Welcome to another episode of the AI Innovators Podcast!

I'm your host, Rob May, and in today's episode, we're wrapping up the year with a bang by looking ahead to AI in 2025. I've gathered a panel of AI experts from Half Court Ventures to discuss the current landscape, future trends, and the evolving challenges and opportunities in the AI sphere.

AI Innovators Podcast is sponsored by Half Court Ventures and the Investing in AI newsletter, this episode promises to provide a wealth of knowledge for anyone interested in the cutting-edge developments in AI and their real-world applications.

So, sit back, relax, and get ready to be inspired by the insights of the Half Court Ventures Team - Byron Galbraith, George Lin, and Eric Koziol

Episode Highlights:

• OpenAI's Market Position: Byron Galbraith, George Lin, and Eric Koziol analyze whether OpenAI remains the frontrunner in the AI industry amidst rising competition, legal challenges from Elon Musk, and varying business models.
• Big Tech's Dominance: We dive deep into the capabilities and future potential of major companies like Google, Nvidia, Meta, and Apple. With insights into compute power, custom chips, and the scaling of AI hardware, our panelists George Lin and Byron Galbraith provide a comprehensive outlook.
• Next-Gen AI Architectures: George Lin introduces his prediction on the shift beyond transformer architectures, citing innovative models like mamba and rwkv. The panel discusses how these could revolutionize AI by extending context and efficiency.
• Generative Media Breakthroughs: Byron Galbraith brings to light the impact of Flux models on the generative media space, the quest to reduce inference costs, and the shifting landscape of local versus cloud-based model training.
• Custom Software Development with AI: Eric Koziol delves into the notion of personalized software, where the cost of software development drops, enabling bespoke applications tailored to specific business processes. This segment examines both the benefits and potential drawbacks of this democratization.

Thanks for tuning in to the AI Innovators Podcast. Stay ahead of the curve by subscribing, and don't miss out on our next deep dive into AI trends and strategies!

0:00:50 - Addressing the status of OpenAI and competitive threats.

0:01:25 - Byron Galbraith on OpenAI's market position and technical challenges.

0:02:17 - Eric Koziol's take on OpenAI's market strength and technical weaknesses.

0:02:40 - George Lin's insights on OpenAI's expanded focus and business strategy.

0:03:38 - Opinions on changing from startup focus to big companies in AI.

0:04:24 - Eric Koziol highlighting Google's investments in energy and compute capacity.

0:05:17 - George Lin talks about Google's internal resources and AI integration.

0:06:06 - George Lin mentions Facebook's advantages concerning human-vetted training data.

0:06:24 - Byron Galbraith supports Google and Meta's positions in the AI space.

0:07:54 - Rob May suggests potential Apple advantages in AI.

0:08:30 - Eric Koziol comments on Apple's potential challenges in AI innovation.

0:08:48 - George Lin discusses Apple's MLX framework and potential in AI.

0:09:35 - Byron Galbraith on benchmarking Apple processors for AI tasks.

0:10:34 - Rob May transitions to individual interesting AI topics for discussion.

0:11:04 - George Lin predicts moving beyond transformer architecture in 2025.

0:12:08 - Byron Galbraith supports re-exploring recurrent models in AI.

0:13:21 - Eric Koziol on the need for evolving AI architecture.

0:13:34 - George Lin mentions advancements in computation and data collection tools for AI.

0:13:53 - Rob May moves to Byron's topic on generative media space.

0:14:29 - Byron Galbraith highlights the introduction and impact of the Flux model.

0:15:34 - George Lin discusses democratization and consumer hardware for AI models.

0:16:26 - Eric Koziol asks Byron about advancements in generative models and their progress.

0:17:44 - Byron Galbraith on the improvements and future of Flux and generative models.

0:18:20 - Byron Galbraith talks about Flux model capabilities and training efficiency.

0:19:26 - Rob May transitions to Eric's interesting topic on personalized software.

0:20:23 - Eric Koziol explains the concept of personalized software in businesses.

