The simulation above is the whole problem of antibiotic resistance in miniature. Nothing in it is unusual or malicious. No bacterium is trying to become resistant. All that happens is that a drug kills the organisms it can kill, and whatever survives goes on to reproduce. Run it three or four times and the colony that comes back is one the drug can no longer touch.
The drug does not create resistance
This is the point people most often get backwards. The antibiotic does not cause the mutation. The resistant bacteria were already there, in tiny numbers, before the first dose. What the drug does is remove their competition.3 Every susceptible bacterium it kills is one less organism competing for space and nutrients, which leaves the survivors free to multiply into the gap.
That is why the resistant fraction climbs so steeply in the simulation while the total population barely changes. You are not creating anything new. You are steadily editing out everything else.
Every dose is a filter, and the things that pass through it are exactly the things you did not want.
Why this makes antibiotics unlike other drugs
A blood pressure drug works about as well on its ten millionth prescription as its first. Antibiotics are the exception. Every course given anywhere in the world, to anyone, applies a small amount of this pressure to the bacterial population at large. The effectiveness of the drug is a shared resource that each use draws down slightly.12
This is why prescribing an antibiotic for a viral infection is not merely useless. It is a course of selection applied for no benefit at all. It is also why finishing a prescribed course matters in cases where stopping early would leave a partially selected population behind, and why hospitals hold certain drugs in reserve, using them only when nothing else will work.
What actually helps
New antibiotics buy time, and several genuinely new classes have reached approval recently Approved. But the simulation shows why new drugs alone cannot solve this: the same process will run again on whatever replaces the current drug. What changes the slope of the curve is using antibiotics less often and more precisely, which means rapid diagnostics that tell you whether an infection is bacterial at all, better infection control so fewer courses are needed, and surveillance that catches resistant strains before they spread.
- Antibiotics do not create resistance. Resistant bacteria already exist in small numbers, and the drug removes their competition.
- Each course acts as a filter, so the surviving population is steadily enriched for resistance.
- This makes antibiotic effectiveness a shared resource that every use depletes slightly.
- New drugs delay the problem, but reducing unnecessary use is what changes its trajectory.
- Murray CJL, et al. Global burden of bacterial antimicrobial resistance in 2019: a systematic analysis. The Lancet. 2022.
- World Health Organization. WHO Bacterial Priority Pathogens List. 2024.
- Baym M, et al. Spatiotemporal microbial evolution on antibiotic landscapes. Science. 2016.
Read the full piece on infectious disease and the antibiotic pipeline.