Who wins the race in Generative AI?

Generative AI will have a material impact on the B2B market, but what is the long term impact?

Tooba Durraze
Tooba Durraze
June 12, 2023
min read
Apple Podcast LinkGoogle Podcast LinkSpotify Podcast Link
Apple Podcast LinkGoogle Podcast LinkSpotify Podcast Link

Meet Dr. Tooba Durraze, PhD

As the VP of Product Strategy at Qualified, Tooba works on building innovative products as well as Qualified’s AI strategy. She has a PhD in Computer Science from MiT, focused on AI model interactions with varying data types.

Follow Dr. Tooba on LinkedIn for more insights as AI continues to evolve.

Who wins the race in Generative AI?

With the proliferation of Generative AI applications in the market right now, things are starting to get muddy. While there may be clear winners in some verticals, in others, the incumbents and the race itself is shifting daily. There are no real differentiators, and the technology is quickly becoming commoditized. So what does this world really look like? It's filled with infrastructure vendors, model providers, and those building applications on top.

The reason this is important is because the market is going to shift based on what is truly differentiated and defensible. And since that's missing from existing incumbents, it becomes impossible to understand the business models around Generative AI. We all know that Generative AI will have a material impact on the B2B market but we are unsure of the shape of its long term impact.

I am extremely excited about the opportunities surrounding Artificial Intelligence, and Generative AI. At Qualified, there are some principles we have chosen to put in place in order for us to have a differentiated vision.

1. An integration is not a product

The first wave of applications are consuming AI model services. It allows developers to iterate quickly, with minimal resources, and even swap out models as technology advances. Anything built on top of this, will be entirely indefensible as there is no differentiator. Most applications and features are primarily using the top five models, meaning their outputs will be very similar.

Generative AI should be thought about as a means to the end and not necessarily as a product in itself. It's important to build applications that are anchored to actual use cases that solve a problem for the target buyer vs. flashy ones that are one part of the story.

Second, is understanding what makes it unique. If looking to solve the problem of optimizing pipeline generation via the website, the applications built should then be looking to solve this for the marketer instead of providing gimmicky outputs with no material impact (as fun as they might be).

The winners will be those who are sitting on a rich data layer that allows our output to be more contextual and is trained to drive results towards specific use cases.

2. Beware of the hype cycle

While I am of course super excited about the wide-spread adoption of Generative AI, we are still cognizant to make sure what we invest in aligns to our footprint around pipeline generation. There is a ton of speculation around how this surge of hype will actually pan out when it comes to delivering value to the buyer.

Under the hood, the investment for the Qualified product team is around continuing to build on our data layer and investing in building towards a robust semantic search engine. Our efforts are uniquely split between investing in data ingestion, treatment, as well as building applications on top.

This means, despite the hype, we take our time in making sure our features are not just flashy, but have a strong data relevance foundation—which takes time. Products and features that are not attached to clear use cases will fizzle out.

Matter of fact, we are so aware of the hype bias, we instituted a penalty of a dollar every time any one mentions the broad term “AI” as a solution to something rather than the actual underpinning model or solution within that umbrella term.

3. Constant experimentation is key

Because of the nature of this technology, it's inevitable that a new model or a new set of solutions will take the space by storm. We believe not only in experimentation as it relates to Generative AI models, but what that means for our overarching objective, which is to drive more pipeline.

What that means in practice is that we often experiment between Generative AI, synthesis AI, and optimization algorithms. We are setting up a system that allows us to experiment not just in a particular output form (generative as an example), but what the system does as a composite to drive better pipeline. The ultimate goal of this experimentation is to allow the flexibility to play around with multiple models that are all working together as a coherent system towards the same objective.

The experimentation also allows us to be fairly flexible when it comes to new models or technology being introduced in the market.

4. Automation and orchestration is the ultimate goal post

While the hype in the world today is around Generative AI, ultimately, we are all headed towards the end goal of doing more with less. Hence the game theory around who wins at the Generative AI race is inherently flawed. The real technology race is around who can use a system of AI to power up their entire solution in a way that doesn't reflect the kind of “pixie dusting” that is currently gaining market momentum.

My personal bet is on advancements around orchestration and automation. The real differentiator is going to be around who can scale and potentially optimize the role that has been traditionally played by a growth marketer.

The real power of AI will sit in the backdrop of most of the key operations and will seamlessly orchestrate the related tasks needed to accomplish the use cases in the most optimized way possible—in a way that is potentially impossible for a human to compute around.

For those paying attention, the race in Generative AI isn’t necessarily a race in Generative AI after all. The real race is toward who can skip ahead and utilize this market swell, to move towards optimization and orchestration. Those staying true to their use cases at core, and utilizing the technology as a way to achieve clear goals related to those problems faster will be the ones who ultimately end up ahead.

Both the level and the number of inventions in Generative AI can only mean one thing—there are incumbents in the market who are already heads down focusing on the third S curve of innovation around orchestration—one that the market is seemingly blind to.

No tags added.

Related articles

Qualified in Action

Quick demo?

Discover how we can help you convert more prospects into pipeline–right from your website.

Contact Us