VentureBeat Transform Day 2: Embracing partnerships for generative AI success

VentureBeat Transform kicked off its second day with wisdom from leaders of startups and large companies, sharing their experience in delivering generative AI for different customer segments, industries and regions.

The panel discussions and fireside chats touched on some important themes in navigating the challenges of the fast-changing world of generative AI. Here are some of the key takeaways from VB Transform Day 2.

Humans in the loop are key to success

One recurring theme in generative AI success stories is keeping humans in the loop. The technology is not mature enough to be left to its own devices. But when combined with human intuition and control, it can do great things.

>>Follow all our VentureBeat Transform 2023 coverage<<

In a session at Transform 2023, Wilko Schulz-Mahlendorf, head of pricing and marketing science at Wayfair, explained that a key tenet of the company’s approach to generative AI is to make sure humans are in control of the process.

The company has already rolled out two successful generative AI products. One is a co-pilot for marketing and sales agents that enables them to retrieve information at a much faster pace than before. The second is a tool that enables copywriters to double and triple their efficiency by providing them with high-quality first drafts that they can then edit to their liking.

Human-centric design

In the same vein, in a separate fireside chat, Daniela Jorge, senior vice president and head of experience design at Capital One, emphasized the importance of human-centered design in creating AI products. Jorge stressed that human-centered design will not only be important for the consumers of AI products but also for the data scientists and engineers who develop AI models.

Generative AI will also create new opportunities for human-centric design. “Before we used to design for broad segments of users,” Jorge said. “With AI, there is an opportunity to have much more one-to-one solutions between humans and systems, which will be interesting.”

Collaboration is key to success

In a panel discussion, NTT VC founding partner Vab Goel and NTT’s CFO and senior EVP Takashi Hiroi discussed some of the trends to look out for in the changing landscape of generative AI.

Hiroi highlighted two key challenges. First, the industry is still trying to figure out the pricing model for AI products. Given AI systems’ huge energy consumption, many companies are currently offering their AI services at a loss to capture market share. Going forward, pricing will be determined by the amount of value created, Hiroi said. At the same time, the enormous compute resources required to run generative AI models will make energy consumption an important indicator in developing and deploying AI products.

Goel stressed that a successful AI strategy will pivot on collaboration. For large companies, he advised looking for partnerships with startups.

“It’s easy to partner with leaders. But remember that OpenAI is a five- or six-year-old company that Microsoft is using to launch their service,” he said. “Broaden the scope. Look at early-stage companies … Partner with them early and shape their vision. You will have a competitive advantage … Meeting a lot of startup companies and taking some risks is going to be critical. It is clear that it will be the partnership of a large company and a small company that will be the winning formula.”

At the same time, AI startups that are raising $100 million and more can minimize risk by partnering with large companies and expanding their market. “Startup companies should find go-to-market partners,” Goel said. “OpenAI and Microsoft is a good example that early-stage companies should early on try to find some of the large companies who can take them and introduce them to customers and build services around that.”

Generative AI opportunities for enterprise success

In a panel discussion, Gerrit Kazmaier, vice president of data and analytics at Google Cloud, and Matt Wood, VP of product AWS, gave their perspective on the opportunities and risks of generative AI.

Wood and Kazmaier identified multiple “buckets” of opportunities for enterprises to benefit from generative AI.

Wood outlined use cases for generating content (blog posts, marketing copy, source code, etc.); new personalization options for search, ranking and relevance; helping experts work more efficiently (e.g., programming co-pilots); and creating opportunities for collaborative problem solving where humans interact with expert systems to drive decision support.

Kazmaier talked about productivity, where generative AI can have a profound impact. One of the key step changes in productivity is enabling non-coders to generate code and create applications, he said. Generative AI will also help work with unstructured data in ways that were previously impossible.

Kazmaier also said that generative AI has the potential to change customer experience by making it easier for users to communicate their demands and intents. And finally, he said that generative AI can help create a new range of products that were inconceivable with previous tech stacks.