To find out more, listen to Pats McDonald on The Short Story Is!

Our question to you: How do you envision generative AI disrupting the creative process in your industry? Share your thoughts in the comments below!

“Our sensors detect that you want more” 🤖

In later editions of Noted, we will continue to explore and dissect the countless ways AI is reshaping our industry and beyond, fostering a deeper understanding and appreciation for its transformative potential....until then, here are some must-reads on AI innovation at dentsu.

• All Players Welcome: A collaborative initiative led by Dentsu Lab Tokyo (DLT), NTT Human Informatics Laboratories, and WITH ALS, showcasing how AI technology and creativity are breaking barriers and empowering people with disabilities to perform live on stage, using only the movement of their eyes. Read more.
• Merkle’s revolutionary generative AI solution: GenCX is Merkle’s unique offering that uses generative AI to deliver more impactful customer experiences. By applying AI to vast amounts of first-party brand data to build large knowledge models (LKMs) and leveraging the power of large language models (LLMs) on brands’ proprietary data assets, GenCX drives insights, audiences, creative, and campaign recommendations, all while improving speed, accuracy, and efficiency. The result is more connected experiences based on a better understanding of customer interactions, behaviors, sentiment, and engagements. Learn more.
• “AI is more than just a technology; it is a study of humans and human nature itself” Dentsu's global creative chief spoke to Campaign Asia on the enduring role of humanity and creativity in the age of AI. Read more.