Since ChatGPT’s release in late 2022, many news outlets have reported on the ethical threats posed by artificial intelligence. Tech pundits have issued warnings of killer robots bent on human extinction^[1], while the World Economic Forum predicted that machines will take away jobs^[2].

The tech sector is slashing its workforce^[3] even as it invests in AI-enhanced productivity tools^[4]. Writers and actors in Hollywood are on strike^[5] to protect their jobs and their likenesses^[6]. And scholars continue to show how these systems heighten existing biases^[7] or create meaningless jobs – amid myriad other problems.

There is a better way to bring artificial intelligence into workplaces. I know, because I’ve seen it, as a sociologist^[8] who works with NASA’s robotic spacecraft teams.

The scientists and engineers I study are busy exploring the surface of Mars^[9] with the help of AI-equipped rovers. But their job is no science fiction fantasy. It’s an example of the power of weaving machine and human intelligence together, in service of a common goal.

Instead of replacing humans, these robots partner with us to extend and complement human qualities. Along the way, they avoid common ethical pitfalls and chart a humane path for working with AI.

The replacement myth in AI

Stories of killer robots and job losses illustrate how a “replacement myth” dominates the way people think about AI. In this view, humans can and will be replaced by automated machines^[11].

Amid the existential threat is the promise of business boons like greater efficiency^[12], improved profit margins^[13] and more leisure time^[14].

Empirical evidence shows that automation does not cut costs. Instead, it increases inequality by cutting out low-status workers^[15] and increasing the salary cost^[16] for high-status workers who remain. Meanwhile, today’s productivity tools inspire employees to work more^[17] for their employers, not less.

Alternatives to straight-out replacement are “mixed autonomy” systems, where people and robots work together. For example, self-driving cars must be programmed^[18] to operate in traffic alongside human drivers. Autonomy is “mixed” because both humans and robots operate in the same system, and their actions influence each other.

However, mixed autonomy is often seen as a step along the way to replacement^[20]. And it can lead to systems where humans merely feed, curate or teach AI tools^[21]. This saddles humans with “ghost work^[22]” – mindless, piecemeal tasks that programmers hope machine learning will soon render obsolete.

Replacement raises red flags for AI ethics. Work like tagging content to train AI^[23] or scrubbing Facebook posts^[24] typically features traumatic tasks^[25] and a poorly paid workforce^[26] spread across^[27] the Global South^[28]. And legions of autonomous vehicle designers are obsessed with “the trolley problem^[29]” – determining when or whether it is ethical to run over pedestrians.

But my research with robotic spacecraft teams at NASA^[30] shows that when companies reject the replacement myth and opt for building human-robot teams instead, many of the ethical issues with AI vanish.

Extending rather than replacing

Strong human-robot teams^[31] work best when they extend and augment^[32] human capabilities instead of replacing them. Engineers craft machines that can do work that humans cannot. Then, they weave machine and human labor together intelligently, working toward a shared goal^[33].

Often, this teamwork means sending robots to do jobs that are physically dangerous for humans. Minesweeping^[34], search-and-rescue^[35], spacewalks^[36] and deep-sea^[37] robots are all real-world examples.

Teamwork also means leveraging the combined strengths of both robotic and human senses or intelligences^[38]. After all, there are many capabilities that robots have that humans do not – and vice versa.

For instance, human eyes on Mars can only see dimly lit, dusty red terrain stretching to the horizon. So engineers outfit Mars rovers with camera filters^[39] to “see” wavelengths of light that humans can’t see in the infrared, returning pictures in brilliant false colors^[40].

Meanwhile, the rovers’ onboard AI cannot generate scientific findings. It is only by combining colorful sensor results with expert discussion that scientists can use these robotic eyes to uncover new truths about Mars^[42].

Respectful data

Another ethical challenge to AI is how data is harvested and used. Generative AI is trained on artists’ and writers’ work without their consent^[43], commercial datasets are rife with bias^[44], and ChatGPT “hallucinates”^[45] answers to questions.

The real-world consequences of this data use in AI range from lawsuits^[46] to racial profiling^[47].

Robots on Mars also rely on data, processing power and machine learning techniques to do their jobs. But the data they need is visual and distance information to generate driveable pathways^[48] or suggest cool new images^[49].

By focusing on the world around them instead of our social worlds, these robotic systems avoid the questions around surveillance^[50], bias^[51] and exploitation^[52] that plague today’s AI.

