Behind the Curtain: NVIDIA and Tastewise on Generative AI and Strategic Innovation
Speakers: Gilad Gershon and Kate Nir
In this talk, Kate Nir and Gilad Gershon explore the revolutionary effect of AI and its transformative potential in the food industry, emphasizing personalized nutrition, enhanced product development, and Tastewise’s innovative chatbot, TasteGPT. They also delve into the importance of strategic partnerships and offer valuable insights for industry leaders who are hesitant about adopting AI technologies.
[00:00:00 – 00:01:00]
Kate: Hi! Welcome, Gilad, so great to have you here.
Gilad: Thank you very much, Kate. I’m happy to be here.
Kate: So I’m Kate, and I’m the VP Product here at Tastewise. I’ve been with Tastewise from the very beginning. So I had the opportunity to build our product from scratch and take part in integrating many AI capabilities into our product. So I’m very excited to have you here today. Gilad, can you tell us a little bit about yourself, and how Nvidia is pioneering the use of generating AI across different sectors?
Gilad: So first of all, I’m honored to play a small part in Tastewise’s annual events and thank you for having me. In a few words. I’m Gilad Gershon, at Nvidia, I lead our generative partnerships with Israeli startups as part of the inception program designed for cross-vertical AI startups. Nvidia is at the forefront of pioneering the use of generative AI across various sectors by leveraging our advanced GPU technologies, our robust AI frameworks and SDKs. I believe that at least one of the reasons that is helping us is our pace of innovation. And our partner focus in understanding what is relevant to be built now and in the future in terms of the needed scale and innovation for our customers.
Kate: Nvidia is really a rocket these days. And talking about the future. How do you see generative AI changing the landscape of consumer behavior and the food industry?
[00:02:00 – 00:03:00]
Gilad: So generative AI is set to revolutionize the food industry by enhancing personalization, optimizing supply chains, improving customer experiences, and driving innovation as technology continues to evolve. And if you ask me, Kate, I’ll say that generative AI is still in its early days. It will likely play an increasingly central role in shaping the future of consumer behavior and the food industry. And I’d like to emphasize two main examples each here. First one is personalized nutrition and recommendations. Generative AI can analyze vast amounts of data on individual preferences, dietary restrictions and health goals to provide personalized meal and product recommendations. These personalizations can enhance customer satisfaction and loyalty by offering tailored solutions.
[00:03:00 – 00:04:00]
Gilad: Second example is enhanced product development. Food companies can leverage generative AI to innovate and develop new products by analyzing consumer trends, ingredient combinations and nutritional values. AI can suggest novel recipes, and food products that’s aligned with current market demands and health trends. Just like you and your team at Tastewise are successfully doing. And, Kate, can you please elaborate more on your team’s great work on this field?
[00:04:00 – 00:05:00]
Kate: So I’m always really proud to say that Tastewise is an AI company, and has been since day one. I think in the past year and a half we’ve really worked to integrate many generative AI capabilities into our product. And I think that what really is unique about those new technologies that we see in our product is how it can help us bring our customers closer to what they’re trying to achieve. So if in the past we could use those AI capabilities to really analyze the data and make sure that we can look at specific segments and kind of predict what is going to happen with a specific trend today. We can also help our customers act upon it so we can provide them with a very specific recipe that they’re trying to create…
[00:05:00 – 00:07:00]
Kate: …or the positioning for their next product innovation, or to actually let them see what the product can look like when it’s on the shelf. And all of these are thanks to the generative capabilities that we’ve worked to integrate.
Gilad: This is super interesting. Can you please share one generative tool, or some kind of generative capability, that is already being used in your product.
Kate: I think that the most straightforward example that we can give is the TasteGPT chatbot that is live in production, and is already being used by many of our customers. And I think that what’s very cool and unique about it is that it’s not only there as a chatbot that’s used to enhance our customers’ experience. It really can take real questions like, “What are the next ice cream flavors we can expect in the market?” or “Create a recipe for pizza.” So it can really be very specific questions that are relevant to our customers’ needs and use cases. And it takes into consideration the user itself, the question that was asked, and the best possible data that we have behind the scenes to answer the question, and I think one of the biggest challenges out there when it comes to GenAI or AI technology in general is trust. Right? So it’s really difficult to understand why you’re seeing what you’re you’re seeing. Why did you receive a specific answer to your question?
[00:07:00 – 00:09:00]
Kate: So Tastewise is always relating you back to that explainability and to the ability to understand why you’re seeing what you’re seeing. So you always have this follow up link where you can understand, where does the data that the answer is based on coming from.
