Christopher Cole

Based on Manufacturing Intelligence Episode 6, Chris Cole and Richard Jaenisch discuss the risks of using AI in Pharma Development and Manufacturing.

In episode 6 of this video interview the series, Manufacturing Intelligence

 Chris Cole, spoke with Richard Jaenisch, senior director of education, outreach, and digital experience at Open Biopharma, about where artificial intelligence (AI) drug discovery actually stands, how the Pope's recent commentary on AI ethics compares to the embryonic stem cell debate, and where AI agents are, and aren’t, worth the security risk in manufacturing.

PharmTech: Where Is the Line Today of What AI Can Do in Discovery?

 There are really 2 approaches under the same umbrella. One mines existing data to identify gaps scientists might have missed, then applies that insight toward a new formulation. Scientists stay in control here, they can look at the output and say, ‘This makes sense,’ because they understand the underlying pathways. The other is closer to generating massive volumes of candidates and hoping something useful emerges, which I'd describe as chimpanzees writing Shakespeare. Give it enough attempts and you'll eventually get something, but I haven't seen that method produce a real success yet. The pathway-driven approach is where I expect real movement, since scientists can still translate and validate the results.

Are There Specific Capabilities Being Exaggerated?

It's not straightforward hype, and it's not fully delivering either. Even the newest generation of tools, like the latest AlphaFold, hallucinates more than its predecessor, and hallucination is inherently not useful, even though generating novel possibilities is part of the point of generative tools. Some of the gap comes down to how people use them. I've seen strong results from tools that weren't designed for the use case, simply because someone found a clever workaround. Benchmarks and real lab use don't always line up.

You've Compared the Pope’s Take on the AI Debate to the Stem Cell Controversy. Where Are You Seeing Those Parallels?

Both debates carry a strong moral charge that isn't really about the underlying science. With embryonic stem cells, the quandary centers on the source of the cells and where life begins. AI has no such component. It's a machine, not a form of life. What carries over is the pattern: a prominent religious voice shapes conversations well beyond people who share that faith. If colleagues in your circle are Catholic, AI ethics becomes a topic of conversation whether you are or not, and that attention compounds with other concerns already circulating, like the environmental footprint of data centers.

Where Are AI Agents Saving Time Today, And Where Is the Risk Not Worth It Yet?

Most agents in active use are doing simple work, scraping and compiling information, or handling short, sequential tasks. The majority deployed right now are coding agents, and that connects to manufacturing more than people expect, since every piece of equipment runs on software employees want to customize. The risk shows up with people who build bolt-ons without really understanding the code underneath, especially once an agent is given broad permissions and starts operating outside its intended scope.

Is There a Company Spending Enough Money to Make That Risk Worthwhile?

I'm not convinced yet. Announcing a partnership or a large infrastructure figure is easy, a $30 billion or $200 billion number tells me a company is buying capacity, but not that it's solving the training and talent problem underneath. I'll believe someone is serious when I see real money going into people doing the work, with the hours to match, not just infrastructure. Until that shows up in quarterly and investor reports, I'd call it mostly announcements rather than execution.

What Does a Pharma Company Doing This Right Look Like in Two Years?

We'll know more in 2 years, honestly. Companies are testing different strategies, and part of what separates them is whether the expertise they're hiring or training matches how they intend to use AI. My suspicion is that team-oriented implementations will outperform ones built around individual empowerment. When AI use is a shared, community resource rather than an individual one, tacit knowledge doesn't walk out the door when one person leaves, the rest of the team keeps it moving. Organizations that build AI adoption around teams, not just individuals, are the ones I expect real results from.

Articles

AI-enabled discovery is bifurcating into pathway-grounded, data-mining approaches that preserve scientific interpretability and high-throughput generative candidate churn with limited demonstrated success. Tool performance remains uneven; newer models can hallucinate, and clever user workflows often outperform benchmark expectations. AI-ethics discourse mirrors embryonic stem cell debates in moral salience and amplification by religious authorities, compounded by sustainability concerns. In manufacturing, agents mostly automate narrow tasks and code customization, but over-permissioned bolt-ons raise security risk. Capital outlays signal capacity, not talent; durable gains likely require team-centered adoption.

