Havana Syndrome is the name given to unexplained health incidents reported by diplomats, intelligence officers and their families, initially in Cuba and later elsewhere, with symptoms including head pain, dizziness, balance problems and cognitive difficulties (National Academies assessment). The question is not whether people have suffered, but what caused their symptoms, whether the same explanation fits different incidents, and what evidence can establish responsibility (NIH findings and limitations).
Our reporting has followed the story from The Sound, Brazen’s 2023 podcast, to the science of a possible weapon, affected families’ accounts and allegations that government agencies failed their own personnel. This guide brings that work together and distinguishes our reporting and analysis from official assessments, scientific findings and questions still unresolved.
Last reviewed: 25 September 2026. An important change since our January report: the 2023 and 2025 intelligence assessments were recalled in June 2026 (House Intelligence Committee statement). The recall does not itself identify a cause or perpetrator; the evidence and limits of the decision are explained below.
Start with our reporting · Listen to The Sound · Developments in 2026 · Read the evidence · Timeline · Reporting notes and contact
These five pieces form the main written investigation, in publication order. Some require a free account; this guide is intended to remain open to everyone.
For the earlier starting point, read “The Mystery of Havana Syndrome,” published 23 January 2023. Bradley Hope introduces a guest contribution by Zachary Dorfman on the intelligence-community disputes surrounding the story and Dorfman’s reporting contribution to The Sound.
The Sound: Mystery of Havana Syndrome is an eight-part investigation hosted by Nicky Woolf, produced by Project Brazen and Goat Rodeo in partnership with PRX. Released from January to March 2023, it follows the early incidents, the search for a mechanism and the human consequences; it is a record of that investigation, not a substitute for the later developments in this guide (episode archive).
The International Spy Museum discussion adds a conversation with Woolf, Polymeropoulos and Zaid, moderated by historian Andrew Hammond. You can also find the series on Apple Podcasts.
In January, CNN reported that the Pentagon had been evaluating a device acquired through a covert operation, with some investigators suspecting a link to Havana Syndrome (CNN, 13 January 2026). Our follow-up the next day recorded responses from sources who said they had heard about such testing.
On 8 March, CBS’s 60 Minutes reported that undercover Homeland Security agents had obtained a miniaturised microwave device through a Russian criminal network and that U.S. military tests on animals produced injuries consistent with those seen in people (CBS transcript). CBS said it had not seen the device: the account depended on confidential sources, the testing remained classified, and the Pentagon declined to comment (CBS’s account and responses).
That reporting changes what readers should know about the search for a device, but does not publicly establish that the reported equipment caused any particular incident (CBS transcript). A reported Russian component or purchase through a Russian criminal network is not, by itself, proof of an operation directed by the Russian state.
On 11 June, House Intelligence Committee chairman Rick Crawford welcomed Director of National Intelligence Tulsi Gabbard’s recall of two assessments on anomalous health incidents (committee statement). CNN identified them as the assessments released in 2023 and January 2025 and reported ODNI’s criticism that they excluded relevant intelligence, suppressed alternative analysis and relied on flawed processes (CNN, 12 June 2026).
Those criticisms are consequential official statements about the assessments’ integrity, not a published technical demonstration of a weapon or an adjudicated finding against a perpetrator (committee statement). Our review for this guide did not locate a public replacement assessment establishing a cause or responsibility; the older assessments are linked here as historical records, explicitly marked as recalled.
The HAVANA Act of 2021 authorised payments to eligible affected personnel and families. In July 2026, the BBC reported nearly $3 million in compensation payments, while describing the cause of the illness as unresolved (BBC, 11 July 2026).
Recognition of suffering and access to care should not be confused with a forensic finding about what caused an individual incident. Our reporting on families’ medical experiences and the government response follows both the health consequences and the fight for accountability.
The National Academies’ 2020 assessment found directed, pulsed radiofrequency energy the most plausible mechanism among those it considered, particularly for the distinctive early symptoms described by some personnel. The committee also said no hypothesis had been proven, lacked individual-level information, and considered it likely that multiple factors explained some cases and differences between them (report summary).
The distinction matters when reading our science and engineering investigation: a mechanism that could produce an effect still needs evidence connecting it to an actual exposure. The article’s device renderings and interactive visualisation are exploratory illustrations, not photographs of recovered equipment or validated reconstructions of an incident.
In March 2024, NIH researchers reported no significant MRI-detectable brain injury or differences in most clinical measures between the participants reporting incidents and comparison groups (NIH summary). NIH stressed that symptoms could be real, disabling and prolonged, and that the absence of an MRI-detectable difference did not exclude an earlier adverse event affecting the brain (NIH’s stated limitations).
In September 2024, NIH said it had stopped the study after finding informed-consent requirements were not met because of coercion, which it said was not perpetrated by NIH researchers (MedPage Today’s report of NIH’s response). NIH also said those findings did not affect the study’s conclusions; stopping the study on ethical grounds should not be described as a scientific retraction (same report).
Our Lenzi investigation reports individual clinical findings and testimony that readers should consider alongside the broader studies. Different patient groups, testing dates and methods complicate direct comparisons; no single scan or group result can settle every reported case (study differences and responses).
The historical January 2025 intelligence assessment, recalled in June 2026, drew a distinction between whether an adversary possessed a relevant capability and whether it had used one. Most components retained the “very unlikely” judgment on foreign responsibility, while one assessed a “roughly even chance” of use against a small, undetermined subset; those were intelligence judgments, not identified perpetrators or resolved individual cases (assessment text).
A separate March 2024 investigation by 60 Minutes, The Insider and Der Spiegel reported evidence potentially connecting Russian military intelligence Unit 29155 to incidents, while acknowledging that the evidence was circumstantial (joint investigation). Russia denied the allegations, with Kremlin spokesman Dmitry Peskov calling them unfounded (Reuters, 1 April 2024).
Our November 2025 reporting argues for Russian responsibility by examining capability, history, motive and interviews. Historical precedent can justify an investigative lead; it cannot substitute for evidence linking a particular actor, device and exposure.
Our government-response investigation points readers to the State Department’s June 2018 Havana Accountability Review Board report. For the older Moscow history, the National Security Archive hosts a 10 December 1975 Kissinger–Dobrynin telephone record; it documents an earlier dispute over microwave signals, not proof of what caused the later incidents.
The next useful evidence would address specific gaps rather than simply repeat a preferred explanation:
This guide draws on the six dedicated Whale Hunting articles linked above, Brazen’s original podcast archive, primary public records and subsequent reporting. It does not reproduce unpublished interviews, claim access to classified device tests, or treat descriptions of records in another publication as our independent examination of those records.
The written investigation credits Bradley Hope and Clara Preve; the 2023 introduction includes Zachary Dorfman’s guest contribution, and The Sound is hosted by Nicky Woolf (written reporting; 2023 introduction; series credits). Guide editor: Bradley Hope. Earlier articles retain their publication dates and arguments; this page supplies a dated route through them rather than silently rewriting the archive.
For information relevant to this investigation, contact syndrome@projectbrazen.com. For corrections or other editorial queries, use Whale Hunting’s contact page; choose a suitable confidential channel before sending sensitive material.
Update record: First guide prepared 25 September 2026, including the reported device testing and June recall of the intelligence assessments. Substantive future revisions should identify what changed and when.
Browse all Havana Syndrome coverage or return to Investigations.