Gilad: Sounds good. Thank you.
Kate: Of course. Where do you see the most significant opportunities for growth and innovation in the next decade with AI technology?
Gilad: I think that 10 years from now AI and humans will be much more integrated. And what I mean by that is that more humans will use more AI tools. And it’s relatively going to start at a very early age. I think that we already found out that AI can do crazy stuff but not every cool feature is a real business case behind it and I think it’s important not to fall in love with your product. I’m aware that some in the organization are trying their best to build solutions for problems that were already being solved without taking into consideration, what does it actually mean to build in terms of time and resources? Today, there are many solutions such as open source solutions and paid solutions that can help companies accelerate and focus on what really matters to them. And in the worst case, it can help failing fast instead of slow.
[00:09:00 – 00:10:00]
Kate: So agility and focus are so important when it comes to product development, whether it be a food product or a tech product. But basically we know how painful it can be to identify too many needs and spread ourselves too thin. And also working on long projects, only to find out one month into the process that we failed. So trying, learning, iterating is key and there are great tools out there that can really expedite the growth and we’ve talked a lot about the future of AI. But I think that your role is very unique because it’s really about connecting the startups to the enabling technology and help your partners drive success. So can you share a little bit about what are the key factors that make a partnership between a tech company and industry players successful.
Gilad: I think that there are many important factors in creating a strategic tech partnership. I’ll highlight two that I believe are very important.
[00:10:00 – 00:15:00]
Gilad: I’ll say that one is a leadership commitment and the reason why it’s so important is that it really takes a deep understanding of the partner needs and how the different departments in the startup organizations are working to achieve the desired outcomes. The other one is clear communication, especially on alignments and expectations. It’s a mutual relationship and responsibility and it really requireds both sides. In my experience, too often both sides of the partnership assume that things are going well without seeing the realities that come with different cultures and different tech backgrounds can be obstacle sometimes. So it is needed to clearly check on expectations.
Kate: I couldn’t agree more. I think that when it comes to product and our role in the organization, the biggest partnerships we have are actually with our customers. So it’s really about trying our best to deeply understand what our customers are trying to achieve and how they go about it. So we can actually go and invent on their behalf, and actually come and develop the best possible product. And yeah, communication is definitely always key in that as well. So what advice would you give to industry leaders who are hesitant about adopting AI technology?
Gilad: So my advice for industry leaders, who are still not sure about AI is to learn more about it. Be curious, read about the AI trends that are affecting your industry. And just to be more practical, those leaders can do some reverse engineering. For example, take a blank paper and write down all your biggest challenges, and how your desired future looks like. Place it on a graph where one axis represents how important is this challenge for you and the other axis represents your resources, how much you actually can invest in terms of money for skill sets. Then you got yourself a map that tells you all about the gaps you currently have and what you can do to act upon it.
[00:15:00 – 00:19:00]
Kate: I love it. Prioritization is really key. And we all have so many different opportunities and challenges right? So having the ability to understand what should we really focus on and maybe go and find solutions that are already out there that can help us. So what’s something that personally really excites you about generative AI right now?
Gilad: That’s a great question. And well, if you think about globalization and major impacts, just in the past few months, we’ve all been introduced to advanced AI translation for those that don’t know. There is a technology that can instantly change your voice to another language. I’m speaking with you English now, but I can switch to any other language using this technology. It is something that people might take for granted now, but remember that a few months ago we didn’t witness anything like that. And this is another testimony to the fast pace of generative AI innovation. And how about you, Kate?
Kate: So what’s really exciting me right now is everything that has to do with agentic models. Today, you can use generative AI to create a set of agents. So each agent could really be an expert in something, and could really focus on a very specific subtask that is relevant to answer a specific set of questions or to be the best possible expert in a specific field. Those agents can then work among themselves to come up with the best answer to your question and provide the best output possible for it as well. I can create my own team of agents and my team can create their own team as well, and so many of those repeating processes and the parts of your job that takes time and effort can now be automated and let you really focus on what matters to you. So in our case, it’s really all about discovering the best opportunities and developing solutions that provide value to our customers. But it could really be anything. I highly suggest everyone to take a look at it and try to find out how you can use it in your day-to-day as well.
00:19:00.00 – 00:19:10
Kate: Thank you so much, Gilad. It was such a pleasure to have this conversation with you today. And I can’t wait to see everything that we’re going to do together in the future.