Richard Jaenisch, Open Biopharma breaks down AI's real limits in drug discovery, agent security risks, and what workforce readiness looks like by 2028.

Q&A: AI Coding Tools Carry the Real Risk

In the second half of a two-part deep dive on major pharma-AI partnerships with 

’s Chris Cole, Richard Jaenisch, senior director of education, outreach, and digital experience at Open Biopharma, argues that generative AI's promise in drug discovery is real but narrower than the hype suggests. Jaenisch draws a sharp distinction between AI tools that mine existing pharmaceutical datasets to surface overlooked structures — where scientists stay in control and can validate the science — and generative approaches he likens to "chimpanzees writing Shakespeare," where sheer volume substitutes for scientific rigor. The former is producing real pipeline candidates, including one drug now in Phase III; the latter, he says, hasn't yet demonstrated meaningful success. (

Watch Part One on why Roche's raw GPU infrastructure buildout and Sanofi's regional talent hub represent fundamentally different bets on where AI adoption actually stalls in pharmaceutical manufacturing.

In this episode, we discuss AI drug discovery hype, agent security risks, and why pharma's AI workforce bets remain unproven.

Manufacturing Intelligence: Sizing Up Pharma's Biggest AI Deals, Part Two 

2:17 – Where the line is today between what generative AI can do in discovery and formulation and what people imagine it does

9:38 – What’s feeding this disconnect, be it exaggerated capabilities or pure hype

11:16 – Similarities between the ethical conversations around AI and those around embryonic stem cell use

20:17 – Whether or not the Pope voicing opinions on AI has any sway, operationally, for pharma companies

22:09 – Where AI agents actually save time today in pharma where the security risk is just not worth it

31:23 – If there’s a pharma company or situation currently “hamfisting” enough money to make the security risk worth it

37:05 – With workforce repositioning as the metric that determines whether any of these investments pay off, what a pharma company doing this correctly looks like in two years

Jaenisch also draws an unexpected parallel between the current AI ethics debate and the embryonic stem cell controversy, noting that Pope Leo's recent commentary on AI has forced pharma leaders, Catholic or not, into conversations about morality and technology they weren't previously having. He connects this cultural moment to concerns about "AI psychosis" and over-reliance on chatbots as substitutes for human judgment and community.

On agentic AI specifically, Jaenisch is blunt about the security exposure created by ungoverned deployment. Most agents in active use today are simple scraping or reporting tools, he says, but coding agents that are increasingly used to build unauthorized software bolt-ons represent a growing risk, since the same capabilities that help non-experts build tools also help bad actors find exploits in them. He points to real-world incidents of agents breaking sandbox environments and warns that context-window limitations can cause agents to quietly drift from their original instructions over time.

Pressed on whether any pharma company has invested enough to meaningfully move the needle on AI-driven workforce transformation, Jaenisch is skeptical of headline infrastructure spending and argues the real signal will be sustained, team-based investment in people, not individual empowerment. "If everyone's stirring the bucket, the bucket keeps moving," he says, describing how team-oriented AI adoption preserves institutional knowledge even as individuals leave. He predicts the industry will have clearer answers on ROI and workforce strategy within two years, once companies begin reporting real financial results rather than partnership announcements.

Videos

The FDA on Aug. 5, 2026, approved Moderna's mFlusiva (mRNA-1010) for adults 50 and older, granting approval for two age groups (adults 50 to 65 and adults 65 and up) on the condition that Moderna run an additional clinical trial in the older group.

 The company said the vaccine will be available for this fall's flu season.

Moderna's mRNA shot was nearly 27% more effective against seasonal flu than a standard flu shot, according to clinical trial data from a study enrolling more than 40,000 adults aged 50 and older.

 Reactogenicity ran higher than with standard-dose comparators, including higher injection-site pain and systemic symptoms, though events were predominantly mild-to-moderate and resolved within one to two days.

 For adults aged 65 and older, immunogenicity data comparing mFlusiva against Fluzone High-Dose showed superior antibody responses at day 29 and at six months, the data underpinning that group's accelerated approval.

