In February 2026, the World Health Organization's Executive Board sat down to adopt an updated Global Action Plan on Antimicrobial Resistance — a document ten years in the making. They couldn't do it. The language that stalled the vote was six words long: voluntary and mutually agreed terms. The fight over those six words reveals everything about why the world is losing the war on superbugs.
The Standoff
The 158th WHO Executive Board convened in early February 2026 to approve the updated Global Action Plan on AMR (GAP-AMR) for 2026–2035. The original GAP was adopted in 2015 — the first globally coordinated framework for tackling antimicrobial resistance. A decade later, the update was supposed to be routine.
It was anything but.
The draft text proposed that technology transfers related to antimicrobial production should occur on "voluntary and mutually agreed terms." This language sounds innocuous. It is not. It mirrors the exact phrasing contested in the ongoing Pathogen Access and Benefit Sharing (PABS) negotiations under the Pandemic Agreement — where it has been used to ensure that pharmaceutical companies retain full decision-making power over who gets access to their technologies, on what terms, and when.
"This is a normative anchoring tactic — inserting language into the GAP-AMR to establish a precedent for PABS negotiations. If 'voluntary and mutually agreed' becomes the standard in AMR policy, it will be cited as established consensus in every subsequent treaty negotiation."
Brazil led the opposition, joined by Indonesia, Colombia, and South Africa. Their argument: this language traps developing nations in permanent dependency on patent holders for access to antimicrobials and manufacturing technology. It restricts their policy space to build local production capacity — the very capacity they need most urgently, because fewer than 1 in 15 bacterial infections are properly treated in low- and middle-income countries.
On the other side: the United Kingdom, Japan, Spain (representing the EU), and Switzerland urged adoption without delay. They pointed to years of extensive consultations. IFPMA welcomed the draft, framing industry as a collaborative partner and emphasizing "effective R&D incentives."
The compromise came from Ethiopia and Nepal, who jointly proposed reopening only the contested technology transfer language — not the entire document. The Executive Board adopted this approach. The GAP-AMR update now goes to informal negotiations before WHA79 in May 2026, where it will face the same fault lines again.
The Ten-Year Report Card
Step back from the Geneva politics and look at what ten years of the original GAP actually achieved. Chen et al. published the first comprehensive governance evaluation in Nature Medicine — a 193-country One Health governance index built through Delphi consultation and analysis of 269 policy documents from 2017 to 2022.
That 44.5 out of 100 is the headline — and it flatters the situation. The study found that policy design consistently outpaces implementation and monitoring. Countries write plans but struggle to execute them. The animal and environmental sectors lag furthest behind, despite the One Health rhetoric. And the five-year delay between NAP adoption and measurable impact on AMR prevalence means that the plans adopted in 2020 are only now beginning to show results — while the crisis has accelerated around them.
Over 170 countries have developed national action plans. The question is no longer whether countries have plans. It is whether plans mean anything without funding, enforcement, and infrastructure.
What the Zero Draft Left Out
GARDP's critique of the zero draft identified four gaps that aren't technical oversights — they are structural blind spots that reveal whose interests shaped the document.
The Commitments on Paper
The 2024 UNGA Political Declaration on AMR was, on paper, the most ambitious set of global AMR commitments ever adopted. It acknowledged the scale. It set targets. It even mandated new institutional structures.
| Target | Benchmark | Deadline |
|---|---|---|
| Reduce AMR-associated deaths | 10% from 2019 baseline (4.95M) | 2030 |
| Human antibiotics from WHO Access group | 70% (currently 52.7%) | 2030 |
| Countries with funded NAPs | 60% | 2030 |
| Basic WASH in healthcare facilities | 100% | 2030 |
| Countries with resistance testing capacity | 80% (all bacterial/fungal pathogens) | 2030 |
| Eliminate non-veterinary use in animals | Complete phase-out | 2030 |
| Independent Science Panel on AMR | IPCC-style body mandated | TBD |
| R&D incentives | Delinked from price/volume | Ongoing |
A follow-up high-level meeting is scheduled for 2029. That means the world gave itself five years to hit targets that a decade of planning has barely moved. The track record is not encouraging.
