Morgan Cheatham
Change happens slowly, and then all at once — especially in complex industries like healthcare.
Just five years ago, venture capital investments in healthcare AI were emerging and exploratory. Half a decade and one global pandemic later, we’re living in a brave and more ambitious new world defined by an unbridled enthusiasm for leveraging revolutionary technologies like AI.
Pointing this technology at previously intractable problems in key industries such as healthcare, life sciences, and beyond is among the greatest opportunities of the century.
The year 2022 was when the broader public bore witness to material advancements in AI research that have matured from lab to life. ChatGPT educated over 100 million people globally about transformers in just two months.
What was once a nascent area of research has now become venture capital’s next platform shift, and with that, investors ask, “How will generational AI companies be built in healthcare, life sciences, and beyond?”
AI-first companies are in the business of advancing AI as a science, whereas AI-enabled companies are implementation and distribution machines. The two company phenotypes establish moats at different layers — AI-first companies innovate just above silicon, while AI-enabled companies create enterprise value at the application level.
For founders, knowing what kind of company you are building is essential for recruiting proper talent, partnering with aligned investors, securing sufficient capital, and deploying a viable business model. AI-first companies require deep AI research acumen, investors willing to take a long view, materially more capital, and potentially less conventional business models than AI-enabled peers.
In reality, this distinction is a spectrum, not a binary. Impactful companies will be built with both approaches. For AI-first companies, though, we believe the fruits will be worth the labors.
Influence over the technology stack from the ground up enables tight control over cost structure, immeasurable product optionality, and greater defensibility relative to AI-enabled companies that defer the exercise of scientific inquiry to those that are AI first.
We can no longer afford to conflate AI-first and AI-enabled companies. So far, the largest AI-first companies have been built for horizontal applications (e.g., OpenAI, Cohere, Anthropic); yet vertical, industry-specific platforms, such as those in healthcare and life sciences, will showcase the expansive capabilities of large-scale models to deliver real-world impact.
For founders, we believe enduring AI-first companies — in healthcare, life sciences, and beyond — will follow these six imperatives.
Create and sustain an undeniable data advantage
AI-first companies exhibit an insatiable appetite for data and employ creative means for acquiring it sustainably. In addition to amassing large and robust datasets, AI-first companies develop designer datasets that are uniquely suited to deliver high performance on specific tasks.
Designer datasets are unique in that they are not easily found in public; they are machine readable, in that they are ingestible by AI models; and they are scalable, in that it is tractable to generate high volumes over time.
Importantly, designer datasets are not simply the exhaust of processes within a given system, and they are not generated by customers alone. For example, the healthcare and life sciences industries generate 30% of the world’s data, and yet companies that train only on existing electronic health record data or resources like PubMed leave material performance gains and capabilities behind.
Designer datasets may require authoring experimental protocols for situations that do not occur naturally but that deliver strong model performance for a given task.
For example, Subtle Medical, an AI-first company focused on imaging acceleration, generated millions of imperfect MRI images captured in 15 minutes, which were later utilized to train deep learning models that could reconstruct and de-noise medical imaging exams taken in shorter periods of time. In practice, imperfect MRI images provide little clinical value; however, as an AI-first company, these images trained deep neural networks that created a data moat for Subtle’s technology.
Reinforcement learning with (expert) human feedback — RL(E)HF — is another critical tool for AI-first companies. RLHF is a technique where an AI system learns and improves its performance by receiving feedback from human input. With RL(E)HF, expert human feedback provided by individuals trained in particular disciplines such as neurology or structural biology can tune model outputs for high performance in that domain.
Abridge, an AI-first company that provides ambient documentation tools for clinicians, leverages clinician feedback on AI-authored notes to enhance note accuracy and quality across specialties.
Data derived from customers also creates flywheels of opportunities for generating novel and defensible data assets. After establishing product-market fit, AI-first companies can leverage this position to serve adjacent customer segments. By capturing and integrating datasets across stakeholders in a given industry, AI-first companies can strengthen data advantages, unlock TAM, and create new categories.
Recruit and empower AI scientists
AI-first companies require “multilingual” teams — meaning they employ scientists deeply skilled in AI research, as well as individuals with industry and business expertise. In healthcare or life sciences, this might look like clinicians and scientists partnering with AI researchers to design models with context-aware representations for a given domain. AI-first companies are also more likely to benefit from an academic or industry laboratory affiliation. For example, Atropos Health, an AI-first company focused on real-world data generation for medicine, initially spun out of a Stanford AI Lab helmed by Dr. Nigam Shah.
