Making & Distributing Pharma 2026
Two days, four keynotes, and one consistent message: the MHRA is moving at pace and it expects the industry to keep up.
An Event That Delivers on Its Name
Making & Distributing Pharma is one of those events that consistently delivers more than its programme suggests. The formal sessions are strong, but the real value accumulates in the corridor conversations and Q&A exchanges that run over time because the room is full of people who actually know what they are talking about. If you work in pharmaceutical quality, regulatory affairs, or manufacturing and you are not yet attending, you should be.
This year’s event was particularly strong. The themes were well-chosen, the speakers were credible, and the regulator’s presence gave the discussions a directness that does not always come through in published guidance alone. We will be back in 2027 when it moves to the NEC in Birmingham.
AI Was the Thread Running Through Everything
AI featured heavily throughout day one, and what stood out was not the breadth of the coverage but the maturity of the conversation. The industry has moved past debating whether AI has a role in pharmaceutical development and regulation. The questions being asked now are more demanding: who is accountable when an AI-assisted output is wrong, how do you validate a system whose behaviour is not fully deterministic, and what does a governance framework look like that satisfies both operational reality and regulatory expectation.
Sessions covered AI applications across drug development, manufacturing process understanding, regulatory affairs, pharmacovigilance, and market surveillance. These are no longer hypothetical use cases. They are operational realities in a significant number of organisations, and the governance infrastructure surrounding them is, in many cases, lagging well behind.
The pattern across sessions
The recurring theme was not AI capability. That case is broadly settled. What organisations are grappling with is the governance gap: the technology is moving faster than the policy infrastructure, and regulators and auditors are increasingly treating that gap as an active compliance risk.
The MHRA Is Deploying AI, and Leading from the Front
One of the more striking disclosures of day one was the extent to which the MHRA is itself an active user of AI. The agency is not positioned as an external regulator watching from a distance. It is building and deploying AI capabilities within its own operations, and that changes the nature of the regulatory conversation considerably.
The MHRA’s market monitoring, data analysis, and risk management functions are increasingly AI-assisted. A regulator with direct operational experience of these tools understands their potential and their failure modes in a way that shapes inspection expectations. When an MHRA inspector asks about your AI governance framework, they are asking from a position of genuine technical familiarity.
The MHRA’s active leadership in the international regulatory AI space also came through clearly. The frameworks and expectations being established now will have reach well beyond the UK, which matters for any organisation operating across multiple regulatory jurisdictions.
The Standout Message: Don’t Wait for Annex 22
The standout session of day one was David Brown, MHRA Expert GDP, and the message was unambiguous. Effective AI governance is a present regulatory requirement. It is not something organisations can schedule for a future sprint or defer until Annex 22 is finalised. Inspectors and auditors are asking questions about it now.
The specific argument David challenged is one we hear regularly: that AI is just a tool, no different in principle from a spreadsheet, and therefore falls outside the scope of GxP governance frameworks. The MHRA’s position is clear. That argument is not defensible and will not survive regulatory scrutiny.
“The proposition that AI is simply a tool, and therefore sits outside the scope of commercial governance, is not a defensible position. The requirement to govern AI usage exists now. Annex 22 is not the starting gun.”
The outputs of large language models are influencing decisions, informing documents, and in some organisations shaping the evidence base underpinning regulatory submissions. The fact that a tool is commercially available does not remove the organisation’s responsibility to govern how it is used. Accountability sits with the individual and the organisation. Not with the model. Not with the vendor.
Key Takeaway — Peter Brown, MHRA Expert GDP
Effective AI governance is a universal necessity for organisations operating in regulated environments. The framing of AI as a tool that sits outside commercial governance is not tenable and will not satisfy regulatory scrutiny. Organisations should not wait for Annex 22 to establish their governance frameworks. The expectation exists now.
TDP has developed a practical AI governance framework comprising a policy, HAZOP, usage decision flow, prompting guidance, and document generation record, that organisations can take and adapt to support their own compliance position. Get in touch to find out more.
Decentralised Medicines, MHRA Transparency, and a Clear Message to Applicants
Day two delivered some of the most direct regulatory messaging we have heard in an open forum for some time.
A Different Kind of Day
Day two has a different quality to the first. The format is more intimate, the conversations more direct, and the regulator’s presence more prominent. This year it delivered on that potential. Both keynotes carried messaging that was specific, frank, and genuinely useful for organisations trying to understand where MHRA expectations are sharpening.