0:21:15 - George Lin questions if enterprises will adopt AI tools over enterprise-grade software.

0:21:39 - Eric Koziol on the trade-offs of homegrown software versus buying software.

0:22:36 - Byron Galbraith discusses the maintenance of software built using AI tools.

0:23:46 - Rob May concludes on the importance of monitoring the trend of personalized software.

0:24:26 - Closing remarks and call for listener feedback and guest suggestions.

Welcome to another exciting episode of the AI Innovators Podcast! I'm your host, Rob May, and today we have a truly special guest joining us: Kevin Novak.

Kevin is a former data scientist at Uber and now a venture capitalist with Rackhouse VC.

In this episode, we dive deep into fascinating topics surrounding the transformative power of AI in early-stage businesses, the evolving landscape of data and infrastructure, and the nuances of AI investment strategies.

Episode Highlights:

• Kevin delves into the shift from viewing data as a byproduct to recognizing its critical role in AI systems, and the implications of increasing AI demands on data centers and power capacity.
• We discuss the performance dynamics of large language models versus small, specialized AI models, and their future potential, including mobile integration.
• Kevin shares valuable insights into the unpredictable impacts of AI across various industries, the importance of continuous experimentation, and the concept of Harvard's "jagged frontier."
• We explore the strategic challenges and opportunities in AI-focused venture capital, Kevin's approach to early-stage investments, and the essential traits of successful startup teams.
• Kevin and I also touch upon the importance of deep customer relationships, managing AI’s operational complexities, and the evolving role of AI in enhancing current business metrics.

Sponsored by Half Court Ventures and the Investing in AI newsletter, this episode promises to provide a wealth of knowledge for anyone interested in the cutting-edge developments in AI and their real-world applications.

So, sit back, relax, and get ready to be inspired by the insights of Kevin Novak on the AI Innovators Podcast.

Kevin Novak's

(rackhouse.vc/)

(Kevin's Linkedin)

Rob's Links

( halfcourt.vc )

( Rob AI Newsletter )

01:00 - Determining genuine AI companies from buzzword claims.

06:40 - Focus on fundamentals, proprietary data, experienced AI team.

11:11 - Large language models grow with more resources.

13:00 - Limitations challenge scaling large language models.

16:52 - Balancing cost, profitability, and AI advancement challenges.

20:52 - Companies often dictate customer roles, causing insecurity.

22:44 - Adapting workflows to technology shifts drives success.

26:23 - Diligence pays off despite emotional misalignment.

29:27 - Invest globally, focus: complete businesses, max $2M.

Welcome to another exciting episode of the AI Innovators Podcast! I'm your host, Rob May, and today we're diving into a truly unique domain as we explore the intersection of AI and the insurance industry.

Our guest is Alex Schmelkin, CEO of Sixfold AI, a company that's revolutionizing how insurance companies underwrite policies.

In this episode, Alex gives us an insider look at the traditional human-led underwriting process and how Sixfold AI is reshaping this landscape with advanced AI technology. We discuss the challenges and opportunities in automating underwriting, the impressive ROI their solutions bring, and the intricacies of adapting AI across different insurance types.

Alex shares valuable insights into the decision-making journey of choosing to enable traditional insurance carriers rather than becoming one, and his perspective on the future of AI insurance. We also touch on the emerging field of AI model insurance and the critical role of document ingestion in their AI tech stack.

Get ready to learn from Alex’s vast experience and gain practical advice that every entrepreneur and AI enthusiast can apply. Whether you're in the insurance industry or simply fascinated by AI's transformative power, this episode is packed with valuable takeaways. So, tune in and join us on this deep dive into the future of AI in insurance!

Alex Schmelkin's Links

( https://sixfold.ai/)

(Linkedin)

(Alex's Linkedin)

Rob's Links

( halfcourt.vc )

( Rob AI Newsletter )

01:00 - Underwriters assess risk using vast data volumes.

04:54 - AI evaluates insurance applications for underwriting decisions.

08:14 - Innovators podcast by Half Court Ventures, AI.