The organizational structure must also reflect an AI-first company’s prioritization of AI from the most senior levels. The R&D organization at an AI-first company will likely operate under a different reporting structure and leadership profile compared to an AI-enabled company. AI-first companies are more likely to have a chief scientific officer (CSO) with deep AI research experience, with AI researchers and software engineering resources reporting into the CSO.
AI-enabled companies are more likely to have chief technology officers with classical software engineering training. The marketing function at AI-first companies also serves AI research as a key business activity. AI-first companies publish work regularly via accessible formats such as peer-reviewed journals or presentations at leading AI conferences. These activities are critical for demonstrating advancements in state-of-the-art (SOTA) and contributing broadly to advancements in the field of AI.
Support a flexible AI stack
AI is advancing at an exponential pace as model sizes scale nonlinearly and facilitate new behaviors. AI-first companies refrain from making rigid, irreversible, or monolithic decisions about the AI stack. AI-first companies do not seek to develop models that perform above SOTA benchmarks for every task or domain required to serve their customers.
Instead, they build modular AI stacks that leverage publicly available models (open source or maintained by other vendors) where incumbents boast best-in-class performance. AI-first companies focus proprietary model development resources in layers of the stack where they have clear advantages due to undeniable data moats, methodological intellectual property, or other contributions of company or affiliated laboratory research.
AI-enabled companies are more likely to rely on incumbent infrastructure, such as GPT, for a majority of AI-related product features. These companies may perform fine-tuning on top of these models, but as a result are more likely to hit a ceiling in terms of product capabilities. By innovating lower in the AI stack, AI-first companies enjoy greater product and feature optionality over time.
Establish distribution moats
Without distribution, the impact of a best-in-class AI model may start and end with a published paper. Deploying AI models into production requires commercialization that enables companies to access customer data and/or contract with end users to perform optimization tasks, such as RL(E)HF. Both AI-first and AI-enabled companies should seek to obtain distribution moats or advantages early on, such that it’s conceivable the product will be able to integrate with or displace incumbent technology providers over the near term.
Unlike AI-first companies, however, distribution is often the only moat for AI-enabled companies. AI-first companies can benefit from both technical and distribution moats, increasing the probability of success. Iterative Health, an AI-first precision gastroenterology company pioneering novel biomarkers for gastrointestinal disease, secured an exclusive partnership with Provation, the leading gastroenterology electronic medical record, to facilitate broad distribution in the specialty.
Business model innovation, including strategies such as product-led growth with core user groups, may accelerate uptake of AI-first products in the enterprise.
Center safety and ethics in model development
As AI permeates all aspects of public and private life, AI-first and AI-enabled companies must grapple with preserving the foundational rights of human users. Investment in AI safety and ethics is non-negotiable for AI-first companies innovating at the foundational layer.
These companies must exercise caution and intention in data custodianship and a relentless commitment to model maintenance. Strategies such as continuous performance monitoring, fail-safes and overrides for human intervention, recurring revalidation against real-world data, and user training that outlines key limitations of AI are employed exhaustively by AI-first companies.
Earn and maintain trust
While AI-first companies are built on data, they thrive on trust. Trust is earned when stakeholders perceive that their problems are met with curiosity and empathy rather than techno-solutionism. It is earned through reliability, accuracy, and respect for the human condition. By adhering to the aforementioned guidance, AI-first companies can earn and maintain the trust of their users, customers, and their industry at large.
Setting the record straight
Building AI-first companies, especially in healthcare and life sciences, is not an easy feat. However, the impact of AI-first companies will be greater, financial returns superior, and moats more enduring than their AI-enabled counterparts. Though we’re in the earliest days of witnessing AI-first companies in the wild, including industry-specific opportunities in healthcare and life sciences, the stark contrast in the approach, capabilities, and leadership of AI-first companies clearly distinguishes them from traditional software or AI-enabled businesses.
AI research will continue to be the lifeblood of generational opportunities in AI. It is imperative that the venture capital and startup ecosystems commit to distinguishing AI-first companies from AI-enabled and become students of how these companies are built and scaled. As we continue to interrogate the role of AI as an agent for problem-solving, I implore founders to think deeply about what kind of AI company they seek to build and what that will mean for the path ahead.
Abridge and Subtle Medical are Bessemer Venture Partners portfolio companies.
Comment