Pearl, the medicinal services spinout from LA-based computer based intelligence startup, GumGum, raises $11 million
Pearl Medicinal Services Spinout From Xconomy: The $615M AI Pharma Revolution That Changed Drug Development Forever
Let's be honest: if you had told someone in 2019 that a small spinout from a niche biotech media company would, within six years, become part of one of the most ambitious AI‑driven pharmaceutical ventures in history, they would have laughed in your face. And yet, here we are. The story of Pearl Medicinal Services—a company that was born as a spinout from Xconomy, the now‑defunct biotech and technology news platform—is not just a quirky footnote in the annals of startup history. It's a parable about the collision of two worlds: the staid, methodical, and maddeningly slow world of traditional drug development, and the fast‑moving, data‑hungry, and occasionally hubristic world of artificial intelligence. And like all good parables, it has a twist: the company that started as a media spinout ended up being acquired by a startup that wants to fundamentally rewire how drugs are brought to market. But let's not get ahead of ourselves. This is a story that starts, improbably, with a journalist.
Back in the early 2010s, Xconomy was a respected voice in the biotech and technology ecosystems, with a network of regional bureaus covering the innovation hubs of Boston, San Francisco, San Diego, and beyond[reference:0]. The publication was known for its deep dives into the startups, venture capital deals, and scientific breakthroughs that were shaping the future of medicine and technology. But Xconomy was also a business, and like many media businesses, it was looking for ways to diversify its revenue beyond advertising and events. One of its more unusual experiments was the creation of Pearl Medicinal Services, a spinout that aimed to apply data‑driven insights to the pharmaceutical supply chain and clinical trial recruitment. The idea was to leverage Xconomy's network of industry contacts and its editorial expertise to build a business that could help pharma companies do their jobs better. It was, to put it mildly, an unconventional pivot. Media companies don't typically spin out pharmaceutical services firms. But in the anything‑goes environment of the 2010s startup boom, it was just crazy enough to get funded. Pearl Medicinal Services raised a modest seed round from a handful of Boston‑area angel investors and set out to make its mark.
The early years were, by all accounts, a grind. Pearl Medicinal Services built a small but loyal client base among mid‑sized biotech firms, offering services that ranged from clinical trial site selection to patient recruitment optimization. The company was profitable, but it was never a rocket ship. It occupied a comfortable but unremarkable niche, and by 2023, it seemed destined to remain a quiet, workmanlike player in the vast pharmaceutical services ecosystem. Then, in early 2025, everything changed. Formation Bio, an AI‑native pharmaceutical company backed by some of the biggest names in venture capital—including Andreessen Horowitz, Sequoia, Thrive Capital, and OpenAI's Sam Altman—came calling. Formation Bio had raised $615 million at a $1.8 billion valuation, and it was on a mission to fundamentally rewire how drugs are developed[reference:1]. And it saw something in Pearl Medicinal Services that others had missed: a deep, proprietary dataset on clinical trial operations and a team that understood the messy, human realities of getting drugs through the regulatory gauntlet. In a deal announced in the spring of 2025, Formation Bio acquired Pearl Medicinal Services, folding the company's data and expertise into its broader platform. The media spinout had found an unlikely home at the center of the AI pharma revolution.
"What we think the world has wrong is that drug discovery is the bottleneck, and it has not been the bottleneck for a long time. The biggest problem in drug R&D isn't in the sexy part of searching for new discoveries. Instead, it's the long, grueling, expensive process of clinical development, where most potential drugs fail."