The Implementation Desert
India offers the clearest case study of the gap between ambition and execution. In November 2025, Health Minister J.P. Nadda launched NAP-AMR 2.0 (2025–2029) — a genuine upgrade over the first plan. It coordinates 20+ ministries under a One Health framework. It proposes blue-stripe packaging to distinguish prescription-only antibiotics. It mandates Schedule H1 prescription enforcement.
But the plan's own architects acknowledge the obstacles. ReAct's analysis of India's AMR response identifies the same pattern that Chen et al. documented globally:
- 20+ ministries coordinated
- Blue-stripe antibiotic packaging
- Schedule H1 prescription enforcement
- State-level action plans
- Surveillance expansion
- No separate budget allocation
- Weak state-level coordination
- Surveillance biased toward tertiary care
- Over-the-counter antibiotic sales persist
- COVID disrupted first NAP implementation
India is not an outlier. It is the median. Most countries with national action plans face the same structural problem: the plan exists, but the money, the enforcement, and the institutional capacity do not. When Chen et al. measured implementation scores separately from policy design, the gap was consistent across income levels and regions.
The Access Paradox
Here is the irony that the EB158 standoff crystallized. The countries fighting hardest to keep "voluntary and mutually agreed" technology transfer language — the UK, Japan, Switzerland — are the same countries that have successfully built pull incentives and subscription models to ensure their own populations have access to new antibiotics. The UK subscription model launched April 1. Japan's delinked contracts show a 6:1 return on investment.
But those models only work for the countries that built them. The pipeline is already thin — 90 agents, only 15 innovative, only 5 targeting critical priority pathogens. When a new antibiotic does reach approval, it arrives in high-income countries years before it reaches the places where the burden is highest. Zero pediatric antibiotic formulations are available in 17 sub-Saharan African countries. The WHO's Access group antibiotics account for just 52.7% of human use globally — well below the 70% target.
The governance paradox is sharp: the updated GAP-AMR is supposed to coordinate a global response, but its technology transfer provisions would entrench the current geography of access — where new drugs flow to wealthy markets and the rest wait.
The Clock
Without robust action, the trajectory is known. The Lancet's GRAM analysis projects 39 million deaths directly attributable to AMR between 2025 and 2050. Annual treatment costs are projected to reach $412 billion, with productivity losses of $443 billion per year. The GRAM study also found that AMR-attributable deaths in people over 70 nearly doubled between 1990 and 2021 — a demographic tsunami that will only intensify as populations age globally.
Meanwhile, the 2026 AMR Benchmark shows the pharmaceutical response contracting: pipeline down 35% in five years (92 to 60 projects from large pharma). The corporate graveyard continues to grow — Venatorx, BiomX, Achaogen, Melinta, Tetraphase, PHAXIAM, and now Synlogic.
The plan to fight antimicrobial resistance is stalled by the same inequities the plan was meant to fix.
What Happens Next
The immediate timeline:
The Pattern
I've now covered 25 aspects of the AMR crisis — from broken economics to policy experiments to the surveillance blind spot to the chemistry that might save us. A pattern has emerged that the EB158 standoff makes explicit.
The science is working. New antibiotic classes are being discovered. Phage therapy is producing clinical evidence. CRISPR-based therapeutics are reaching Phase 2/3 trials. AI is accelerating drug design.
But the governance — the systems that translate science into globally accessible medicine — is not keeping pace. Markets fail to reward antibiotic development. Regulators are fragmented. Surveillance is incomplete. Access is geographically determined by wealth. And when the world's health ministers sit down to update the one document that's supposed to coordinate all of this, they deadlock over whether developing countries should have the right to manufacture the drugs their populations are dying without.
That is the governance gap. It is not a knowledge gap. It is not a science gap. It is a will gap — and it will cost 39 million lives if it doesn't close.