WASHINGTON—From the rolling hills of California to the bustling cities of India, the households of Australia, and frozen penal colonies in the Russian Arctic, the ancient specter of toxins—poisons derived directly from plant or animal life—has risen to stalk the modern world.
Five European countries recently stated that Russia assassinated opposition leader Alexei Navalny using a lethal toxin called epibatidine, which is found in South American dart frogs. The idea of murdering a political rival with an exotic frog toxin feels like a plot lifted from a spy thriller. In the real world, it echoes the infamous 1978 assassination of Georgi Markov, a Bulgarian dissident writer who was fatally injected with a ricin-filled pellet fired from a modified umbrella. But the exotic toxin and delivery device in these specific cases belie the much more common nature of the threat that toxins pose.
Moreover, weaponized toxins are not just the domain of nation-states with offensive biological weapons programs. They are increasingly the weapon of choice for terrorists, too. In late 2025, the leader of an international extremist group pleaded guilty in New York to soliciting mass casualty hate crimes, including a horrific plot to distribute ricin-laced candy to Jewish children and racial minorities.
In India this past November, authorities disrupted a plot by suspects tied to the Islamic State–Khorasan Province to weaponize and use ricin to attack crowds and water supplies. A highly potent toxin derived from castor beans, just a few milligrams of ricin can cause significant sickness and death.
Rapid advances in artificial intelligence can now also make this threat exponentially worse.
The threat spectrum also continues to expand as nation-states explore novel agents. A recent US State Department assessment, for example, highlighted Chinese military involvement in research about dual-use marine toxins, raising serious concerns about Beijing’s compliance with the Biological and Toxin Weapons Convention.
This threat also extends to localized murders and tragic accidents. In Australia, the 2025 Erin Patterson case resulted in a life sentence for killing three people with death cap mushrooms during a family luncheon. Difficulties with detection and attribution of toxins made this case, and others like it, particularly difficult to prosecute.
And in California, families recently paid a devastating price for mistaking a death cap mushroom for a safer, edible one. Death cap mushroom poisoning often begins with vomiting and diarrhea, but after a brief period during which patients improve, liver failure occurs. This month, state health officials reported the largest surge of severe poisonings in years, with more than three dozen hospitalizations, three liver transplants, and four deaths as of February.
While the world banned biological and toxin weapons decades ago, that agreed-upon norm never eliminated the materials, knowhow, or intent that make toxins attractive to criminals, terrorists, and rogue states alike. Rapid advances in artificial intelligence can now also make this threat exponentially worse. These technologies lower the barrier to entry, making the complex knowledge required to design, synthesize, and deploy lethal toxins more accessible to bad actors than ever before.
There are those who want to believe that the problem of toxins no longer exists or that it is far less challenging than, for example, the threat of infectious diseases. But ongoing ricin events in the United States and overseas, repeated food contamination with aflatoxin, and growing concern about nation-states weaponizing toxins alongside viruses and bacteria—all demonstrate that the threat of toxins persists and appears to be increasing.
In response, the world needs to treat toxins as a main problem, not a subordinate issue.
Specifically, governments need to invest in and support biological attribution. These investigations are far from easy. They begin with mysterious illnesses, suspected food poisoning, unexplained deaths, and rumored causes. Researchers need forensic methods and investigations that can differentiate among natural, accidental, and intentional toxic exposure. Governments need credible results to move forward with prosecutions, inform intelligence assessments, and allow elected leaders to speak with confidence to the public about what is going on and what they are doing about it.
The United States should make diagnostics for toxins a higher priority, with the ability to distribute them quickly and effectively down to the state and local level. The US government needs to alert the healthcare and public health communities about the current threat that toxins pose so that they can learn to recognize (or at least suspect) this sort of poisoning and use readily available, rapid point-of-care diagnostics during the very short window of time available to treat and save victims. It is not possible to send specimens to reference labs and get results back in time to save lives. Responsible toxidromic biosurveillance, too, can play an important part in identifying public health events and allowing health professionals to better inform citizens. After all, the United States cannot address the threat if it does not pay attention to it.