The review was not straightforward. The FDA initially refused this year to review Moderna's application before it reversed course a week later, with agency officials saying they wanted more data because Moderna had compared its vaccine with a standard flu shot in adults aged 65 and older whereas the CDC recommends a high-dose vaccine for that age group.

 An amended, age-stratified application followed, and the FDA's Vaccines and Related Biological Products Advisory Committee voted 9-0 that benefits outweighed risks ahead of the final decision.

That regulatory maneuvering echoes what our 

 series has tracked this year in tariff- and funding-driven shifts affecting mRNA programs, and it lands against a specific policy backdrop: HHS canceled roughly $500 million across 22 BARDA-funded mRNA vaccine projects in 2025, a decision that also forced a stop-work order on 

. The approval is likely to face pushback from allies of the health secretary, underscoring that mFlusiva's clearance came despite, not because of, the current federal posture toward the platform.

What Does the Approval Mean for Manufacturing?

mFlusiva extends Moderna's mRNA-lipid nanoparticle platform beyond COVID-19 into a second major respiratory indication, following the 2

That trajectory has been built on real infrastructure: Moderna's 

, all now underpinning a third product line rather than a single pandemic response.

The core manufacturing argument covered in our look at 

engineering challenges in commercial-scale mRNA production

—shared production infrastructure and rapid reformulation to match circulating strains—gets a real-world test as Moderna scales mFlusiva for fall 2026 distribution.

1. Lovelace, B. "FDA approves 1st mRNA flu shot, from Moderna." NBC News, Aug. 5, 2026. 

2. "FDA approves first mRNA-based flu vaccine." CNN, Aug. 5, 2026. 

3. "FDA Advisory Committee Supports Moderna mRNA Flu Vaccine Review." AJMC, 2026. 

4. "FDA Advisory Panel Votes 9-0 in Favor of Moderna's mRNA Flu Vaccine." BioPharm International, 2026. 

5. "A new type of flu shot is coming after FDA approves mRNA-based vaccine." Washington Post, Aug. 5, 2026. 

FDA approval of Moderna’s mFlusiva (mRNA-1010) introduces the first mRNA-based influenza vaccine for adults ≥50 years, with age-stratified authorization and a postmarketing trial requirement in those ≥65. In a >40,000-participant study, relative vaccine effectiveness was ~27% higher than a standard comparator, with increased but mostly mild-to-moderate, short-lived reactogenicity. For ≥65, accelerated approval relied on superior immunogenicity versus high-dose influenza vaccine at day 29 and six months. The decision arrives amid mixed federal signals toward mRNA programs.

FDA approves Moderna's mFlusiva, the first mRNA flu vaccine, testing the platform's manufacturing case amid federal funding cuts.

What Does Moderna's mRNA Flu Vaccine Approval Signal?

In late July 2026, the White House Office of Science and Technology Policy (OSTP) released 

, a 123-page report to the President authored by OSTP Director Michael Kratsios.

 Framed explicitly as a modern sequel to Vannevar Bush's 1945 report 

—the document that created the postwar federal science funding system and led to the founding of the National Science Foundation (NSF)—the new report diagnoses what it calls a slowdown in American scientific productivity and lays out a reform agenda covering federal research funding, regulatory policy, domestic manufacturing, and the role of AI in science.

Much of the report is broad science policy: restructuring NSF grant mechanisms, opening national laboratories to industry, reforming graduate education. But several sections speak directly to the CMC, manufacturing, regulatory, and supply-chain concerns that matter those working in pharmaceutical manufacturing and development. Here's what's in it, and what to watch for.

The Productivity Problem: Eroom's Law Goes Official

The report leads its diagnosis of American science with a well-recognized statistic: the steady decline in drug-development efficiency often called "Eroom's Law" (Moore's Law spelled backward). According to the report, pharmaceutical R&D efficiency, measured as new drugs approved per billion dollars spent, has fallen roughly eightyfold in inflation-adjusted terms since 1950, with efficiency roughly halving every nine years.

 That figure traces directly to the 2012 paper that coined the term, by Jack Scannell and colleagues in 

, which reported the identical eightyfold decline and nine-year halving period.

