What is one specific practice or strategic choice that enables biotech companies to accelerate innovation, and why?

Many answers highlight leveraging AI, organoids, and parallelization to speed up drug development, while others focus on regulatory streamlining, organizational structure, and early decision-making to accelerate innovation.

18 answers

Answers

  • Michael Ehlers

    Entrepreneur Partner · MPM BioImpact

    Shared AI Platform

    Michael Ehlers says Formation's drug programs all run on the same development organization and shared AI platform, which he says gives the company both scale and speed.

    Decoding Bio · Decoding Bio · Read
  • Samir Kaul

    Founding General Partner · Khosla Ventures

    Use organoids

    Samir Kaul says the FDA is considering using organoids instead of animal models, and argues that if allowed, this could compress drug development cycles because organoids are more relevant, quicker, easier to manipulate, and yield data faster.

    The answer starts at 7:13.

    MTS · Listen on Apple Podcasts
  • Kevin Weil

    CEO

    Build large-scale automated labs

    Kevin Weil says the way to make AI models strong in science is to build large-scale automated labs that can do science at volume, because the data they generate will help make AI models incredible in the sciences.

    The answer starts at 11:01.

    MTS · Listen on Apple Podcasts
  • Richard Socher

    CEO · Recursive

    Organoids for testing

    Richard Socher says organoids are tiny mini organs that mimic parts of a real organ, and cites Pearl Bio's FDA approval to skip animal trials for immune therapy toxicity testing, which he says can get drugs to market faster.

    Super Data Science: ML & AI Podcast with Jon Krohn · Watch on YouTube
  • Joshua Weiner

    Senior Vice President Consumer Engagement & Analytics · Cvs Health

    Speed up iteration cycles

    Joshua Weiner says the key is to compress the timeline by linking the stages, noting that a better, more potent molecule could make the clinical trial easier.

    The answer starts at 18:55.

    Stanford Online: University Courses & Lectures · Listen on Apple Podcasts
  • Eric Newcomer

    Founder / Writer · Newcomer

    Wet-lab sandbox

    Eric Newcomer says Anthropic's wet lab is a sandbox for pursuing basic research as an end in itself, where a basic research team tests their tools and maximizes use of Claude.

    The answer starts at 28:33.

    Stanford Online: University Courses & Lectures · Listen on Apple Podcasts
  • Patrick Hsu

    Co-Founder and Core Investigator · Arc Institute

    Faster trial enrollment

    Patrick Hsu says innovations in clinical trial regulations, enrollment, recruitment and endpoints are needed to test things faster in people, since human response data is the most valuable thing in the industry.

    The answer starts at 1:19:16.

    Dialectic · Listen on Apple Podcasts
  • Simeon George

    Chief Executive Officer and Co-Founder · Sr One Capital Management

    Reduce regulatory friction

    Simeon George says cutting friction in regulatory frameworks and pathways is a way to move molecules forward efficiently, citing clarity about end posts and consistency in applying rules as what matters to him.

    Life Science Connect · Watch on YouTube
  • Scott Galloway

    Author, “Notes on Being a Man” · Simon & Schuster

    AI for admin tasks

    Scott Galloway says AI helps drug companies with the administrative tasks that slow down drug development, discovery and testing, citing Moderna's openness about its AI use.

    The Prof G Pod – Scott Galloway · Watch on YouTube
  • Peter H. Diamandis

    Founder & Curator · Abundance360

    Parallel organoid testing

    Peter H. Diamandis cites Parallel Bio, which builds tiny organoids from pluripotent stem cells and tests drugs on them in parallel; he says this is more predictive of how drugs interact in real human bodies, and that the company got FDA approval to skip animal trials.

    The answer starts at 17:52.

    Moonshots with Peter Diamandis · Listen on Apple Podcasts
  • Mukhtar Ahmed

    Chief Executive Officer · BC Platforms

    Start with the decision

    Mukhtar Ahmed says companies should start by defining the decision rather than buying a broad AI platform, because technology creates more value when applied to an explicit decision framework and cannot compensate for ambiguous criteria.

    The answer starts at 15:46.

    The AI in Business Podcast · Listen on Apple Podcasts
  • Allan Shaw

    CFO & CBO · CREATE Medicines

    Build credibility via feedback

    Allan Shaw suggests biotech companies develop credibility by outlining what they intend to do and getting feedback, since the people they talk to will often say what they are looking for or might need.

    Life Science Connect · Watch on YouTube
  • Sylke Poehling, PhD

    SVP and Global Head: Therapeutic Modalities, pRED; Roche Gene Therapy Center of Excellence · Roche

    AI-driven high-throughput design

    Sylke Poehling says AI's clearest value is in design, where large libraries of variants can be built and tested in parallel so a model can learn which sequences have the needed properties, because AI helps most where ideas can be tested quickly and cheaply.

    Life Science Connect · Watch on YouTube
  • Jake Auchincloss

    U.S. Representative · Us Congress

    Fund valley-of-death companies

    Jake Auchincloss says it is essential to keep funding valley-of-death stage companies, because planting those seeds takes 10 to 15 years to produce innovations.

    The answer starts at 28:03.

    BiotechTV - News · Listen on Apple Podcasts
  • Noubar Afeyan

    Founder and CEO · Flagship Pioneering

    Parallel company creation

    Noubar Afeyan says his firm chose to start a few companies at once rather than one at a time, pooling capital to get them jointly created and advanced, and then see what happens.

    The answer starts at 14:45.

    Building One with Tomer Cohen · Listen on Apple Podcasts
  • Olivier Litzka

    Partner · Andera Partners

    Cut losses early

    Olivier Litzka says biotech investors should always think about how to cut losses early, because when the underlying scientific rationale does not deliver, there may be no way forward, and that is how risk is balanced.

    Business Of Biotech · Business Of Biotech · Listen at 16:36
  • Fiona H. Marshall

    President, Biomedical Research · Novartis

    Data scientists in leadership

    Fiona H. Marshall says she required each of her leadership team members to have a data scientist on their team, so that someone who understands the technology is at every table and can help design experiments fit for AI analysis.

    Vanguards of Healthcare by Bloomberg Intelligence · Vanguards of Healthcare by Bloomberg Intelligence · Listen at 44:51
  • Eric Nguyen

    Co-Founder & CEO · Radical Numerics

    Tokenize multi-omics data

    Eric Nguyen says his company figured out how to tokenize modalities such as proteins, RNA, epigenomics and metabolomics so large language models can read them, aiming to integrate them into a single system rather than model one siloed modality at a time.

    The Neuron: AI Explained · The Neuron: AI Explained · Listen at 17:44

More Biotechnology questions

View all →