11:38 - Enhancing insurance with advanced underwriting models.

15:16 - AI exclusions in insurance policies raise concerns.

17:44 - Reinsurance differs between life and PNC sectors.

21:39 - Focus on your core strength and excel.

23:12 - Target buyer's conference indicates optimal selling approach.

Welcome back to another episode of AI Innovators with Rob May, the podcast that deep dives into how AI is reshaping industries and the groundbreaking minds behind it.

In today's episode, we're joined by Pawel Zimoch, co-founder of Featrix AI, a company known for pioneering vectorization solutions for the enterprise.

As we explore the world of vectors and embeddings, Pawel unpacks the significance of data predictability and why some data sets work better with AI than others.

We'll also delve into the evolution of Featrix, from its founding story to its innovative strides in custom embeddings and multi-modal data analysis.

Whether you're an entrepreneur, a data scientist, or simply curious about the future of AI, this episode is packed with insights you won't want to miss. Tune in for a fascinating conversation on the complexities, challenges, and exciting advancements in the realm of AI.

Pawel Zimoch's Links

( Featrix.ai )

(Linkedin)

( Featrix)

Rob's Links

( halfcourt.vc )

( Rob AI Newsletter )

00:00 - Embeddings use cosine similarity for comparative analysis.

04:57 - Predictive data value varies; feature engineering necessary.

06:28 - Efficient AI system predicts project viability quickly.

12:04 - Human judgment and problem-solving skills prevail.

13:59 - Learning computer science is essential for understanding AI.

16:26 - Managing models requires significant context from data scientists.

22:28 - Connect with others for broader learning perspectives.

23:28 - Success in entrepreneurship relies on understanding people.

Welcome to another exciting episode of AI Innovators with Rob May!

In today's episode, we are thrilled to have Kojo Asai from Matrix Capital as our guest. Kojo dives deep into his intriguing thesis on why inference costs are plummeting towards zero and its massive implications for the AI industry.

We’ll explore how dropping costs can unlock groundbreaking applications, revolutionize generative media, and shape the future of AI.

Kojo also shares his insights on the hottest layers in the AI stack, evaluates the potential of companies like OpenAI, and offers valuable advice for aspiring entrepreneurs and investors.

Plus, stay tuned for a speed round where Kojo gives his buy or sell verdict on emerging AI trends. Whether you're an AI enthusiast, investor, or entrepreneur, this episode is packed with thought-provoking discussions and valuable insights you won't want to miss. Let's dive in!

Kojo Osei's Links

( Kojo's Blog )

( Matrix VC )

( Kojo Linkedin )

( Kojo's Twitter / X )

Rob's Links

( halfcourt.vc )

( Rob AI Newsletter )

01:00 - Competition drives inference costs reduction significantly.

05:35 - Evolution of tech parallels Internet's media expansion.

09:28 - Open source enhances adaptability; application layer underrated.

11:42 - OpenAI achieved rapid growth, comparable to Google.

15:55 - AI needs business model to capture value.

18:07 - AI impersonation threats impact identity-centric businesses.

22:34 - Lead with curiosity; choose inspired over anxious.

23:52 - Gratitude expressed; requests blog URL for more.

Welcome to another exciting episode of 'AI Innovators with Rob May.' This is the podcast where we dive deep into the transformative world of artificial intelligence and its impacts across various industries.

I'm your host, Rob May, and today, we have a fascinating episode lined up for you.

In today's episode, we're thrilled to be joined by two incredible guests, Steve Peltzman and Sam Koch. Steve is here to shed light on the limitations of traditional surveys in identifying specific issues, particularly in public restrooms, and to emphasize the unparalleled insights provided by real-time data.

Sam, on the other hand, has extensive experience in automating workflows using AI within the private equity space and will be sharing some innovative applications of AI in enhancing operational efficiency at Feedback now—a company renowned for its groundbreaking restroom feedback buttons and real-time data collection through IoT devices.

We'll explore how Feedback now's technology extends beyond restroom feedback to improve services in hospitals and other facilities.

