A crisis that is already here

The World Health Organization projects that drug-resistant bacteria could kill 10 million people per year by 2050 if nothing changes. That is more than cancer, more than traffic accidents, more than most diseases that dominate public attention.

Josh Baylin’s Stansberry Research presentation, which we reviewed in full here, spends a surprising amount of time on antibiotic resistance. It is not the sexiest investment topic, but it may be one of the most consequential, and it illustrates exactly why medical AI matters.

The problem with the old way

Antibiotics are one of the most important discoveries in human history. Before penicillin, a simple infection could be a death sentence. But bacteria evolve, and over decades of antibiotic use, many strains have developed resistance to the drugs we have.

The solution is straightforward: develop new antibiotics faster than the bacteria can evolve resistance. The problem is the timeline. As Josh notes in the presentation, it takes 10 to 15 years and more than $2 billion to develop one new drug. Over the past 45 years, we have produced only a small number of new antibiotics, most of which were variations of drugs we already had.

The bacteria are evolving faster than the drug development pipeline can keep up. If that continues, the WHO’s projection of 10 million annual deaths by 2050 becomes less a worst-case scenario and more a mathematical certainty.

The MIT breakthrough

The Stansberry presentation describes what happened when an MIT lab turned AI loose on the antibiotic resistance problem. Instead of spending years in a lab, the AI screened molecular candidates and, within hours, discovered an entirely new class of antibiotic that could destroy some of the most drug-resistant bacteria on the planet.

This is not a hypothetical projection. It is a published result. The AI found something that human scientists had not found in decades of searching, and it found it in hours.

The scale difference matters. As Josh notes, “we can now come up with hundreds of thousands of these candidates in just three hours.” When the bottleneck shifts from discovery to testing, the timeline shrinks from decades to months.

The investment angle

Antibiotic resistance is not traditionally an investment thesis. It is a public health crisis. But the Stansberry presentation reframes it as an investment opportunity, and the logic has some weight behind it.

The key insight is that AI can potentially solve a problem that traditional drug development cannot. If AI discovers new antibiotic candidates at 10,000 times the traditional pace, the supply of new drugs can potentially keep pace with the evolution of the bacteria. The companies that make this possible, the AI drug discovery platforms and the supply chain companies we profile in our guide to medical AI stocks, stand to benefit from the acceleration.

The broader thesis is that medical AI is not just creating new drugs. It is solving problems that traditional drug development could not solve. And the antibiotic resistance crisis is the clearest example of a problem that traditional methods have failed to address.

Beyond antibiotics

The antibiotic resistance story is one example of a broader pattern. Josh presents several others:

A lab in Scotland used AI to screen 4,000 molecules for candidates that could reverse aging in human cells. In five minutes, the AI identified 21 promising candidates. Scientists then narrowed those to the three best, a process that would have taken years using traditional methods.

Scientists in Seattle used AI to create a new cobra antivenom. A drug called rentosertib, which treats idiopathic pulmonary fibrosis, became the first drug discovered entirely with medical AI, now in human trials after 50 years without a treatment.

There are 173 AI-discovered medicines in human trials, according to the presentation. These are treatments for diseases that traditional drug development failed to address, sometimes over decades of effort.

The $367 trillion context

The Stansberry presentation claims medical AI could be worth $367 trillion, or $1 million per American. As we noted in our review, that number is more than three times global GDP and is not a credible market sizing. But the underlying point, that AI applied to medicine creates enormous value, has real merit.

The antibiotic resistance crisis alone illustrates the scale. If AI can prevent 10 million annual deaths, the economic value of that prevention is genuinely enormous, even if it is not $367 trillion. The healthcare industry consumes about 20% of America’s entire economy, as the presentation notes. Even a modest improvement in the efficiency of drug discovery represents a massive market.

What to watch

For investors, the antibiotic resistance story is a microcosm of the broader medical AI thesis. AI is solving a problem that traditional methods could not solve, and the companies that make the solution possible stand to benefit.

The specific investment plays are in the AI drug discovery platforms and the supply chain companies that support them. The drug makers themselves are the obvious targets, but as we explore in our supply chain analysis, the companies that make the drugs possible may offer better risk-adjusted returns.

The antibiotic resistance crisis is not going away. If AI can genuinely outpace the evolution of drug-resistant bacteria, the companies that make that possible are solving one of the most important public health problems of the century. That is an investment thesis with real substance behind it.

NewsletterVetter is an independent publication. We receive compensation from some of the services we review through affiliate links. Nothing on this site is investment advice. Always do your own research.