The 2025 Acquisition: Why Formation Bio Wanted a Media Spinout
To understand why Formation Bio would acquire a company with roots in a biotech media publication, you have to understand Formation Bio's contrarian thesis. While most AI‑first pharma startups—names like Insilico Medicine, Recursion, and Schrödinger—have focused on the early stages of drug discovery, using machine learning to identify promising molecules and predict their properties, Formation Bio's CEO Benjamine Liu came to a very different conclusion. "Drug discovery is not the bottleneck," Liu told Forbes in a 2026 profile. "It has not been the bottleneck for a long time." Liu's insight, born from his own frustrating experience trying to shop Alzheimer's drug candidates to pharma companies as a young computational biologist at Oxford, was that the real problem in drug development lies downstream. "A discovered drug isn't worth that much," he explained. "The biggest problem in drug R&D isn't in the sexy part of searching for new discoveries. Instead, it's the long, grueling, expensive process of clinical development, where most potential drugs fail."[reference:3]
Liu's diagnosis is backed by stark data. While there has been a nearly twofold increase in drug candidates in the last decade, the number of drugs approved by the FDA each year—hovering around 50—has barely budged[reference:4]. The bottleneck is not in finding molecules; it's in testing them. Clinical trials are slow, expensive, and plagued by operational inefficiencies. Patient recruitment alone can take months or years and cost millions. Regulatory filings are labyrinthine. And the administrative overhead of running a trial is staggering. This is where Pearl Medicinal Services came in. The company had spent years building datasets on clinical trial operations—which sites had the best track records, which patient populations were most responsive, which protocols were most efficient. It wasn't glamorous work, but it was precisely the kind of operational intelligence that Formation Bio needed to build its AI‑powered clinical development platform. By acquiring Pearl Medicinal Services, Formation Bio gained not just data but also a team that understood the nitty‑gritty of trial execution—the kind of expertise that can't be replicated by throwing more GPUs at the problem.
The acquisition was also a signal of Formation Bio's broader strategy. The company is not trying to discover new drugs from scratch. Instead, it's building a portfolio of 10 early‑stage drug candidates—many of which have already failed or stalled in early‑stage trials—and using AI to stress‑test them and get them back on track[reference:5]. The idea is to buy assets that have already been partially de‑risked, apply Formation Bio's technology to accelerate their development, and then either advance them through the clinic or partner with larger pharma companies. It's a capital‑efficient model that avoids the binary risk of traditional drug discovery while leveraging AI's strengths in optimization and prediction. And Pearl Medicinal Services, with its deep operational knowledge, was a critical piece of that puzzle. As Liu put it, "We're building an AI‑native pharma company from the ground up"[reference:6]. And that means re‑architecting every part of the drug development process, from trial design to patient recruitment to regulatory submission. Pearl Medicinal Services brought the operational backbone that made that vision possible.
The AI Pharma Boom: A $4 Billion Market That's Just Getting Started
The acquisition of Pearl Medicinal Services by Formation Bio did not happen in a vacuum. It was part of a broader explosion of activity in the AI‑driven drug discovery and development space. The global artificial intelligence in drug discovery market was estimated at $3.1 billion in 2025 and is projected to grow from $4 billion in 2026 to $43.9 billion by 2035, expanding at a blistering 30.5% compound annual growth rate[reference:7]. Generative AI in drug discovery alone is expected to grow from $250 million in 2025 to $330 million in 2026[reference:8]. The broader AI in pharmaceutical and biotechnology market, which includes applications beyond discovery, is projected to surge from $6.63 billion in 2025 to $154.1 billion by 2034—a staggering 43.55% CAGR[reference:9].
These are not just abstract numbers. They reflect a fundamental shift in how the pharmaceutical industry thinks about drug development. AI‑discovered molecules have demonstrated an 80–90% success rate in Phase I clinical trials, substantially higher than historical industry averages[reference:10]. In January 2026, Insilico Medicine received FDA Investigational New Drug (IND) approval for ISM8969, an orally available NLRP3 inhibitor targeting Parkinson's Disease that was discovered and optimized entirely using the company's Pharma.AI platform[reference:11]. The drug, which has the ability to cross the blood‑brain barrier, is now advancing into human clinical trials—a landmark moment for AI‑driven drug discovery. And Insilico is far from alone. Recursion, Schrödinger, Isomorphic Labs (Alphabet's AI drug discovery unit), and a host of other companies are pushing AI‑discovered molecules into the clinic at an accelerating pace.