Additionally, we'll discuss a three-tiered AI adoption strategy that businesses can use to integrate AI effectively, the importance of measuring and optimizing facilities like older airport bathrooms, and the critical role of agility and real-time feedback in smaller companies.

Whether you're an executive looking to adopt AI in your operations or simply curious about the future of AI, this episode is packed with insights, real-world examples, and actionable advice. So, without further ado, let's jump right into our conversation with Steve Peltzman and Sam Koch. Stay tuned!"

Contact Feedback Now

https://feedbacknow.com/

[email protected]

feedbacknow

Sam's Linkedin

Steven Linkedin

Rob Links

( halfcourt.vc )

( Rob AI Newsletter )

04:23 – Leveraging AI to automate private equity deal sourcing.

09:34 – Exploring the customer journey: interconnected experiences and lack of control.

13:50 – Integrating diverse data sets with real-time AI for enhanced insights.

16:41 – Accelerating transactions to boost spending and elevate customer experience.

19:31 – Encouraging feedback participation to influence system improvements.

21:44 – How schedule changes and feedback lead to greater efficiency and savings.

26:41 – Managing egos: Sam’s leadership transformation focused on innovation.

29:59 – Overcoming barriers: Avoiding comfort in victimhood.

31:26 – Enhancing decision-making by learning from past mistakes.

Welcome to AI Innovators podcast with your host, Rob May.

In this episode of the AI Innovators podcast, I chat with Jason Patel, founder of Openforge.

Jason shares his intriguing journey of building an AI-powered email sales tool, highlighting the dynamic decision-making process about the right level of AI integration into their tech stack. Openforge aims to enhance the email sales experience, and Jason discusses the company's evolving strategy, navigating whether to adopt platform models and capturing market momentum.

We delve into Openforge's choice of using Claude from Anthropic as their primary Large Language Model (LLM) and compare it to other AI models, like OpenAI. Jason reveals how Claude excels in parsing natural language and providing nuanced responses in sales communications, emphasizing the arms race in AI development.

The conversation also addresses the potential future of AI in sales and customer acquisition. Jason speculates on the evolving role of AI agents, suggesting that while AI might handle much of the grunt work, the indispensable human touch in sales and customer interactions will continue to hold significant importance.

Additionally, we discuss the rapid pace of AI advancements and how this new wave influences entrepreneurial strategies. Jason reflects on his experiences as a second-time founder, contrasting his previous edtech venture with the current AI-driven landscape, and underscores the necessity for founders to remain agile and pivot swiftly in response to market changes.

Tune in to gain insights on how to navigate the ever-changing AI ecosystem, the importance of embracing market feedback, and the future trajectory of AI in transforming industries.

Guest Links:

( Open Forge Ai/ )

( Jason's Linkedin )

Rob Links

( halfcourt.vc )

( Rob AI Newsletter )

05:10 AI development is an ongoing competitive arms race.

07:14 AI supports tasks; human connection remains essential.

12:29 Embrace change; future communication methods are unknown.

15:58 AI ecosystem evolution prompts shift to platform.

18:55 Potential AI monopolies threaten economic diversity.

21:24 Entrepreneurship: Celebrated, yet challenging and financially dire.

23:46 Tune in to AI Innovators podcast, subscribe.

AI Innovators Podcast : interviews technical leaders, investors, and business executives about the impact AI is having on business models, markets, products, and consumer behavior.

We investigate impacts on public and private markets, from startups, to large companies like Amazon, Google, and Facebook. If you invest at any level and are looking to understand how AI will impact your world, please subscribe. Presented by HalfCourt Ventures

I'm your host, Rob May, a seasoned entrepreneur and investor in over 100 AI startups. In this series, we dive deep into candid conversations with the trailblazers at the forefront of AI technology. From technical leaders to business executives and academics, we uncover how AI is reshaping everything from startup ecosystems to global giants like Amazon, Google, and Meta.