The event is free to attend and genuinely worth a place on your calendar. It is one of the few forums where you can hear directly from MHRA inspectors and assessors in an open setting and ask them direct questions. That access is not something to take for granted.
Martine Powell, Decentralised Medicines and Point of Care Testing
A detailed and informative session on one of the most complex and rapidly evolving areas of pharmaceutical regulation, where the product, the patient, and the site of manufacturing or testing are no longer in the same place at the same time.
Martine Powell’s session on decentralised medicines was technically substantive and delivered a necessary reality check. The framework exists and the regulatory pathway is established, but this is a new legal mechanism that has not yet been deployed at scale. Teams whose development planning is ahead of where the mechanism has been tested in practice need to factor that in.
Two points from the session deserve particular attention. First, the DSCMF must be treated as a live, controlled document managed within the organisation, not prepared for submission and then filed. Second, the regulatory expectations that apply to a conventional manufacturing or testing site apply equally to a decentralised one. The site is still a site, regardless of its geography or scale.
“The application requires a product to be considered in full. And remember your blinding controls.”
The specific callout on blinding controls reflects a complexity that is easy to underestimate in a decentralised model. When the blinding function is separated from the site of administration, variables are introduced that do not exist in a conventional setting. Teams designing decentralised protocols should treat blinding control as a question that needs answering from first principles, not an assumption carried over from standard practice.
Key Takeaways — Decentralised Medicines
The decentralised medicines framework is a new legal mechanism not yet deployed in practice. Manage the DSCMF as a live internal document, not a submission artefact. Decentralised sites carry the same regulatory expectations as conventional sites. When you get to the application: your product must be considered in full, and your blinding controls must be explicitly addressed.
Lewis Corbett, Team Updates and a Direct Message to Applicants
A candid and practically useful session delivering some of the most direct guidance on licensing applications that we have heard from the agency in an open forum.
Lewis Corbett’s session was the most directly practical of the two days. The first message was one the MHRA has communicated before but which clearly still needs repeating: engage early. Pre-submission dialogue is available, it is encouraged, and organisations that use it produce better-prepared applications. It remains underused.
The second point was straightforward: ensure your key licences are current. Lapses create friction that is entirely avoidable and, in some cases, create barriers that fundamentally complicate an application that would otherwise be straightforward.
“If you are applying for a licence, please postpone if you are not ready. We see too many applications that just are not ready.”
This is worth sitting with. An application submitted before it is ready does not simply fail and reset. It consumes agency resource, delays the programme, and creates a record that the organisation carries into every subsequent interaction with that team. The reputational cost of a premature submission is significantly higher than the cost of waiting until the application is genuinely ready.
The underlying message
The MHRA wants to grant licences. The agency is moving quickly and is well-resourced to assess applications. The bottleneck is not on the agency’s side. It is in the quality of what is being submitted. Organisations that engage early, prepare thoroughly, and submit only when they are ready will get the best outcomes.
Transparency from the MHRA, A Landmark Commitment for 2026
The most significant announcement from day two was that the MHRA will soon be publishing full observation statistics and observation detail.
The MHRA is working quickly and will soon be publishing full observation statistics and observation details for transparency.
This represents a genuine shift toward regulatory transparency in the UK, moving toward a model where the industry can see not just that inspections happened, but what was found, in aggregate and in detail. For organisations working to understand where regulatory expectations are sharpening, this is valuable intelligence.
This matters significantly. Until now, organisations have had to infer where MHRA inspection focus is sharpening: from guidance signals, from peer experience, from the occasional warning letter. When observation data is published, inference is replaced with evidence. Compliance programmes can be calibrated against real inspection findings rather than educated guesswork.
The publication of MHRA observation statistics and detail is a genuine shift toward regulatory transparency. It is also a prompt for action. Organisations that have not fully aligned their quality systems with current MHRA inspection expectations should use the period before publication to close those gaps. When the data is visible to the whole industry, so are the patterns. TDP helps organisations benchmark their compliance programmes against current inspection expectations and address gaps before they become findings.
Closing Thoughts
Two days, two distinct formats, and one consistent signal: the MHRA is engaged, is moving at pace, and has clear expectations of the organisations it regulates. Its willingness to attend events like this, to speak directly, to answer questions openly, and to give the industry honest and specific feedback is not something every regulator offers. It should be used well.
Making & Distributing Pharma continues to grow because the industry needs exactly what it provides: an expert-led, open forum where the regulator and the regulated can have a direct conversation. The move to the NEC in Birmingham for 2027 reflects that growth. TDP will be there. If the themes covered here are live issues in your organisation, we would be glad to talk.