Artificial Intelligence

Everything Everywhere All at Once: There is no such thing as an AI company.

Chip Hazard1 Comment

I’ve been obsessed with AI and its impact on our world for decades. This obsession led to several investments in the field, as I described four years ago in my blog posts on Applied AI:  Beyond the Algorithm and The AI Paradox. So like many others, I have watched the evolution of generative AI and ChatGPT with keen interest.

At our Flybridge investment team meeting earlier this week, my colleagues asked how the explosion in the AI market compares to previous tech trends I have seen emerge over the last nearly 25+ years. Spoiler alert: I’ve been a VC for a looong time – see my reflections on 25+ years here).  

While the move to the cloud (looking at you, MongoDB), mobility, and the rise of consumer social apps were all significant developments over the previous 15-20 years, they pale in comparison to the sudden change in the market over the last 12 months brought about, primarily by OpenAI and the launch of first GPT-3, more recently ChatGPT, and the rapid advances from other players such as Google.

In particular, I am struck by the speed of adoption and the incredible ubiquity and breadth of impact that the Large Language Models ( LLMs) and Generative AI already have.  Today Groupthink is our always-on collaborative AI research assistant (picture a stateful marriage between ChatGPT, Google Search, and your favorite collaborative apps, and that’s Groupthink); we use the AI-powered no-code tool Blaze to build internal apps, Aiera provides me real-time transcriptions and AI-generated summaries of Wall Street events, Brighthire streamlines our hiring insights and processes with AI-powered interview summaries and transcripts, Proscia’s AI pathology platform will analyze my skin biopsy from the dermatologist, and my college-age son can use Teal to customize cover letters based on his resume and a potential job’s needs.  And that is just an illustrative sample from our small Flybridge portfolio.  Everything Everywhere All at Once, indeed.  

(It should be noted that Everything Everywhere All at Once was an incredible movie and my clear Best Picture Winner winner from 2022.)

The closest analogy is the excitement I felt when I first opened the Mosaic browser to explore the worldwide web in early 1994.  At the time, there was a rush to define companies as “web or dot com” companies, but that quickly became a meaningless distinction as every company leveraged the global connectivity and accessibility the web uniquely enabled to solve problems and create new markets.  The same will hold for today’s new “dot AI” companies, as very quickly, there will be no such thing as an AI company, as the underlying technology will be leveraged across every new and existing application to drive unique value for customers and end-users.  As with the first web applications, there will be value in being a first mover, but the real value will come from harnessing the underlying technical enablers in unique and new ways to solve real problems.  For B2B founders, this will typically mean starting with a vertical focus, incorporating the AI into existing business processes and workflows, and in a way that is not AI for AI’s sake but rather with a laser focus on driving business value. 

There is a lot of hand-wringing about whether this technical step function will change the nature of the economy, and perhaps not for the better.  Similar concerns were raised with the rise of the internet in the mid to late 90s, but instead, it unleashed a creative explosion of new ideas, tools, and solutions.  As my favorite movie of 2022 shows, there is an incredible opportunity to harness humankind’s unique creativity to build magical experiences and creative solutions to real-world problems. We could not be more excited to back this generation of founders harnessing the power of AI to build everything.

The AI Paradox

Chip Hazard2 Comments

Nobel Prize-winning Economist Robert Solow once quipped that “You can see the computer age everywhere but in the productivity statistics”. This famous observation has become known as the Productivity Paradox – the conundrum that economists face when trying to explain how is it that productivity in the US slowed down just as investments in information technology exploded in the 1980s and 1990s.

Today, we face a similar paradox, the AI Paradox. Just as corporations are investing billions of dollars in AI infrastructure, the actual impact and absorption of AI in the enterprise seems relatively muted.

Earlier this month, the MIT Technology Review published an excellent article entitled “This is why AI has yet to reshape most businesses”.  Some of the themes rhymed with our recent post about the potential for Applied AI companies, but also the real adoption barriers. It is striking that, although AI has been a central area of focus for may enterprises for years, actual adoption has lagged the hype. PWC recently conducted a survey of 1000 enterprises that are currently implementing or investigating AI and learned that only 20% plan for enterprise-wide adoption in 2019. Many projects are thus still stuck in pilot purgatory.

This dichotomy between promise and reality points to the power of Applied AI startup companies that focus intensely on the path to the adoption, and absorption, of AI into the enterprise.  Internally, we are calling this type of startup “AAA” grade – Absorbable, Applied AI.