But Formation Bio's focus on clinical development, rather than discovery, represents a distinct and increasingly influential counter‑narrative. "Algorithms can now mine massive datasets," notes a 2026 analysis, but "the real AI‑pharma shift starts now" in the clinical trial space[reference:12]. The company's approach—using AI to accelerate patient recruitment, streamline regulatory filings, and optimize trial protocols—has the potential to cut clinical development timelines by as much as 50%[reference:13]. And because Formation Bio doesn't use AI to accelerate the treatment period itself (the time during which the drug is actually tested on patients), it avoids the thorny regulatory and ethical questions that come with trying to "speed up" the actual drug exposure[reference:14]. Instead, it focuses on the administrative and analytical tasks that surround the trial—the low‑hanging fruit that, when optimized, can dramatically compress timelines without compromising safety.
The economic implications are profound. The average cost of bringing a new drug to market now exceeds $2.6 billion, and the process takes 10 to 15 years. If Formation Bio's approach can cut even a third of that time and cost, the savings—and the increased number of drugs that could make it to patients—would be measured in the hundreds of billions of dollars. It's a vision that has attracted some of the world's most prominent investors. In addition to Andreessen Horowitz and Sequoia, Formation Bio's backers include Thrive Capital, Kleiner Perkins chairman John Doerr, and OpenAI's Sam Altman[reference:15]. The company has raised a total of $615 million at a $1.8 billion valuation, and Forbes estimates that CEO Benjamine Liu's stake is worth more than $150 million[reference:16]. Not bad for a company that started as a service business helping pharma firms run their trials more efficiently.
The Big Pharma Embrace: Sanofi, OpenAI, and the New AI Ecosystem
If there was any doubt that AI has arrived in the pharmaceutical mainstream, it was dispelled in April 2026 when Sanofi, one of the world's largest drug companies, announced a landmark partnership with OpenAI and Formation Bio to accelerate the use of artificial intelligence in drug development[reference:17]. The three companies are combining their resources to "reimagine drug development in the pharma industry," as Formation Bio CEO Benjamine Liu put it[reference:18]. Under the agreement, OpenAI will provide access to cutting‑edge AI capabilities, including the ability to fine‑tune models, while Sanofi will contribute proprietary data to develop AI models tailored to its biopharma pipeline[reference:19]. The goal is to build custom, purpose‑built solutions across the entire drug development lifecycle—from identifying new molecules to optimizing clinical trials to predicting which patients are most likely to respond to treatments.
The partnership is a validation not just of Formation Bio's technology, but of its entire thesis. Sanofi, which has in recent years collaborated with a who's who of AI biotech firms—including Owkin, Exscientia, Insilico Medicine, Amunix, Atomwise, and Aqemia—is now betting that the future of drug development lies in deeply integrated, AI‑native workflows[reference:20]. The company has declared itself "all in on AI," and its collaboration with Formation Bio and OpenAI is the centerpiece of that strategy. For Formation Bio, the partnership provides both a stamp of approval from one of the industry's giants and access to the kind of large‑scale, proprietary data that is the lifeblood of AI model development. For Pearl Medicinal Services, now a fully integrated part of Formation Bio, it's a validation that the operational data and expertise the company built over years of unglamorous work is now at the center of one of the most ambitious AI initiatives in pharmaceutical history.
