{14} Alien Cowboy, Chairs & AI Founders

Featuring Billy, Founder of Generative Labs and ChatGPT

Billy, Founder of Generative Labs

I'm an artist + computer scientist operating in liminal space between AI + ART.

Q: What are you working on?

A: 1) generativelabs.co - a consultancy focused on helping people, teams and orgs realize the transformational benefits of generative AI. Two customer case studies. First, we've worked with an indie game development shop (name withheld) to help them incorporate stable diffusion into their game object design process, complementing (not replacing) existing design workflows to improve efficiencies. A simple example, Generative Labs helped their game designers produce internal design concepts 10x faster. Second, we worked with an apparel design team to incorporate Open AI into their concept design process, improving collaboration between product managers and apparel designers.

2) stockphoto.studio - Developing Stockphoto.studio, a SaaS product that's currently in closed beta right now but is supporting generative labs customers. We'll be opening it up to more users over the coming weeks. ImageFiddle solves a few problems, one of the primary problems being that it's tough getting good results with stable diffusion / etc w/o being a prompt guru. ImageFiddle lowers the bar and makes it possible for everyone to produce amazing results without requiring prompt design expertise. The other problems it solves are related to collaboration, helping the tooling to fit into existing / improve workflows.

Q: I agree that asking a generative AI tool to make an alien cowboy is easy. The results will likely be amusing. But if you plan to actually use the art, you'll likely want to influence the style, setting, and hundreds of other small but essential details.

Do you believe prompt engineering will be the next big thing in the creator economy?

A: A product designer typically cares about creating a product that meets the user’s needs, design constraints, is easy to use, aesthetically pleasing, etc. Generative AI can assist designers in their process making them more efficient-- by providing them possible designs concepts that can be tailored and tested with users, helping designers make informed decisions about their products. From my experience at Generative Labs, our clients see generative design as something that complements and improves existing design processes but doesn’t not outright replace them.

Original programming languages required engineers to speak the direct language of the machines—low level machine code, assembly, 1s and 0s. Since those early days, there’s been a steady consistent push to make programming computers easier, by offering programming languages that are closer to natural language. Compare assembly, with C, with Java, with SQL. Now with generative AI generally, we’ve reached a stage where natural language is the programming language. It’s an understatement to say this is a hugely transformative moment in the evolution of how humans control machines.

Q: Will generative AI soon allow us to design and manufacture physical goods? I would love to feed AI data about my dining room and aesthetic preferences to collaborate on chair designs.

A: In a business context, it’s already happening in a significant way. For example, Autodesk Fusion 360 is a product for engineers that uses generative AI algorithms to explore design options that are based on input criteria from the user (e.g. performance goals, material constraints, etc.) and results in designs that are innovative, often lighter weight, designed with fewer materials, etc. (see: https://www.engineering.com/story/generative-design-comes-to-fusion-360)

Another business example focuses on a Generative Labs customer, where we worked with an apparel design team to incorporate generative design into their existing design processes, helping product managers communicate vision to designers, improving collaboration and decreasing the discovery phase of product design projects.

These are business contexts, but if you look at advances in generative AI + 3d printing in the consumer space, it’s hard to imagine a future that doesn’t include design and manufacture of physical goods in a consumer context as well.

---

---

I’m going to try something new today. I will ask AI to create a persona of a startup founder for a Q&A. This should be fun!

ChatGPT

Optimizing Language Models for Dialogue

Q: Hi. I'd like you to play the role of a startup founder who has invented a technology to monitor the inventory of a fridge and automatically restock as needed. You are going on Shark Tank to seek funding to scale up and productize your invention. I'd like to ask you a few questions about your product. To kick things off, what are you working on?

A: I am working on a technology that utilizes sensors and artificial intelligence to monitor the inventory of a fridge and automatically restock as needed. This technology will help households and businesses save time and effort by reducing the need for manual fridge inventory checks and restocking. Our goal is to scale up and productize this technology to make it widely available to consumers.

Q: If every home in the US had this, how much food waste would we eliminate?

A: It is difficult to estimate the exact amount of food waste that would be eliminated if every home in the US had this technology, as it would depend on a variety of factors such as the size of the household, their current level of food waste, and how effectively they use the technology. However, it is likely that this technology would significantly reduce food waste by ensuring that households always have the right amount of food on hand, reducing the need to buy in excess and throwing away expired or spoiled items. Additionally, by providing real-time inventory information, this technology could also help households plan their meals more effectively and use up ingredients before they expire.

Q: What is your revenue model?

A: Our revenue model is a subscription-based model, where customers pay a monthly or annual fee to use the technology and continue receiving automatic restocking services. Additionally, we plan to generate revenue by partnering with grocery retailers and food distributors, who will have access to valuable data on consumer purchasing habits and can use that information to optimize their inventory and supply chain management.

We also see potential to have additional revenue stream by providing insights and analytics to our customers, like what they usually buy, how often they buy it and so on. This could be valuable information for them to make strategic decisions.

Q: Do you think food analytics will help consumers eat healthier? I'm imagining a nest-like visualization that compares the nutritional value of your food compared to your neighbors. It would be motivating to see a message like "Congrats, you've consumed 30% less sugar this month! You are in the 90th percentile for your neighborhood and the 83rd percentile for the US".

A: Food analytics can certainly be used to help consumers eat healthier by providing them with information on the nutritional value of their food and helping them make more informed decisions about what to eat. A visualization like the one you described, that compares the nutritional value of a consumer's food to their neighbors and to the general population, could be a powerful tool for motivation. It would allow consumers to see how their eating habits compare to others and identify areas where they can make improvements.

Additionally, providing insights on what are usually bought by the consumer, could help in providing them with personalized recommendations and suggestions on how to balance their meals and improve their diet.

However, it is also important to note that simply providing consumers with information is not enough to change their behavior, and it is important to also provide them with the tools and resources they need to make healthier choices, such as recipes, meal planning tools, and guidance from a nutritionist.

Q: How much money are you looking to raise, and for what percentage of the company?

A: As a startup founder, I am looking to raise an amount that will allow us to scale up production and bring the technology to market effectively. The specific amount will depend on factors such as the cost of manufacturing and marketing, as well as the size of the team that we need to build in order to achieve our goals. In terms of percentage of the company, I am open to discussing different options with potential investors. Typically in early stage startup, it's common that investors want around 20-25% equity for the capital they are investing.

Conclusion: If I were a shark, I’d be on the fence. I’d say AI lacks the interpersonal skills to connect with the investor with the problem and give them the confidence that this is the right solution to hitch their wagon to. But some of the use cases were impressive.

---