The ethics of care

Robots can unite the groups^[53] that work with them by eliciting human emotions when integrated seamlessly. For example, seasoned soldiers mourn broken drones on the battlefield^[54], and families give names and personalities to their Roombas^[55].

I saw NASA engineers break down in anxious tears^[56] when the rovers Spirit and Opportunity were threatened by Martian dust storms.

Unlike anthropomorphism^[58] – projecting human characteristics onto a machine – this feeling is born from a sense of care for the machine. It is developed through daily interactions, mutual accomplishments and shared responsibility.

When machines inspire a sense of care, they can underline – not undermine – the qualities that make people human.

A better AI is possible

In industries where AI could be used to replace workers, technology experts might consider how clever human-machine partnerships could enhance human capabilities instead of detracting from them.

Script-writing teams may appreciate an artificial agent that can look up dialog or cross-reference on the fly. Artists could write or curate their own algorithms to fuel creativity^[59] and retain credit for their work. Bots to support software teams might improve meeting communication and find errors that emerge from compiling code.

Of course, rejecting replacement does not eliminate all ethical concerns^[60] with AI. But many problems associated with human livelihood, agency and bias shift when replacement is no longer the goal.

The replacement fantasy is just one of many possible futures for AI and society. After all, no one would watch “Star Wars” if the ‘droids replaced all the protagonists. For a more ethical vision of humans’ future with AI, you can look to the human-machine teams that are already alive and well, in space and on Earth.

It can be challenging to create a treatment plan for depression. This is especially true for patients who aren’t responding to conventional treatments^[1] and are undergoing experimental therapies such as deep brain stimulation. For most medical conditions, doctors can directly measure the part of the body that is being treated, such as blood pressure for cardiovascular disease. These measurable changes serve as an objective biomarker of recovery that provides valuable information about how to care for these patients.

On the other hand, for depression and other psychiatric disorders, clinicians rely on subjective and nonspecific surveys^[2] that ask patients about their symptoms. When a patient tells their doctor they are experiencing negative emotions, is that because they are relapsing in their depression or because they had a bad day like everyone does sometimes? Are they anxious because their depression symptoms have lessened enough that they are experiencing new feelings, or do they have some other medical problem independent of their depression? Each reason may indicate a different course of action, such as altering a medication, addressing an issue in psychotherapy or increasing the intensity of brain stimulation^[3] treatment.

We are^[4] neuroengineers^[5]. In our study, newly published in Nature, we identified potential biomarkers^[6] for deep brain stimulation that could one day help guide clinicians and patients when making treatment decisions for those using this approach to alleviate treatment-resistant depression.

Biomarker for depression

Clinical depression does not respond to available therapies in a significant number of patients. Researchers have been working to find alternative options for those with treatment-resistant depression^[7], and many decades of experiments have identified specific brain networks with abnormal electrical activity in those with depression.

This notion of depression as abnormal brain activity rather than a chemical imbalance led to the development of deep brain stimulation^[8] as a depression treatment: a surgically implanted, pacemaker-like device that delivers electrical impulses to certain areas of the brain. Studies testing this technique have found that it can decrease depression severity^[9] over time in most patients.

Our research team wanted to find specific changes in brain activity that could serve as a biomarker that objectively measures how well deep brain stimulation is helping patients with depression. So we monitored the brain activity^[10] of 10 patients receiving deep brain stimulation for severe treatment-resistant depression over six months.

At the end of six months, 90% of the patients responded to the therapy – defined by a reduction of symptoms by at least a half – and 70% were in remission, meaning they no longer met the criteria for clinical depression.

To identify a potential biomarker, we developed an algorithm that looked for patterns in brain activity changes as patients recovered. The algorithm was based on data from six out of the original 10 patients who had usable data from the experiment. We found that there are coordinated changes in different frequencies^[11] present in the electrical activity within the area of the brain being stimulated. Using these patterns, the algorithm was able to predict whether someone was in a stable recovery with 90% accuracy each week.

Interestingly, we observed some parts of this pattern moved in the^[12] opposite direction^[13] later in stimulation therapy compared with the patterns at the start of therapy. This finding provides evidence that the long-term recovery is due to the brain adapting to the stimulation in a process called plasticity^[14] rather than as a direct effect of the stimulation itself.

We also saw other potential biomarkers worth investigating further.

For example, abnormalities in brain imaging taken before implanting the electrodes in specific parts of the brain correlated with how sick each patient was. This could provide clues about what’s causing depression in some people, or help develop imaging methods to determine who might be a good candidate for deep brain stimulation.