 The report also notes that despite the NIH budget more than doubling since the 1990s, breakthrough treatments, citation impact per dollar, and overall scientific productivity have not scaled proportionally.

This isn't a new observation to industry insiders, but its inclusion in a formal White House policy document elevates it from an industry talking point to an official justification for reform of grant structures, regulatory review, and how the government funds biomedical research generally.

Reproducibility and the Alzheimer's Cautionary Tale

One of the more pointed case studies in the report involves a 2009 Alzheimer's research paper published in a top journal that proposed a promising treatment target.

 The report recounts that other researchers demonstrated the finding was not reproducible by 2012, and that a pharmaceutical company's internal review had already ended a drug development program based on the work. Even so, the report notes, the paper continued to accumulate more than 800 citations and helped misdirect research priorities and federal funding for roughly another decade before being formally retracted 15 years after its original publication, and only after its lead author resigned amid a separate investigation into data manipulation.

While the report doesn't name names, the details identify the case precisely: it's the 2009 

 paper "APP binds DR6 to trigger axon pruning and neuron death via distinct caspases," co-authored by Marc Tessier-Lavigne while he was a scientist at Genentech. Genentech's own internal review had flagged reliability problems with the paper's central finding as early as 2008, before publication.

 The paper drew more than 800 citations by the time Tessier-Lavigne, president of Stanford University by then, resigned that post in July 2023 following a Stanford investigation into data manipulation across several of his papers; the Alzheimer's paper itself was formally retracted in December 2023, about 15 years after it appeared.

The report uses this episode to argue for what it calls "Gold Standard Science," a push, tied to an existing "Restoring Gold Standard Science" executive order,

 to mandate reproducibility, transparency, data sharing, and falsifiability across federally funded research.

For those involved in sponsored biomedical research or NIH-funded programs, this points toward tighter expectations around data transparency and replication going forward, and toward AI-assisted "verification infrastructure" intended to catch irreproducible results before they propagate.

Chapter III of the report calls for "restoring permissionless innovation," reweighting regulatory review to account for the cost of inaction, not just the risk of action, and expanding regulatory sandboxes that let companies test new technologies under controlled conditions. Notably, the report cites the administration's existing reforms in "nuclear, pharmaceuticals, and drones" as the model it wants to extend into other sectors.

That's a signal worth flagging: pharma is being held up in this document as a precedent-setting example of the administration's deregulatory approach, meaning further changes to FDA review processes, expanded use of real-world evidence, or new regulatory sandbox frameworks could follow, framed as extensions of work already underway rather than new policy territory.

New Funding Models that Could Reach Bio/pharma

A large portion of the report is devoted to reforming how federal science funding actually works, and several of the proposed mechanisms are directly relevant to bio/pharma researchers used to navigating NIH's standard R01 grant structure.

The report highlights "Focused Research Organizations" (FROs), time-bound, non-profit research entities designed to break specific scientific bottlenecks that are too large for an academic lab but not commercially attractive enough for a pharmaceutical company to pursue on its own. It gives the example of platform technologies for decoding basic biology, noting that pharmaceutical companies "face much stronger incentives to chase the next drug breakthrough than to build platform technologies for decoding basic biology," precisely the kind of infrastructure work FROs are meant to fill.

The report also points to NSF's newly launched "

" program, the agency's first initiative explicitly designed to fund independent research organizations outside traditional academic institutions, offering full-time research teams operational autonomy and milestone-based funding.

 Other funding mechanisms discussed include "golden ticket" systems (piloted in Denmark and now being tested by NSF) that let individual reviewers champion high-risk proposals outside of consensus review, prize challenges and advanced market commitments that pay for outcomes rather than effort, and "fast grant" models capable of turning around funding decisions in 48 hours rather than 6-9 months.

For those involved in academic-industry partnerships or grant-funded biomedical research, these mechanisms represent a widening menu of funding pathways beyond the traditional NIH grant cycle, worth tracking as they scale.

Manufacturing Reshoring Gets Elevated to a "National Character" Issue

Chapter IV argues that scientific discovery alone doesn't guarantee that the resulting manufacturing jobs, supply chains, and process knowledge stay in the United States. The report points to lithium-ion batteries and EUV lithography as cautionary examples, technologies the U.S. helped invent but where manufacturing capacity and supply chains are now dominated by companies headquartered in Asia and Europe.