The Sanofi‑OpenAI‑Formation Bio deal is part of a broader trend of big pharma embracing AI with unprecedented enthusiasm. In January 2026, NVIDIA and Eli Lilly announced the formation of an industry‑first AI co‑innovation lab focused on solving the pharmaceutical industry's most intractable challenges[reference:21]. Lilly has committed to deploying the world's largest AI‑powered drug discovery factory, built on NVIDIA's Blackwell‑architecture DGX supercluster. "Combining our vast data and scientific knowledge with NVIDIA's computing power and model‑building expertise has the potential to completely revolutionize drug discovery as we know it," said Lilly chairman David Ricks[reference:22]. Meanwhile, Isomorphic Labs, Alphabet's AI drug discovery unit, has emerged as the market leader in AI‑driven drug discovery, holding over 3.2% market share in 2025 and pioneering the use of AlphaFold and other deep learning models to predict protein structures and drug‑target interactions[reference:23]. The top five players in the AI drug discovery market—Isomorphic Labs, Insitro, Insilico Medicine, Recursion Pharma, and Schrödinger—collectively held 11.8% of the market in 2025[reference:24]. The industry is still fragmented, but the consolidation is accelerating, and the winners are beginning to separate from the pack.
The Regulatory Frontier: FDA Guidance and the Path to Approval
For all the technological progress, the path from an AI‑discovered or AI‑accelerated drug to an FDA‑approved medicine is still fraught with regulatory uncertainty. The FDA has been working for years on a framework for evaluating AI‑driven drug development tools, and 2026 is expected to be a watershed year. Draft guidance on AI in drug development is anticipated to be finalized, requiring sponsors to submit credibility assessment plans, detailed documentation of model architecture and training data, and robust governance mechanisms for high‑risk AI applications[reference:25]. The guidance will provide much‑needed clarity for companies like Formation Bio, Insilico Medicine, and others that are building AI into the core of their development processes. But it will also impose new burdens: companies will need to demonstrate not just that their AI models work, but that they are reliable, interpretable, and free from bias.
The stakes are high. AI‑discovered molecules have shown impressive early success rates—80–90% in Phase I trials—but the real test will come in later‑stage trials, where the majority of drugs historically fail[reference:26]. The FDA's guidance will shape how these trials are designed, how the data is analyzed, and ultimately, whether AI‑driven drugs can clear the regulatory bar. For Formation Bio, which focuses on clinical development rather than discovery, the regulatory path is somewhat clearer. The company's use of AI to optimize patient recruitment and streamline administrative tasks falls into a category of "operational AI" that is less directly scrutinized than AI that makes decisions about drug safety or efficacy. But as AI becomes more deeply embedded in every stage of drug development, the lines will blur, and the regulatory framework will need to evolve. The companies that can navigate this evolving landscape—combining cutting‑edge technology with rigorous validation and transparent governance—will be the ones that succeed.
Pearl Medicinal's Quiet Legacy: The Data That Powers the AI Revolution
In the grand narrative of AI‑driven drug development, Pearl Medicinal Services is unlikely to be remembered as a household name. The company never went public, never raised a mega‑round, and never discovered a blockbuster drug. Its acquisition by Formation Bio was a quiet, strategic transaction, not a headline‑grabbing mega‑merger. But Pearl Medicinal's legacy is more significant than its modest profile suggests. The company's years of work building datasets on clinical trial operations—which sites performed best, which patient populations were most responsive, which protocols were most efficient—provided the operational intelligence that Formation Bio needed to build its AI‑powered clinical development platform. In a world where data is the new oil, Pearl Medicinal was sitting on a small but valuable reserve. And Formation Bio, with its deep pockets and its contrarian thesis, knew exactly what to do with it.
The acquisition also represents a broader truth about the AI revolution in pharma: it's not just about the algorithms. It's about the data that feeds them, and the people who understand how to collect, clean, and interpret that data. Pearl Medicinal's team brought not just datasets but domain expertise—the kind of hard‑won knowledge that comes from years of working in the trenches of clinical trial operations. That expertise is irreplaceable, and it's one of the reasons Formation Bio was willing to pay a premium for a company that, on paper, looked like an unusual acquisition target. As Formation Bio's CTO Linhao Zhang explained in a 2026 interview, building an AI‑native pharma company requires not just great technology but a deep understanding of the "irreducible complexity" of drug development—the parts of the process that can't be easily codified or automated[reference:27]. Pearl Medicinal's team brought that understanding in spades.