For another example, we found that the facial expressions of patients changed as their brains changed over the course of their treatment. While physicians often report this anecdotally, quantifying these changes may provide a way to develop objective markers of recovery that incorporate a patient’s behavior with their brain signals.

Because the results of our study are based on a small sample of patients, it’s important to further investigate how broadly they can be applied to other patients and newer deep brain stimulation devices.

Improving decision-making for depression

Clinical depression is a debilitating condition that causes significant personal and societal suffering^[16]. It is one of the largest contributors to the overall disease burden^[17] of many countries. Despite the many approved treatments available, nearly 30% of the 8.9 million U.S. adults^[18] taking medications for clinical depression continue to have symptoms.

Deep brain stimulation is one of the alternative therapies for treatment-resistant depression that researchers are investigating. Studies have shown that deep brain stimulation can offer effective and long-term relief^[19] for some patients.

Although deep brain stimulation is an approved treatment for other conditions like Parkinson’s disease^[20], it remains an experimental therapy for treatment-resistant depression. While the results from small experimental studies have been positive, they have not been successfully replicated in large-scale, randomized clinical trials^[21] necessary for approval from the U.S. Food and Drug Administration.

Finding an objective biomarker that measures recovery in depression has the potential to improve treatment decisions. For example, one patient in our study had a relapse after several months of remission. Were a biomarker available at the time, the clinical team would have had warning that the patient was relapsing weeks before standard symptom surveys showed that anything was wrong. Such a tool could help clinicians intervene before a relapse becomes an emergency.

From the aroma of fresh-cut grass to the smell of a loved one, you encounter scents in every part of your life. Not only are you constantly surrounded by odor, you’re also producing it. And it is so distinctive that it can be used to tell you apart from everyone around you.

Your scent is a complex product influenced by many factors, including your genetics. Researchers believe that a particular group of genes, the major histocompatibility complex^[1], play a large role in scent production. These genes are involved in the body’s immune response and are believed to influence body odor by encoding the production of specific proteins and chemicals.

But your scent isn’t fixed once your body produces it. As sweat, oils and other secretions make it to the surface of your skin, microbes break down and transform^[2] these compounds, changing and adding to the odors that make up your scent. This scent medley emanates from your body and settles into the environments around you. And it can be used to track, locate or identify a particular person, as well as distinguish between healthy and unhealthy people.

We are^[3] researchers who^[4] specialize in^[5] studying human scent through the detection and characterization of gaseous chemicals called volatile organic compounds^[6]. These gases can relay an abundance of information for both forensic researchers and health care providers.

Science of body odor

When you are near another person, you can feel their body heat without touching them. You may even be able to smell them without getting very close. The natural warmth of the human body creates a temperature differential with the air around it. You warm up the air nearest to you, while air that’s farther away remains cool, creating warm currents of air^[7] that surround your body.

Researchers believe that this plume of air helps disperse your scent by pushing the millions of skin cells you shed over the course of a day off your body and into the environment. These skin cells act as boats or rafts^[8] carrying glandular secretions and your resident microbes – a combination of ingredients that emit your scent – and depositing them in your surroundings.

Your scent is composed of the volatile organic compounds present in the gases emitted from your skin^[9]. These gases are the combination of sweat, oils and trace elements exuded from the glands in your skin. The primary components of your odor depend on internal factors such as your race, ethnicity, biological sex and other traits. Secondary components waver based on factors like stress, diet and illness. And tertiary components from external sources like perfumes and soaps build on top of your distinguishable odor profile.

Identity of scent

With so many factors influencing the scent of any given person, your body odor can be used as an identifying feature. Scent detection canines^[10] searching for a suspect can look past all the other odors they encounter to follow a scent trail left behind by the person they are pursuing. This practice relies on the assumption that each person’s scent is distinct enough that it can be distinguished from other people’s.

Researchers have been studying the discriminating potential of human scent for over three decades. A 1988 experiment demonstrated that a dog could distinguish identical twins living apart^[11] and exposed to different environmental conditions by their scent alone. This is a feat that could not be accomplished using DNA evidence, as identical twins share the same genetic code.

The field of human scent analysis has expanded over the years to further study the composition of human scent and how it can be used as a form of forensic evidence. Researchers have seen differences in human odor composition that can be classified based on sex, gender, race and ethnicity. Our research team’s 2017 study of 105 participants found that specific combinations^[12] of 15 volatile organic compounds collected from people’s hands could distinguish between race and ethnicity with an accuracy of 72% for whites, 82% for East Asians and 67% for Hispanics. Based on a combination of 13 compounds, participants could be distinguished as male or female with an overall 80% accuracy.