 It argues explicitly that "discovery without domestic manufacturing leaves America paying the research bill while rivals develop the process improvements and capture the economic, strategic, and knowledge returns."

The report frames reshoring not simply as an economic policy goal but as a national security and "national character" issue, which suggests continued (and possibly expanded) federal attention to incentives for domestic pharmaceutical manufacturing, alongside workforce development proposals, including new apprenticeship pathways, practitioner-in-residence programs pairing skilled technicians with PhD researchers, and expanded community college partnerships aimed at rebuilding the "hands-on" manufacturing workforce alongside the research workforce.

The report's final chapter calls for launching and scaling what it calls the "Genesis Mission," a flagship national AI-for-science initiative intended to integrate supercomputers, AI models, scientific instruments, and datasets across national laboratories, with a stated goal of doubling the productivity of U.S. science within a decade.

 It also calls for accelerating investment in autonomous, closed-loop laboratories and robotics-driven experimentation, arguing these could compress discovery timelines "by orders of magnitude."

Expect new federal funding vehicles and infrastructure investments aimed at AI-driven discovery and automated experimentation, which biologics developers, CDMOs, and drug discovery teams may be able to access through partnerships with national laboratories as these programs scale.

Taken together, a few practical threads stand out:

Watch for FDA guidance or executive actions that build on the report's framing of pharmaceutical deregulation as an already-successful model to extend elsewhere.

New mechanisms like FROs, X-Labs, golden-ticket review, and fast grants could open funding pathways for platform technology development that don't fit neatly into a standard NIH grant.

Expect increased emphasis on reproducibility and data-sharing requirements in federally funded biomedical research, tied to the "Gold Standard Science" executive order.

Reshoring incentives and new workforce training models are likely to remain a policy priority, reinforcing trends already underway in API and biologics manufacturing.

The Genesis Mission and related AI-for-science investments may create new collaboration opportunities between national labs and industry for AI-driven drug discovery and automated experimentation.

One important caveat: this is a policy vision document from the OSTP director, not enacted legislation. Many of its central proposals—the Genesis Mission, new institutional funding models, expanded regulatory sandboxes—are recommendations rather than current law. How much of this translates into actual FDA guidance, NIH funding structures, or budget appropriations remains to be seen and will depend on subsequent executive and legislative action.

Kratsios, M. Science: A New Golden Age. Office of Science and Technology Policy, A Report to the President, July 2026, 

https://www.whitehouse.gov/wp-content/uploads/2026/07/Science-A-New-Golden-Age.pdf

Bush, V. Science, the Endless Frontier, 75

 Anniversary Edition. National Science Founcation. 

https://nsf-gov-resources.nsf.gov/2023-04/EndlessFrontier75th_w.pdf

Scannell J, Blanckley A, Boldon H, Warrington B. Diagnosing the Decline in Pharmaceutical R&D Efficiency. 

Wosen J. Stanford President to Resign After Investigation of Research. 

https://www.statnews.com/2023/07/19/marc-tessier-lavigne-stanford-president-resignation

Retraction Watch. Former Stanford president retracts Nature paper as another gets expression of concern. December 2023, 

https://retractionwatch.com/2023/12/18/former-stanford-president-retracts-nature-paper-as-another-gets-expression-of-concern

The White House. Restoring Gold Standard Science – Executive Order 14303. Available at 

https://www.whitehouse.gov/presidential-actions/2025/05/restoring-gold-standard-science/

OSTP’s Science: A New Golden Age frames declining U.S. scientific productivity—highlighting Eroom’s Law and stagnant output despite NIH growth—and proposes reforms spanning funding, regulation, manufacturing, and AI. A high-profile Alzheimer’s reproducibility failure is used to justify “Gold Standard Science” requirements for transparency, data sharing, and verification. Pharma is positioned as a deregulatory model, with potential expansion of sandboxes and real‑world evidence. New funding vehicles (FROs, X-Labs, fast grants, prizes) and domestic manufacturing workforce initiatives accompany a national AI-for-science “Genesis Mission.”