Here’s why the approach of AAA startups matters so much in driving AI adoption::

  • First, as the article points out, the “initial payoff is often modest” of AI projects. Thus, we think it is important for Applied AI companies to “manage expectations along the way, as many Applied AI use cases falter based on overselling the potential and customers expecting too much when in reality improvements come incrementally over time”.
  • Second, the article mentions how if “companies don’t stop and build connections between such systems, then machine learning will work on just some of their data”, which strikes us as an approach a company focused solely on developing narrower, application-specific solutions can take as they can often derive insights gleaned from data that reflects “the real-world variance and dimensionality of the problem space” and can work aggressively with “input from domain experts to identify the logical gaps that exist and how to fill those gaps”.
  • Third, as “Ninety percent of the work is actually data extraction, cleansing, normalizing, wrangling”, this provides an advantage to third-party Applied AI companies, as opposed to internal DIY AI efforts, as the third party application provider–which are typically cloud-based platforms that scale across customers—is able to leverage this data wrangling cost across a wide swath of enterprises.  
  • Finally, the article points to several aspects that impede adoption including how end users need to be “attuned to how AI works and where its blind spots are” and how “In order for them to trust its judgments, they needed to have input into how it would work” which is why we feel it is critical for Applied AI companies to combine domain expertise with technical expertise, carefully collect the domain-specific requirements and use cases, use those requirements to “instantiate the models into a broader application that fits into customer’s broadly defined workflows”, and make transparency and explainability a core feature of the application, not an afterthought.  

If AI startups can apply these AAA techniques, they are going to continue to achieve superior customer traction as compared to their competitors. And, in doing, attract plenty of attention from AI-focused venture capital firms like Flybridge!

Applied AI: Beyond The Algorithms

Chip Hazard5 Comments

One of the primary areas of focus for Flybridge over the years has been to be the first institutional investor behind companies looking to transform the enterprise technology landscape with modern software.  Given the explosion in the volume of data being generated globally, this theme has led to investments in companies such as MongoDB (databases) and Nasuni (storage) that operate at the data infrastructure layer of the enterprise tech stack.

More recently, we have been investing in further advances in data management, analytics, machine learning, and artificial intelligence.  While the potential for artificial intelligence has been written about extensively, what is less well understood is that the algorithms and underlying tools are only a fraction of the value and are unlikely to be a source of long-term differentiation.  Fully realizing the power of AI requires a deep understanding of the domain and the specific workflows that AI will seek to improve and optimize. In other words, the application layer of AI ultimately drives the business value. And we believe the window of opportunity for the AI application layer is now.

This evolution from platform to applications is not an uncommon one: when a new technology platform is immature and not well understood, there is a lot of room for innovation at the underlying technology layer, but as the platforms mature the application layer is where value accrues.  For example, in the PC era, once Windows and its associated tools were well developed, apps accumulated a huge amount of value; and similarly in the early Internet era, once the browser, server, and app server infrastructure were well established, apps became super valuable.

This “Applied AI” investment thesis resulted in four new investments by Flybridge in 2018: Aiera, which is using AI to drive fundamental equity analysis; Kebotix, which is using AI to discover and create advanced chemicals and materials; Looka, which is an AI-powered graphic design platform; and Proscia, an AI-powered digital pathology solution.  We also made follow-on investments in our existing, scaling, Applied AI portfolio companies Bitsight, Bowery Farming, Datalogue, and DataXu.  

Given the breadth of this emerging portfolio, we thought it would be helpful to expand on what we look for in Applied AI companies and some of the keys to success in building a company in this exciting field.  We believe that it is important to:

  1. Go Old School. Opportunities for Applied AI companies lie outside of the markets targeted by traditional web-scale companies.  As Andrew Ng recently observed, “a lot of the stories to be told next year [2019] will be in AI applications outside the software industry”.  Many of these markets, such as Real Estate, Finance, Healthcare, Oil & Gas, Agriculture, Manufacturing, and Logistics, have the advantages of being A) extremely large, B) where innovative AI driven approaches can drive massive levels of improvement versus the status quo, C) where potential customers may not be able to access AI talent on their own, such that build versus buy is less attractive, and D) not being an area of focus of the Googles and Baidus of the world where their massive troves of data can be a source of significant competitive advantage.  
  2. Combine Talents. The most successful teams in Applied AI will have a unique combination of an understanding of the domain and the technical capabilities to realize the vision.  Even more specifically, it is doubly helpful if one of the founders was formerly a practitioner in the field. For example, Dawson Whitfield, the founder of LogoJoy, was previously a top-notch graphic designer himself; Ken Sena, was a top-ranked equity analyst before founding Aiera, and Kebotix co-founder Professor Alan Aspuru-Guzik holds a Ph.D. in Chemistry and is a leader in the field of computational chemistry. In other words, AI experts will do better when teamed up with someone that comes from the field in which they are seeking to operate.
  3. Drive continued technical innovation.  Given how quickly the field is advancing, a deeply technical co-founder who is up to speed on, and willing to continually learn about the latest advances in AI, and see the application of new approaches to the problems their company is seeking to solve is essential.  Whether it is “few-shot” learning approaches, ensemble models, GANs, CNNs, transfer learning, explainability, and a myriad of other developing techniques, knowing and understanding the strengths, weaknesses, and applicability of different approaches is critical. We often see the domain expert mentioned in point 2) renting or borrowing their AI expertise in the form of advisors and part-time experts, but this approach is not good enough given the need to have a tight feedback loop between market-driven customer needs and the AI-driven technology insights and art of the possible.
  4. Create Data Network Effects. The most successful companies will have a clear understanding and angle on how to start and continue to spin the data network effects flywheel.  Generally, this requires having access to initial datasets that can begin the model building process, and a well thought out and focused strategy on how to increase the quantities of data available for analysis.  The initial data sets, which Proscia refers to as “inorganic data”, might be acquired, and are used to overcome the cold start problem of training a new model from scratch. In contrast, “organic data” that comes from the ongoing use of the platform can help hone and refine the algorithms over time.  Taken together, this means the cost of data should decline over time because organic data is typically free (or even negative if you can get users to pay you for the service delivered while the data is collected). In the pursuit of data, it is important to remember that the sheer volume is not always inherently better. Yes, size matters, but quality matters more.  The data should reflect the real-world variance and dimensionality of the problem space and a data strategy should incorporate input from domain experts to identify the logical gaps that exist and how to fill those gaps. Further, when assessing a technical team (per point 3), we believe it is important that they know how to build an AI infrastructure that can be monitored for changing performance and updated accordingly as the scale and scope of the datasets increase.  For example, when Aiera first started making buy-sell calls on stocks, they only did so on 16 companies based on a model that analyzed 10,000 documents a week from 300 data sources. Today, they cover nearly 2,000 securities with a model that analyzes 500,000 documents a week from 22,000 data sources. Perhaps not surprisingly, the accuracy, breadth, and duration of their buy-sell calls increased significantly over this time.
  5. Absorb The Algorithm. The specific algorithms and AI techniques themselves are not the sources of defensible value so the most successful AI companies will instantiate the models into a broader application that fits into customer’s broadly defined workflows.  We call this “Absorbable AI”, which means customers can incorporate the AI into their business and realize the benefits of the operational insights. Successful AI Applications need to not only explain why the model is generating certain results (and, importantly, explainability also helps understand and highlight bias such as gender-based or racial bias), but also integrate into the customer’s business in a logical and systematic way.  It’s also important to manage expectations along the way, as many Applied AI use cases falter based on overselling the potential and customers expecting too much when in reality improvements come incrementally over time. These application and process skills are often found coming from the more traditional application development space in fields such as UX, visualization, workflow/BPM, and integrations.
  6. Craft the Business Model. Thinking through the business model of a company is critically important.  Depending on the domain, the openness of customers to new approaches, their willingness to pay for innovation, and the scale of the level of AI being incorporated, the best way to realize the value of an Applied AI company could be by selling an application that makes human work more efficient and accurate–an end-to-end automation stack that replaces humans–or it could be by selling a complete product.  For example, our indoor farming portfolio company, Bowery, leverages a significant amount of AI to drive efficiency and quality in their operation, but they decided the best way to realize the value of that AI was to sell incredibly tasty, locally grown, pesticide-free green vegetables versus selling an AI-powered Farm Operating System to other growers. A similar example would be Tesla, where the vision is the sell a complete, AI-powered, autonomous vehicle, as opposed to say Cruise, which chose to realize the value in selling the application (and the company) to other automobile producers.  

Points 5 and 6 can be better visualized in the following matrix:

Slide1

With the continued explosion of data availability, advances in AI techniques, and the accessibility and performance of computing (GPU) cycles, we believe the trend of AI as the next great application enabler will continue for some time, and we look forward to finding more Applied AI companies with passionate domain experts and technical founders to invest behind in the coming year.

Thanks to my partner Jeff Bussgang, our advisor Harini Suresh, David West of Proscia, and Bryan Healey of Aiera for their input and insights in developing these thoughts.