Researchers are also producing models to predict the characteristics of a person based on their scent. From a sample pool of 30 women and 30 men, our team built a machine learning model^[13] that could predict a person’s biological sex with 96% accuracy based on hand odor.

Scent of health

Odor research continues to provide insights into illnesses. Well-known examples of using scent in medical assessments include seizure and diabetic alert canines^[14]. These dogs can give their handlers time to prepare for an impending seizure or notify them when they need to adjust their blood glucose levels.

While these canines often work with a single patient known to have a condition that requires close monitoring, medical detection dogs can also indicate whether someone is ill. For example, researchers have shown that dogs can be trained to detect cancer^[15] in people. Canines have also been trained to detect COVID-19 infections^[16] at a 90% accuracy rate.

Similarly, our research team found that a laboratory analysis of hand odor samples^[17] could discriminate between people who are COVID-19 positive or negative with 75% accuracy.

Forensics of scent

Human scent offers a noninvasive method to collect samples. While direct contact with a surface like touching a doorknob or wearing a sweater provides a clear route for your scent to transfer to that surface, simply standing still will also transfer your odor into the surrounding area.

Although human scent has the potential to be a critical form of forensic evidence, it is still a developing field. Imagine a law enforcement officer collecting a scent sample from a crime scene in hopes that it may match with a suspect.

Further research into human scent analysis can help fill the gaps in our understanding of the individuality of human scent and how to apply this information in forensic and biomedical labs.

Uncommon Courses^[1] is an occasional series from The Conversation U.S. highlighting unconventional approaches to teaching. Title of course:Art & Science from Aristotle to InstagramWhat prompted the idea for the course?The idea for this course came out of my own research on intersections between art and science in the early modern period^[2], roughly 1400-1700. In this time, the division between the arts and sciences was not as stark as people perceive it to be today. Many natural philosophers – the scientists of their day – like Galileo Galilei^[3] made images in the process of conducting their studies. However, they also relied on artists and artisans to communicate their ideas to a wider audience – they needed engravers, draftsmen and other graphic arts practitioners to make the images that would go into their books and published works.In addition, throughout history the development of new technologies has affected artistic practices. The invention of the printing press and new photographic technologies allowed scientific ideas to be communicated in new ways to new audiences, but these inventions simultaneously created new artistic media. What does the course explore?In contemporary society, art and science are often characterized as diametrically opposed. However, knowledge making has been inextricably linked to image making since antiquity.

This image, made by Maria Sibylla Merian in 1705, is both a naturalist’s documentation and a work of art. Maria Sibylla Merian via Minneapolis Institute of Art^[4] One way we explore this relationship is by studying people from antiquity to the present who cross these realms. Leonardo da Vinci is a great example. People think of him as a master Renaissance painter, and he painted what is widely considered the most famous painting in the world, the Mona Lisa^[5]. But at the same time, he also pursued scientific questions about anatomy^[6], botany^[7] and motion^[8] and was an inventor^[9].But there were other examples of people who pursued science and art together. In the 19th century, Anna Atkins^[10] was one of the first people to use an early photographic technique – the cyanotype – to study British plants and algae. The images she created^[11] are aesthetically beautiful but also created new knowledge within botany.In the course, we also explore different technological developments that affected the arts, creating new materials and media. These include technologies such as the printing press^[12], camera obscura^[13], daguerreotype^[14] and digital art^[15].Why is this course relevant now?We live in a visually saturated world, yet we often take in these images uncritically. My students encounter images in every aspect of their lives, in greater quantity and at a greater rate than ever before. Yet, people frequently accept these images as true depictions of reality, even when they are not.Why do people assume a scientific image is divorced from the same aesthetic choices and manipulation that are applied to the image on a magazine cover? Why do people accept a scientific image as objective and not a created object like a painting? Issues like photoshopped images or AI-generated artworks may seem unique to the modern moment, but concerns about manipulation and deception have a long history. What’s a critical lesson from the course?Today, the perceived division between empirical and quantitative science and creative and qualitative arts is even more pronounced than in the past. In my classes, I find science students often think that a scientific image made today is strictly true or objective. Yet during the course they discover that many choices get made in constructing that image. What information should be included? What information should be left out? The art students in the class soon come to realize that many of the artistic materials and media they rely on, be it synthetic pigment or digital technologies, were developed for scientific or engineering purposes. What materials does the course feature?“The Republic^[16]” (fourth century BCE) by Plato, where we consider his skepticism of the arts due to their ability to deceive. “De Humani Corporis Fabrica^[17]” (1543) by Andrea Vesalius, an important book on human anatomy where the illustrations and text were equally influential. Images from the Hubble Space Telescope^[18], and how they can be considered both works of art and science. What will the course prepare students to do?It is my hope that after taking this course, students will have gained the skills to be more discerning in how they think about the ways the visual information around them is created. They will not only have a greater appreciation for the processes of creating artistic and scientific knowledge but also have gained a critical lens for assessing the images they see around them.