OSTP's July 2026 report to Trump diagnoses U.S. science slowdown and pushes deregulation, new funding models, reshoring, and AI-driven "Genesis Mission."

What the OSTP’s New Policy Blueprint Means for Pharma Manufacturing and Development

As the pharmaceutical industry crosses the midpoint of 2026, the landscape is defined by a complex interplay of rapid technological advancement, shifting geopolitical alliances, and evolving regulatory frameworks. Following a series of interviews conducted in July 2026 with seven pharmaceutical manufacturing and development experts, a clear picture emerges: while the industry has embraced the potential of next-generation modalities and AI, the "last mile" of execution—particularly regarding regulatory compliance, data consistency, and organizational readiness—remains a formidable hurdle (

Continued below the infographic - click to enlarge

Infographic: Pharma Manufacturing 2026 Midyear Reality Check

This report synthesizes these expert perspectives to provide a comprehensive look at the trends defining H1 2026 and the challenges awaiting the industry in the final months of the year.

We also wanted to hear from you. Findings from a poll of our LinkedIn followers reflect the core tensions identified by the seven experts in their midyear 2026 assessments, particularly regarding the gap between technological potential and operational reality (Table).

I. The Rise of Next-Generation Modalities: ADCs, AOCs, and the GLP-1 Surge

The first half of 2026 has seen a significant surge in the development of complex biologics and bioconjugates. 

, notes that his earlier predictions regarding the rise of these molecules have been largely borne out by clinical progress and regulatory approvals.

"We're seeing a lot of excitement, a lot more investment, in these next-generation molecules, and we're seeing that kind of borne out in the clinic and through some of the approval processes as well," Bunce observes.

Bunce specifically highlights the "surge in ADCs" (antibody-drug conjugates), noting that there are now approximately 23 approved ADCs worldwide. Recent milestones include the approval of AbbVie’s Decnupaz, an anti-CD123 conjugate for dendritic cell neoplasia, which utilizes a novel PPD dimer toxin payload. Beyond traditional ADCs, the industry is witnessing intense interest in antibody-oligoconjugates (AOCs), with companies like Avidity and Dyne progressing treatments for muscular dystrophies (Figure 1).

Bunce predicts this trend will only strengthen as biotechs face immense pressure to reach "value inflection points" by securing patient safety and efficacy data quickly. To meet this demand, CDMOs are shifting toward high-throughput "front-end design" processes and integrated manufacturing platforms that truncate timelines by combining protein production and conjugation activities.

Simultaneously, the "GLP-1 boom" continues to reshape manufacturing priorities and patient access models. 

Mike Stenberg, vice president of Business Development at LGM Pharma

, points to the rapid normalization of direct-to-consumer (DTC) and telehealth channels as a major development in how patients access these drugs.

"The biggest story for me is what's spilling over into how people access these drugs. Direct-to-consumer platforms have gone from a novelty to nearly universal in under two years,” Stenberg states. “Meanwhile, we continue to see new routes of dosage administration being explored."

While transdermal and subdermal delivery systems are still in early development, the approval of the first oral solid dose GLP-1s represents a milestone in administration routes. 

Saharsh Davuluri, CEO and managing director of Neuland Laboratories

, adds that while small molecules remain a core modality, the need to invest in new modalities like peptides and oligos is creating a challenge in capital allocation. He notes that Neuland is currently commissioning new peptide manufacturing facilities to meet increasing demand, particularly for large-scale production.

Continue reading this exclusive members-only content!

(Free, 30-second registration required for new users)

Pharma manufacturing and development experts reassess their 2026 predictions for the industry and look ahead to the second half of the year.

Midyear 2026 Industry Report: Taking Stock of Pharma Manufacturing, AI, and Regulatory Trends

II. Geopolitics and Global Supply Chain Reshaping

The "China+1" strategy has evolved into a more aggressive push for multi-region redundancy and onshoring. Experts agree that geopolitical tensions and new tariffs have transformed supply chain resilience from a strategic choice into an operational necessity.

serialization, regulatory compliance, and track and trace expert, observes that the framing he used at the end of 2025 was already too conservative by early 2026. "I'm seeing companies building genuine multi-region redundancy, not just adding one backup supplier and calling it done,” Garg explains. “The supply chain resilience conversation is no longer primarily about geography. It's increasingly about data visibility."

