The UK government says it wants to phase out animal testing. There is just one rather important detail missing. There is no overall timetable for doing it.
In a parliamentary answer published on the 8th of September, the government confirmed that its Replacing Animals in Science strategy contains deadlines for replacing particular tests, but no overall deadline for ending animal testing itself.
In 2025, 2.54 million scientific procedures involving living animals were carried out in Great Britain. That was down just 4% on the previous year. And some numbers went the wrong way. Procedures involving monkeys increased by 15%. Procedures involving dogs increased by 9%. Those involving cats increased by 11%. Horses were used in 12,101 procedures, up from 11,483. Overall procedures involving dogs, cats, horses and primates increased from 16,147 to 17,317. This is what “progress” currently looks like.
The government’s strategy, published in November 2025, talks about eventually eliminating animal use “in all but exceptional circumstances” and replacing animals wherever possible. It also acknowledges that some alternatives can offer greater species relevance, sensitivity and speed. There are deadlines for some individual practices. Regulatory animal testing for skin and eye irritation is supposed to be replaced where validated alternatives exist. Mouse tests for botulinum toxin potency are targeted for replacement. Pharmacokinetic studies involving dogs and monkeys are supposed to be reduced. But reducing individual tests is not the same thing as having a plan to end animal experimentation. Several important 2026 commitments have little or no publicly available evidence of progress. Training regulatory assessors. Increasing scrutiny of justifications for animal experiments in funding applications. Training early-career researchers in non-animal methods. These are not administrative footnotes. They are some of the exact barriers keeping animal experimentation embedded in science.
A 2025 review found that researchers face institutional resistance when trying to move away from animals. In one survey, 31% of researchers said they had used animals pre-emptively because they expected journal reviewers to demand it, while 44% had actually been asked by reviewers to add animal data.
The science itself is already moving faster than policy.
A large analysis of research published between 2003 and 2022 found that non-animal-only research outnumbered animal-based research across seven major areas, including cancer, heart disease, diabetes and toxicology. By 2022, there were around 2.7 non-animal publications for every animal-based publication. The UK had the highest overall ratio among the five countries studied, rising to almost 4.5 non-animal publications for every animal-based publication by 2022.
We are not waiting for somebody to invent alternatives. They are already here. Human tissues. Organoids. Organs-on-a-chip. Stem cells. Computer models. AI. Researchers have developed human heart models that can outperform animal tests. Generative AI systems are being developed that researchers say could cut animal use in some preclinical testing by 30% to 50%. A “virtual mouse” has already been created to model nanomedicine experiments using existing datasets.
The government itself has now committed £22 million towards human-specific technologies, including organoids, AI modelling, lab-grown heart tissue and projects specifically aimed at replacing dog and monkey cardiovascular safety studies. And the reason to move faster is not only ethical.
Animal experiments are not some flawless scientific gold standard.
A major review found that more than 92% of novel therapies fail during clinical trials, with differences between humans and other animals contributing to failures in translating safety and efficacy.
Then there is what happens to the animals themselves.
A 2026 UK study examined cervical dislocation, one of the most commonly reported methods used to kill laboratory rodents. There is no standardised technique and no commercially available purpose-built tool scientifically validated for consistently performing it. Researchers reported using pens, scissors, metal rods, cage scrapers and even spanners to break rodents’ necks. The authors warned that inaccurate dislocation can mean delayed unconsciousness, prolonged death and unnecessary suffering.
This is why percentages and strategy documents cannot become a substitute for the actual objective. The government says it wants to phase out animal testing. Good. Then give the phase-out a deadline. Fund the alternatives properly. Train researchers. Remove institutional barriers. Stop treating animal experiments as the automatic default. Replace tests as soon as scientifically possible. Because a phase-out without an endpoint risks becoming something else entirely.
An indefinite continuation.
*sigh*
I'll wait in the corner with bated breath tapping my foot. 😒