How corporations and startups can flourish together in AI era

In today's rapidly evolving business landscape, a steady stream of startups is securing VC funding to develop generative AI applications. However, established companies across various industries—corporates—seem to be missing in action when it comes to pioneering category-defining products.

Per a Q1 2024 EY analysis, 20% of the billions in VC funding is allocated to AI startups, creating potential competitors for corporations. While startup competition is not a new phenomenon, the rapid growth of AI startups requires novel approaches to manage the ensuing disruption. The top three sectors of VC investment are Healthcare, IT, and Business Services.

Managing corporate disruption

Historically, corporations have managed disruptions by leveraging their financial resources: "Let them build, we will buy." In other instances, they made minority investments in early-stage potential competitors—if they succeed, we get a share.?However, in the AI era, startups are leveraging advanced technologies to drive rapid and transformative industry disruption. To stay competitive, corporations need to become customers and co-build with startups, influenced by factors such as:

• Cost of acquisition: Acquiring valuable AI startups will be expensive.
• Learning and transforming: Deep engagement in building the AI required for their sector is essential for corporate teams to learn and transform.?
• Cultural integration: Acquiring AI startups without the company’s own transformation can cause a collision of vastly different work cultures, accelerating disruption.

Corporate versus startup advantage

Corporations are well-positioned to invent category-defining products due to their financial resources, vast data repositories, and established distribution channels (massive reach through existing products and platforms). Startups, on the other hand, often operate with limited funds, must acquire data through various means, and build distribution networks from scratch.

While the traditional startups typically innovate within the framework of more established technologies, the AI startups are stretching the limits of our imagination and possess a culture that is learner, more agile, and collaborative. If corporations can integrate some aspects of this culture, they are ideally positioned to invent new product categories, leading to new revenue streams.

The layers of AI tech stack

The AI landscape is complex, with numerous models, tools, platforms, and applications emerging across sectors. Understanding the layers of AI tech stack is crucial for mapping applications to their expected outcomes and making informed decisions. The diagram below outlines the key layers. Excluded is the data layer and traditional AI for simplicity.

Emerging product categories

Multimodal AI and the capability to handle extensive data context are shaping new product categories. Startups, with their culture of innovation, collaboration, and higher risk tolerance, are likely to lead these innovations. Corporations must overcome challenges like hierarchical cultures, technical debt, and workforce upskilling to compete effectively.

An approach for strategic building and investment

Deciding what to build is a deeply analytical exercise involving corporate strategy, ventures, business, and technology teams. To navigate the AI ecosystem successfully, corporations should:

1. Assess VC funding trends: Analyze VC funding in the sector and map it to the layers of the AI tech stack, from series A to pre-IPO. Gaining insights into relevant activities within the startup ecosystem is crucial.
2. Leverage proprietary data: Utilize unique data assets to address customer pain points and fine-tune pre-trained models.?
3. Build down the tech stack: Start with simpler layers first for faster time to market and ROI. Build or co-build as you progress down the layers of the AI tech stack (Layers 7, 6, 5, 4). Leverage managed LLM and co-build with the startup as a partner. Becoming a customer of the startup ecosystem, enables corporations to bring transformative products to market sooner while upskilling their own organizations.
4. Invest up the tech stack: Make minority investments in more complex layers as you move up the tech stack (Layers 4, 5, 6, 7). Leave the foundational layers 1, 2, and 3 to tech giants.

Product distribution

AI-powered data products developed with startups can be marketed through various channels and integrated into existing platforms to scale. They can be offered as standalone subscriptions in freemium or premium business models, as well as through sales and customer success teams, or digital marketplaces.?

Conclusion

Business growth and innovation go hand in hand. Corporations must actively engage with the rapidly evolving AI ecosystem, build new products, and adopt the innovative and agile culture of startups. By doing so, corporations and startups can flourish together in the AI era.

The above is adapted from my discussion with the CTO/CIO forum for AI at The Millenium Alliance where I shared a framework for how corporations can invent new product categories and develop win-win models by integrating with the startup ecosystem.