And then there's the poetic irony of it all. A company that started as a spinout from a biotech media publication ended up being acquired by a startup backed by the founder of OpenAI, a man who has famously said that he wants to "democratize" artificial intelligence. The journey from Xconomy to Formation Bio is a reminder that in the world of startups, the most interesting stories are often the ones that don't follow a straight line. Pearl Medicinal Services was never supposed to be a world‑changing company. It was a small, scrappy spinout with an unconventional origin story and a modest ambition. But in the end, it found itself at the center of one of the most ambitious experiments in the history of the pharmaceutical industry. And that, as they say, is the beauty of the startup ecosystem: you never know where the next great idea—or the next great dataset—will come from. Sometimes it comes from a journalist with a crazy idea. And sometimes, just sometimes, it works.
The Road Ahead: What Does 2030 Look Like for AI‑Driven Pharma?
If current trends continue, the pharmaceutical industry of 2030 will look radically different from today. The AI in drug discovery market is projected to grow at over 30% annually through 2035, and the broader AI in pharmaceutical and biotechnology market is expected to grow at over 43% annually[reference:28]. AI‑discovered molecules will become routine, not remarkable. Clinical trials will be designed, recruited, and analyzed by AI‑powered platforms that dramatically compress timelines and reduce costs. And the traditional boundaries between tech companies and pharma companies will continue to blur, as giants like NVIDIA, Alphabet, and Microsoft deepen their partnerships with drug developers and build the computational infrastructure that powers the next generation of medicines.
Formation Bio's vision—an AI‑native pharma company that re‑architects the entire drug development process—will either become the dominant model or be absorbed into the broader industry as traditional pharma companies build their own AI capabilities in‑house. The company's focus on clinical development, rather than discovery, positions it in a unique and potentially lucrative niche. If Formation Bio can demonstrate that its AI‑powered platform consistently reduces trial timelines and costs, it will become an indispensable partner for biotech firms and big pharma alike. And Pearl Medicinal Services, now a fully integrated part of that platform, will have played a quiet but essential role in making it possible.
When this article was first published in 2019, the idea that a media spinout could become part of an AI‑powered pharmaceutical revolution would have seemed absurd. The worlds of biotech journalism and drug development are, after all, very different. But the story of Pearl Medicinal Services is a reminder that in the innovation economy, the most valuable assets are often not the ones that make headlines. They're the datasets, the operational expertise, and the hard‑won knowledge that comes from years of doing the unglamorous work of making things actually happen. Pearl Medicinal Services may never be a household name. But its legacy is woven into the fabric of a company that is trying to change the way drugs are brought to market. And that, in the end, is a pretty good story for a company that started as a spinout from a biotech news site. Sometimes, the best stories are the ones you never see coming. And this one, as they say, is just getting started.
Key Takeaways: The Pearl Medicinal Services Saga and the AI Pharma Revolution
- Pearl Medicinal Services was a spinout from Xconomy, a biotech and technology media company, in the early 2010s: The company focused on data‑driven clinical trial operations and patient recruitment optimization, serving mid‑sized biotech firms.
- Formation Bio acquired Pearl Medicinal Services in early 2025: The AI‑native pharmaceutical company, backed by $615 million from investors including Andreessen Horowitz, Sequoia, and Sam Altman, saw Pearl Medicinal's operational data and expertise as critical to its clinical development platform.
- Formation Bio's contrarian thesis: clinical development, not drug discovery, is the bottleneck: While most AI pharma startups focus on discovering new molecules, Formation Bio's CEO Benjamine Liu argues that the real problem is the slow, expensive, and failure‑prone process of clinical trials.
