CBP Wants AI-Powered ‘Quantum Sensors’ for Finding Fentanyl in Cars

United States Customs and Border Protection is paying General Dynamics to create a prototype of “quantum sensors” alongside a “database with artificial intelligence” designed “to detect illicit objects and substances (such as fentanyl) in vehicles, containers, and other devices,” according to a contract justification published in a federal register last week.

“This database and sensor project will integrate advanced quantum and classical sensing technologies with Artificial Intelligence and ultimately deploy proven concepts and end products anywhere in the CBP environment,” the justification document reads. “Under this requirement, CBP will take additional steps to enhance its ability to detect, and thus, significantly reduce the harms of illicit contraband entering the United States of America, thus bolstering national security.”

The document redacts the name of the company developing the prototype; however, contract details included in the federal register entry reveal that the justification is for a $2.4 million General Dynamics contract that has been public since December 2025.

CBP and General Dynamics did not respond to WIRED’s requests for comment.

CBP’s request for a prototype of “quantum sensors” with an AI database—which comes amid a widespread push within the Department of Homeland Security (DHS) “to support the adoption and scaling of AI technologies,” according to a strategy memorandum published last year—involves a real and growing area of scientific and technological research.

Last week’s justification does not get into detail about which methods its “quantum sensors” would use or what information the AI database would store and analyze. However, it does provide hints about detection methods the agency has considered.

The document claims that CBP conducted market research from April through October of 2025. In July, CBP published an information request seeking a vendor for exactly 35 handheld "Gemini” analyzers, sold by Thermo Fisher Scientific, which are designed to identify unknown chemicals and narcotics.

DHS has also tested the Gemini in previous years, according to reports published in 2021 and 2023. The July request—which notes that the devices would be used to identify substances like fentanyl, ketamine, cocaine, methamphetamine, diazepam, and MDMA—makes no mention of artificial intelligence or a database.

“The detection equipment will be used by CBP Officers in non-intrusive testing to detect a wide range of narcotics, controlled substances, unknown substances, and general organic materials,” the request reads, noting that the agency “continues to seize an increasing number of opioids at the nation’s borders.”

The July request for information claims that the Gemini analyzers use “Fourier Transform Infrared Spectroscopy (FTIR),” which measures how much infrared light a sample absorbs, and “Raman spectroscopy,” which measures how light scatters off the surface of a sample when a laser is directed at it.

Last week’s contract justification says that the agency found an American company that creates a “handheld analyzer” for identifying dangerous chemicals but claimed it “cannot detect fentanyl.” It’s unclear whether this was referring to Gemini or one of the more than 10 other devices DHS tested in 2021 and 2023. But when reached for comment, Thermo Fisher Scientific said that its Gemini analyzers “are designed to detect fentanyl.”

It’s also unclear whether the General Dynamics prototype may use FTIR or Raman spectroscopy. But a 2024 working paper about a laboratory-based fentanyl-detection method (unrelated to CBP, General Dynamics, or Thermo Fisher Scientific) notes that “portable Raman spectrometers” and other handheld devices—though convenient, fast, and inexpensive—can “struggle with detection of fentanyl” and may be prone to “false-positive and false-negative results.”

While it remains ambiguous what exactly last week’s justification was referring to with its mention of “quantum” sensors, there are fentanyl detection methods based in quantum chemistry. The 2024 paper, for instance, explains how “quantum dots” and fluorescent dye can be used to detect fentanyl and 58 of its analogues.

Matthew Webber, an engineering professor and molecular science researcher at the University of Notre Dame and one of the coauthors on the 2024 paper, tells WIRED that the “quantum dots” are an artificially made, graphene-based nanomaterial. When combined with a fluorescent dye and a synthetic molecule that acts like a “basket,” Webber says, there’s visible and quantifiable fluorescence. When a few micrograms of fentanyl are added to this mixture, Webber says, it sticks to the quantum dots and “outcompetes” the fluorescent dye, causing a loss of fluorescence.

Webber notes that all fentanyl-related lab research is conducted in a “super-controlled environment,” and never entails large quantities of fentanyl for legal, practical, and safety reasons. He emphasizes the sensitive nature of discussing fentanyl research, and says that labs like his never have bags of fentanyl laying around, an incorrect impression some people may have.

When asked about CBP’s request for an “AI database” accompanying the quantum sensors, Webber says that AI could potentially help with matters like “spectral deconvolution.”

“If you have signals coming from multiple agents within a mixture, the eye may not be able to see to deconvolve the spectra into individual components, Webber says. “But there’s potentially AI-based specialty convolution frameworks that could be powerful in that context.”

In other words, since new street drugs can constantly move the goalposts of detection, an AI-based system that processes images and recognizes patterns may catch fentanyl in situations where a human could miss it.