Join us as we discuss the impact of AI on markets, innovative products, and the evolving ways we work and live. Whether you're an investor looking for insights, a tech enthusiast, or simply curious about the future, this podcast is your guide to AI's profound influence. Sponsored by HalfCourt Ventures, subscribe now to hear real-world stories from those driving the AI revolution.

( halfcourt.vc )

( AI Newsletter )

Welcome to AI Innovators podcast with your host, Rob May.

In this episode of the **AI Innovators podcast**, I chat with Alex Gunnarson and Zach Hudson, co-founders of Onton (formerly known as Deft).

Onton was born out of Alex's frustration when he spent 30 hours searching for a specific couch, an experience shared by many online shoppers. The duo identified significant inefficiencies and trust issues in existing e-commerce search engines, which often resulted in shoppers spending extensive time across multiple websites to make a purchase over $50.

Alex and Zach describe their journey from basic vector search techniques to developing advanced AI integrations, including natural language processing and color data sets, to enhance search capabilities.

They introduce their self-learning system that improves search results based on user queries, utilizing the principle of compositionality. This innovative approach allows their search engine to generate outputs far beyond its training set, facilitated by a highly expressive knowledge graph database.

The conversation explores philosophical influences like Douglas Hofstadter and symbolic logic, advocating for a neuro-symbolic AI system as the next frontier.

They highlight the recent shift towards consumer AI driven by genuine pain points and the novelty of AI solutions, making such investments appealing and less risky. Onton's success in reducing search friction and enhancing personalization demonstrates the value of consumer AI.

Addressing the long-term defensibility of Onton, they emphasize their unique, clean data pipeline and search index. They also discuss the necessity of solving current AI limitations to progress toward AGI, expressing skepticism about immediate breakthroughs.

Guest Links:

(https://onton.com/ )

(Alex's Linkedin)

(Zack's Linkedin)

Rob Links

( halfcourt.vc )

( Rob AI Newsletter )

01:00 Co-founders solved e-commerce search problem differently.

04:52 Developed robust parser for natural language processing.

08:04 Shopping is time-consuming and depends on data accuracy.

12:33 Blended symbolic and connectionist approaches enhance AI.

15:00 AI drives renewed interest in consumer startups profitability.

19:12 Core business products are defensible, not minor improvements.

21:29 AGI's actions depend on differing objective functions.

23:34 Finding value in small moments amid life's anxieties.

Welcome to AI Innovators podcast with your host, Rob May.

In today's episode, we're excited to delve into the dynamic landscape of AI with Derek Larson, a seasoned entrepreneur and founder of multiple AI ventures. Derek also leads the New York AI group, a thriving community for AI enthusiasts and professionals.

Together, Rob and Derek explore Derek's journey from his early entrepreneurial days, building marketing tools for small businesses, to his latest endeavors in mental health and therapist-focused AI tools. They discuss the evolution of AI acceptance, the hype around LLMs, and the practical challenges faced by AI startups today.

Derek sheds light on the unique tech ecosystem in New York, comparing it to the Bay Area, and shares his insights on the trends that excite him the most and those he believes are overhyped. They also discuss the potential and limitations of AGI, the complexities of AI applications, and the importance of selecting the right problems to solve with AI.

Finally, Derek offers valuable advice for aspiring entrepreneurs, emphasizing the necessity of passion and resilience in the startup world. This episode is packed with insights for anyone interested in the AI industry, from seasoned professionals to newcomers. So, tune in and join us as we navigate the fascinating world of AI innovation.

Guest Links:

(https://kin.is/)

(https://newyorkai.org)

(https://www.linkedin.com/in/derek-larson/)

Rob Links

( halfcourt.vc )

( Rob AI Newsletter )

06:27 New York forces diverse problem-solving, unlike Bay Area.

09:52 Skeptical about $100 billion valuation and competitiveness.

10:38 Doubts about AGI potential and inherent limitations.

14:10 Osmo uses AI to interpret chemical structures.

17:23 Structured outputs for language models; efficiency crucial.

20:29 Earn well and enjoy building unique tech.