When 17 people^[1] were in orbit around the Earth all at the same time on May 30, 2023, it set a record. With NASA and other federal space agencies planning more manned missions and commercial companies bringing people to space, opportunities for human space travel are rapidly expanding.

However, traveling to space poses risks to the human body. Since NASA wants to send a manned mission to Mars^[2] in the 2030s, scientists need to find solutions for these hazards sooner rather than later.

As a kinesiologist who works with astronauts, I’ve spent years studying the effects^[3] space can have on the body and brain. I’m also involved in a NASA project that aims to mitigate the health hazards^[4] that participants of a future mission to Mars might face.

Space radiation

The Earth has a protective shield called a magnetosphere^[5], which is the area of space around a planet that is controlled by its magnetic field^[6]. This shield filters out cosmic radiation^[7]. However, astronauts traveling farther than the International Space Station will face continuous exposure to this radiation – equivalent to between 150 and 6,000 chest X-rays^[8].

This radiation can harm the nervous and cardiovascular systems^[9] including heart and arteries^[10], leading to cardiovascular disease. In addition, it can make the blood-brain barrier leak^[11]. This can expose the brain to chemicals and proteins that are harmful to it – compounds that are safe in the blood but toxic to the brain.

NASA is developing technology that can shield travelers on a Mars mission from radiation by building deflecting materials such as Kevlar and polyethylene into space vehicles and spacesuits^[12]. Certain diets and supplements such as enterade^[13] may also minimize the effects of radiation. Supplements like this, also used in cancer patients on Earth during radiation therapy, can alleviate gastrointestinal side effects of radiation exposure.

Gravitational changes

Astronauts have to exercise in space to minimize the muscle loss they’ll face after a long mission. Missions that go as far as Mars will have to make sure astronauts have supplements^[14] such as bisphosphonate^[15], which is used to prevent bone breakdown in osteoporosis. These supplements should keep their muscles and bones in good condition over long periods of time spent without the effects of Earth’s gravity^[16].

Microgravity also affects the nervous and circulatory systems. On Earth, your heart pumps blood upward, and specialized valves in your circulatory system keep bodily fluids from pooling at your feet. In the absence of gravity, fluids shift^[17] toward the head.

My work and that of others has shown that this results in an expansion of fluid-filled spaces in the middle of the brain. Having extra fluid in the skull and no gravity to “hold the brain down” causes the brain to sit higher in the skull^[18], compressing the top of the brain against the inside of the skull.

These fluid shifts may contribute to spaceflight associated neuro-ocular syndrome^[20], a condition experienced by many astronauts that affects the structure and function of the eyes^[21]. The back of the eye can become flattened, and the nerves that carry visual information from the eye to the brain swell and bend. Astronauts can still see, though visual function may worsen for some. Though it hasn’t been well studied yet, case studies suggest^[22] this condition may persist even a few years after returning to Earth.

Scientists may be able to shift the fluids back toward the lower body using specialized “pants^[23]” that pull fluids back down toward the lower body like a vacuum. These pants could be used to redistribute the body’s fluids in a way that is more similar to what occurs on Earth.

Mental health and isolation

While space travel can damage the body, the isolating nature of space travel can also have profound effects on the mind^[24].

Imagine having to live and work with the same small group of people, without being able to see your family or friends for months on end. To learn to manage extreme environments and maintain communication and leadership dynamics, astronauts first undergo team training on Earth.

They spend weeks in either NASA’s Extreme Environment Mission Operations^[25] at the Aquarius Research Station^[26], found underwater^[27] off the Florida Keys, or mapping and exploring caves with the European Space Agency’s CAVES program^[28]. These programs help astronauts build camaraderie with their teammates and learn how to manage stress and loneliness in a hostile, faraway environment.

Researchers are studying how to best monitor and support behavioral mental health^[29] under these extreme and isolating conditions.