Garg argues that geographic diversification alone is insufficient; companies must combine it with real-time traceability and serialization data infrastructure to respond to disruptions quickly. Davuluri echoes this sentiment, highlighting India’s sustained role as a critical manufacturing hub, particularly for APIs and peptides. However, he notes a surprising lack of new CDMO investment commitments on US soil compared with the "flurry of announcements last year."

A significant challenge identified by Davuluri is the increasing volume of out-licensing deals coming from Chinese innovators to “big pharma.” This creates a complex manufacturing dilemma for non-Chinese CDMOs, he notes. "When you have big pharma in-licensing some of these new drugs from Chinese innovators, who determines the manufacturing plans? Is it the Chinese company that is licensing out, or is it the big pharma company?" Davuluri asks.

Despite this complexity, Stenberg sees a clear trend toward onshoring drug product manufacturing closer to the end market. He notes that while APIs and raw materials often remain globally sourced, finished dose is where domestic investment makes the most sense for gaining operational, regulatory, and economic control over the "last mile."

III. The AI Evolution: From Productivity to Operational Participation

Artificial Intelligence has moved past the "hype" phase into practical application, though its deployment remains uneven due to validation and auditability requirements. 

Megha Sinha, founder and CEO of Kolter AI

, reports that "agentic systems" are now actively participating in plant floor operations.

"Inside the plant, agentic systems are already detecting out-of-spec results, pulling batch records, drafting root cause analyses, and cutting investigation time by 50% or more in validated GxP environments,” Sinha notes (Figure 2). “That is a shift from AI as a productivity layer to AI as an operational participant."

However, she argues that these systems stop short of solving the industry's costliest problem: translating manufacturing changes into regulatory consequences across dozens of markets. Current digital twins model shipments and capacity but cannot compute the specific filing requirements or grace periods triggered by a process change.

Henrik Johanning, senior vice president of quality & strategy at Epista Life Science

, notes that AI governance has matured rapidly. "The conversation has matured from 'What can AI do?' to 'Can the regulated process defend what AI helped create?' That is a significant shift," he observes.

Regulatory expectations now require AI tools to sit within the pharmaceutical quality system, necessitating defined intended use, risk assessments, and human oversight. Stenberg warns that the pace of AI adoption will continue to look slow because the industry proceeds carefully given the regulatory scrutiny. He points to the 

first FDA warning letter citing improper use of AI in pharma manufacturing

IV. Data Quality: The "New Moat" and the NDC-12 Identity Transformation

The experts we spoke with all agree that technology is no longer the primary bottleneck in pharma manufacturing; rather, it is the quality and governance of the underlying data. Garg emphasizes that data quality has become the central competitive differentiator for AI in manufacturing, stating, "The companies that are realizing real AI value in early 2026 are almost uniformly the ones that invested in their [enterprise resource planning (ERP)], serialization, and master data infrastructure in 2023 and 2024. Data quality isn't just a barrier; it's the new moat."

This focus on data is critical as the industry faces 

. Garg describes the FDA's 12-digit National Drug Code rule as an "enterprise identity transformation" that will affect virtually every system in the supply chain, from ERP and serialization to labeling and reimbursement platforms. "The parallel I keep using is DSCSA... History is about to repeat itself for organizations that make the same mistake with NDC-12 [by treating it as check-the-box compliance]," Garg warns.

Remco Munnik, founder of Arcana Life Sciences

, notes that while management has a "strong appetite to 'solve everything with AI,'" the reality is that AI only works when foundational data governance, such as identification of medicinal products, or IDMP, is solid. He identifies "digitalization with consistency" as the primary challenge for the remainder of the year, ensuring that data are interoperable across regulatory, quality, and manufacturing systems.

V. The Regulatory Landscape: Annex 1, Reform, and Convergence

In Europe, the implementation of EU GMP Annex 1 continues to drive facility investments. Johanning reports a strong focus on HVAC, water systems, and barrier technologies as companies move from theoretical risk management to real capital decisions. "The real maturity test is whether [quality risk management] changes decisions, not whether the risk assessment exists," he remarks (Figure 3).