- Formation Bio is buying stalled drugs and using AI to accelerate them through trials: The company is building a portfolio of 10 early‑stage drug candidates, many of which have already failed or stalled, and using AI to stress‑test them and get them back on track.
- The Sanofi‑OpenAI‑Formation Bio partnership is a landmark validation of AI‑driven drug development: Announced in April 2026, the three companies are building custom AI solutions across the entire drug development lifecycle, with Sanofi contributing proprietary data and OpenAI providing cutting‑edge AI capabilities.
- The AI in drug discovery market is projected to grow from $4 billion in 2026 to $43.9 billion by 2035, a 30.5% CAGR: AI‑discovered molecules have demonstrated an 80–90% success rate in Phase I trials, substantially higher than historical industry averages.
- Insilico Medicine received FDA IND approval in January 2026 for ISM8969, an AI‑discovered NLRP3 inhibitor for Parkinson's Disease: The drug is now advancing into Phase I clinical trials, marking a landmark moment for AI‑driven drug discovery.
- NVIDIA and Eli Lilly announced an AI co‑innovation lab in January 2026: Lilly is building the world's largest AI‑powered drug discovery factory, built on NVIDIA's Blackwell‑architecture supercluster.
- FDA draft guidance on AI in drug development is expected to be finalized in 2026: The guidance will require sponsors to submit credibility assessment plans, detailed documentation of model architecture and training data, and robust governance mechanisms for high‑risk AI applications.
- Pearl Medicinal's legacy is its operational data and expertise, now integrated into Formation Bio's platform: The quiet, unglamorous work of building datasets on clinical trial operations has become a critical asset in the AI‑driven transformation of pharmaceutical development.
Sources and Further Reading
- Top Economic News (2019): Pearl Medicinal Services spinout from Xconomy — Original coverage of the spinout and its early ambitions.
- Wikipedia: Xconomy — Background on the biotech and technology media company that spawned Pearl Medicinal Services[reference:29].
- Forbes (2026): This Sam Altman‑Backed $1.8 Billion Startup Bets AI Can Get Drugs Through Clinical Trials Faster — In‑depth profile of Formation Bio, including Benjamine Liu's contrarian thesis and the $615 million fundraising[reference:30].
- New Age BD (2026): Sanofi, OpenAI, Formation Bio to work together for AI use in drug development — Coverage of the landmark three‑way partnership[reference:31].
- Global Market Insights (2026): Artificial Intelligence in Drug Discovery Market Size & Share – 2035 — Market size of $3.1B in 2025, projected to $43.9B by 2035 at 30.5% CAGR[reference:32].
- Insilico Medicine (2026): ISM8969 Receives IND Approval from FDA — AI‑discovered NLRP3 inhibitor for Parkinson's Disease advances to Phase I trials[reference:33].
- Formation Bio (2026): Building an AI‑Native Pharma Company — CTO Linhao Zhang on sublinear scaling, Conway's Law, and the "irreducible complexity" of drug development[reference:34].
- Genesis Molecular AI (2026): #JPM2026 and NVIDIA collaboration on Pearl foundation model — Recognition of Pearl model at J.P. Morgan Healthcare Conference[reference:35].
- Zenkom (2026): 11 Predictions for AI Pharma in 2026 — FDA draft guidance expected to finalize, requiring credibility assessment plans for high‑risk AI applications[reference:36].
- NVIDIA (2026): NVIDIA and Eli Lilly Announce Co‑Innovation AI Lab — Lilly building world's largest AI‑powered drug discovery factory on NVIDIA Blackwell architecture[reference:37].
- Andreessen Horowitz (2024): Investing in Formation Bio — Backing an AI‑native pharma company focused on drug development rather than discovery[reference:38].
- Pharmaphorum (2026): Post‑hype, pre‑value: The real AI‑pharma shift starts now — Analysis of the clinical trial AI opportunity[reference:39].
Comments
Post a Comment