While space travel comes with stressors and the potential for loneliness, astronauts describe experiencing an overview effect^[30]: a sense of awe and connectedness with all humankind. This often happens when viewing Earth from the International Space Station.

Learning how to support human health and physiology in space also has numerous benefits for life on Earth^[32]. For example, products that shield astronauts from space radiation and counter its harmful effects on our body can also treat cancer patients receiving radiation treatments.

Understanding how to protect our bones and muscles in microgravity could improve how doctors treat the frailty that often accompanies aging. And space exploration has led to many technological achievements advancing water purification^[33] and satellite systems^[34].

Researchers like me who study ways to preserve astronaut health expect our work will benefit people both in space and here at home.

Online shopping isn’t just a convenient way to buy batteries, diapers, computers and other stuff without going to a brick-and-mortar store.

Many Americans also use the internet to quietly acquire illegal, fake and stolen items^[1]. Guns^[2], prescription drugs no doctor has ordered and checks^[3] are on this long list, as well as cloned credit cards^[4], counterfeit passports and phony driver’s licenses^[5].

Because buyers and sellers alike realize that the authorities can detect illegal online transactions, criminals and their customers prefer covert online platforms that protect user anonymity, such as Tor^[6], or encrypted messaging applications like Telegram and WhatsApp^[7]. Buyers and sellers also use digital wallets^[8] and cryptocurrencies to further conceal^[9] their identities.

As scholars of^[10] high-tech crime^[11], we were eager^[12] to solve a riddle. Having these items shipped to the buyers’ homes or offices would make it easy for authorities to catch them. So how do people who buy these illegal items maintain their anonymity when they take possession of items they purchased on the dark web^[13]?

They mostly use vacant residential properties, called “mule addresses^[14]” or “drop addresses^[15].” Once the illegal goods or phony documents get delivered – presumably without the owners’ knowledge – to the doorstep of the uninhabited home, the buyer or a middleman picks it up. This practice makes it very hard to trace these transactions.

Penchant for sharing

To discover where these items change hands, we took advantage of the inclination of some of the criminal vendors to share images on Telegram of the parcels they send, along with the illicit items.

They use this strategy to build their reputations, earn the trust of buyers and market their services.

Not all users of online underground markets do this, but we still spotted thousands of packages delivered this way over a period of two years.

In one case, we found a photo of a forged or stolen check alongside the mailed envelope used for its delivery on a Telegram channel dedicated to trading stolen and counterfeit checks.

The label on the envelope bears not only the shipping date but also the Wyoming address where it was sent. Armed with this information, anyone can retrieve related details by searching online. We found an apartment complex at that address with several units for rent.

Guns, drugs and rentals

We also found that criminal vendors use mule addresses as their sender address. In one example, we found a video, uploaded in April 2023, of an assault rifle shipped from an Arizona address. At the time, that property was for sale.

The video displays an assault rifle apparently shipped from that address after being purchased online on an underground gun market. At the time, that property was for sale.

We found a similar video documenting the punctual delivery of what we believe to be illegal drugs. Considering that the video has been circulating in illegal drugs markets that we monitor, it’s reasonable to assume that the package contains narcotics or prescription drugs.

The footage portrays a satisfied customer who has just gotten the drugs. We looked up the recipient’s address, which is discernible in the video.

It’s a property in North Las Vegas, Nevada, which was listed for sale at the time of delivery – although it seems to have later been sold. The anticipated delivery date, March 28, 2023, coincided with the day the package in the video was received.

One of the illegal digital marketplaces we identified is a hub for prescription- sales of OxyContin, Viagra, Adderall and Valium. It’s linked to an administrator who presides over several Telegram channels.

The administrator has shared photos on those channels that allowed us to see tracking numbers associated with packages they’d mailed. By collating the tracking numbers from April 20 to May 23, 2023, we compiled a comprehensive database of those addresses and the statuses of those properties when the packages were delivered.

We found that 72% of the 650 deliveries in this database were to properties listed for sale, and the rest were to properties unoccupied for other reasons. The average time that elapsed between a property listing and an illicit package being delivered there was nine days.

Be on guard

We haven’t yet learned of any criminals who were convicted of criminally using mule addresses to deliver illegal packages.

Because criminals take advantage of vacant residential properties listed for sale or rent by unsuspecting homeowners to protect their anonymity, we believe that it’s important for landlords and people who are selling or renting homes to protect themselves from these crimes of commerce.

Some of the same strategies that enhance safety in other regards can help, such as installing surveillance cameras and employing property managers.