While global regulatory convergence has helped align terminology around data integrity and quality systems, Johanning notes that "operational friction" persists due to regionalization and local release models. Munnik adds that the 

 is making significant progress toward a future in which lifecycle changes are handled via structured database updates rather than document-based submissions, with tangible success targeted for 2027 (Figure 4).

On the inspection front, Stenberg analyzed recent FDA data and found that while inspections increased in 2025/2026, the growth was almost entirely domestic. "Inspections did increase, but growth was almost entirely domestic, with foreign inspections in countries like China and India seeing far more modest growth despite the number of registered facilities there continuing to increase," Stenberg notes. Critically, the core GMP deficiencies found during these inspections, such as issues with written procedures and investigations of failures, remain unchanged from previous years, according to Stenberg.

VI. The Human Factor: Skills Gaps and Change Management

The transition to digital-first manufacturing has exposed a significant skills gap and the need for rigorous change management. Johanning notes that while companies are upskilling, quality, engineering, and IT still too often operate in separate, deep lanes. He argues the constraint is the ability to make "good GMP risk decisions" across these disciplines.

Munnik observes that economic pressures often cause companies to postpone the rotational programs and skill-building needed to bridge this gap (Figure 5). "When budgets tighten, rotations and skill-building are often the first to be postponed, even though they are exactly what is needed to close the gap," he says.

Sinha adds that the "orchestrator" roles often touted by companies are frequently "title-deep rather than skill-deep." She frames the shift as a fundamental change in the job itself, moving from 

 them. Davuluri emphasizes that a major part of this transition is ensuring that scientific talent feels secure and views AI as a tool to aid their effectiveness rather than a replacement.

VII. The Road to H2 2026: Challenges and Strategic Outlook

Looking ahead to the remainder of 2026, several critical challenges remain (Figure 6):

 Stenberg warns that early-stage biotech financing is on pace for its "worst year this decade," which could lead to a thin manufacturing pipeline in the future.

 Garg predicts that as the stabilization period ends, organizations should expect greater operational scrutiny from trading partners regarding data accuracy and interoperability.

 Johanning argues the biggest challenge is "turning risk-based intent into executable control strategies" without falling into the trap of over-control.

 Davuluri emphasizes the need for CDMOs to be more agile and responsive, utilizing flexible manufacturing lines to meet crunched development timelines.

 Sinha predicts that the "connective tissue" between functions must be built now to survive the pace that AI-native discovery is about to create.

: Munnik identifies the core challenge as "digitalization with consistency," warning that digitalization only works when "underlying data are stable, governed, and consistent across regulatory, quality, and manufacturing systems."

: Bunce highlights the pressure to offer "innovative ways of rapidly developing molecules" through "high-throughput sort of design" to help biotechs reach critical patient data as quickly as possible.

The first half of 2026 has demonstrated that the pharmaceutical industry is no longer limited by its technology, but by its organizational readiness and data integrity. As Sinha aptly summarizes, "Strategy moves at the speed of boardroom. Execution has moved at the speed of a spreadsheet.” The choice, she says, is whether to wire operations into one auditable backbone now or remain fragmented.

The winners in the second half of 2026 will be those who treat digitalization as a data consistency challenge rather than a technology project, ensuring that their manufacturing, regulatory, and quality systems are truly aligned. As the industry moves toward 2030, the shift from "point solutions" to integrated, data-driven operational backbones has become a requirement for survival in a rapidly evolving global market.

Geopolitical tension and tariffs are accelerating multi-region redundancy and onshoring, making real-time traceability and serialization central to resilience across global networks. Agentic AI is already reducing investigation cycle times in validated GxP operations, yet adoption is constrained by auditability and quality-system governance. Data governance now differentiates winners, especially as NDC-12 forces enterprise-wide identity remapping akin to DSCSA. Annex 1 capital upgrades, uneven FDA inspection growth, and persistent GMP findings add pressure. Skills gaps and cross-functional integration will define H2 2026 execution.

Six months in, pharma manufacturing and development experts reassess their 2026 outlooks and predict what comes next.