Skip to content
Sensemaking
Confronting UnknownsPaper

Working paper · June 2026

Confronting Unknowns: A Framework for Anomalous Aerospace Events

The Confronting Unknowns ’26 Program · MIT
Download PDF (opens in new tab)

Abstract

How should institutions act when something appears in the sky that resists immediate attribution? This paper proposes the Confronting Unknowns Framework: Detect, Identify, Respond, Explain (), a pipeline whose stages each fail when the adjacent stage fails. The problem it addresses is institutional, not phenomenological: the United States lacks the detection architecture, the inference methodology, and the communication infrastructure to say what it does and does not know about objects in its own airspace. The framework is applied to the 2024–2025 drone-wave incidents and is agnostic about what the objects ultimately are, and opinionated about what we owe each other when we cannot yet say.

Contents

Read the paper

13 chapters · about 81 min end to end

  1. Overview

    Executive Summary

    1. 01Executive Summary3 minWhy the institutional work of confronting anomalous aerospace events is overdue regardless of what the objects are: the three converging forces, what New Jersey cost, the four-stage Detect-Identify-Respond-Explain pipeline, and the four core recommendations.

  2. Part I

    The Problem

    1. 02Why Now Is Different3 minThree forces have converged that did not previously exist together: a crowded airspace, a contradictory legislative response, and a compressed information environment. None depends on what the objects turn out to be; together they leave less margin for getting the next response wrong.
    2. 03Something Above New Jersey1 minThe New Jersey drone wave of late 2024: a 76-day arc from the first Picatinny sighting to a White House non-answer, and the institutional failures it exposed.
    3. 04A Pattern of Incidents Worldwide3 minNew Jersey was not isolated. A worldwide pattern of unexplained airspace incidents reveals the same detection, coordination, and communication gaps.
    4. 05The Confronting Unknowns Framework6 minThe Confronting Unknowns Framework carries uncertainty through a four-stage loop: Detect, Identify, Respond, Explain (D-I-R-E). Each stage targets the uncertainty left by the one before it.
    5. 06The CU Crisis Playbook for Leaders5 minWhen an anomalous event breaks, leaders need a response in minutes: classify the situation, work the contact chain, guard against decision traps, and match the response to the threat.

  3. Part II

    Detect & Identify

    1. 07Detecting Unknowns10 minThe science of seeing clearly: observation credibility, the observer-education feedback loop, hotspots and bias, and sensor capabilities and their limits.
    2. 08Identifying Unknowns13 minFrom binaries to probabilities: ranking five mutually exclusive hypotheses (H1–H5) under explicit uncertainty, AI-assisted anomaly resolution, and dual-use detection pathways.

  4. Part III

    Respond & Explain

    1. 09The New Jersey Media Melee7 minHow institutional communication broke down during the New Jersey wave: conflicting explanations, conspiracy dynamics, and the escalation ladder from watching to shooting.
    2. 10A Century of Anomaly Miscommunication6 minStructural communication failures repeat across a century of anomalous events. The case for unified incident command and threat planning.
    3. 11Communicating Uncertainty with Clarity5 minHow to communicate uncertainty with clarity, what each audience needs to hear, and what leaders owe the public when answers do not yet exist.
    4. 12Diagnostic Markers for the Next Incident1 minThe observable markers that tell you, in real time, which failure mode the next incident is in, so the response can match the situation rather than the headline.

  5. Part IV

    An Action Agenda

    1. 13An Action Agenda for New Unknowns18 minA four-part action agenda: support science for a new era, enhance detection and identification, improve incident readiness and response, and engage the public, closing with the argument that does not depend on the answer.

Where do you start?

This research is written to be useful across disciplines. Choose the path matched to how you engage with the material.

Essentials

Congressional staff, journalists, policy generalists

The fastest credible orientation: why this moment is different, what leaders owe the public, and how to recognize the next incident as it unfolds.

Operator

Emergency managers, agency leads, congressional-oversight staff

The operational core: the decisions you have to make in the first fifteen minutes, the agenda that fixes the system, and the objection worth taking seriously.

Researcher

Researchers, sensor specialists, ML practitioners

The technical spine: the science behind detection and identification, a fully worked Bayesian case, and the sensor capabilities the framework depends on.

The framework

D

Detect

Is something there? Capture and preserve the evidence.

I

Identify

What is it? Rank hypotheses under explicit uncertainty.

R

Respond

What to do, and who decides?

E

Explain

What to say, by whom, and when?

  1. Detect: Is something there? Capture and preserve the evidence.
  2. Identify: What is it? Rank hypotheses under explicit uncertainty.
  3. Respond: What to do, and who decides?
  4. Explain: What to say, by whom, and when?
The Confronting Unknowns framework: D·I·R·E

The hypothesis space

Every observation is weighed across five hypotheses, not sorted into “explained” or “unexplained.”

  1. Data or Sensor Artifact

    The apparent event is primarily a product of sensor behavior, processing error, or interpretation failure.

  2. Natural Physical Source

    A known, real physical phenomenon from atmospheric, astronomical, oceanic, or geophysical causes.

  3. Human-Made Physical Source (unclassified)

    A real object consistent with conventional or commercially accessible technology.

  4. Human-Made Physical Source (classified)

    A real object associated with restricted state or defense activity.

  5. Unknown Physical Phenomenon

    A real physical phenomenon not adequately explained by H1–H4: the open residual that holds probability for explanations not yet named.

The five hypotheses

Worked example · interactive

Worked Example: A Bayesian Case Assessment

A toy probabilistic case, drawn from the New Jersey wave, worked in two stages: first the reported civilian and investigation evidence, then a single uncorroborated anomalous-kinematics report. The lesson is what does not happen: a dramatic lone sensor reading widens the range of live possibilities rather than resolving the case toward "unknown."

Open the worked example

Figures

The New Jersey drone-wave timeline, November 2024 to January 2025.Figure by the authors of Confronting Unknowns, illustrated by Jonathan Miller.
Major U.S. incursions and the New Jersey wave across geography and time.Figure by the authors of Confronting Unknowns, illustrated by Jonathan Miller.
Europe's 2025 drone wave and related incidents.Figure by the authors of Confronting Unknowns, illustrated by Jonathan Miller.
The Confronting Unknowns Framework (D-I-R-E), with feedback at every junction.Figure by the authors of Confronting Unknowns, illustrated by Jonathan Miller.
The sensor shortcut: evidence quality over quantity.Figure by the authors of Confronting Unknowns, illustrated by Jonathan Miller.
Algorithmic blind spots in legacy radar processing.Figure by the authors of Confronting Unknowns, illustrated by Jonathan Miller.
Typical likelihoods across the hypotheses H1 to H5 (illustrative).Figure by the authors of Confronting Unknowns, illustrated by Jonathan Miller.
A toy example of Bayesian likelihood-ratio analysis.Figure by the authors of Confronting Unknowns, illustrated by Jonathan Miller.

Foreword

In early 2026, dozens of engineers, scientists, intelligence analysts, pilots, policy researchers, and communications specialists gathered during a snowstorm at the Massachusetts Institute of Technology to contribute to a new, independent activity on sensemaking and the challenge of confronting anomalous aerospace phenomena. What they produced is in your hands. We offer it to anyone who finds it useful, though we wrote with specific audiences in mind: congressional staff, agency analysts, journalists, scientists, and the concerned citizens trying to make sense of objects that move between sky and sea, seemingly undeterred.

Our program emphatically sidesteps evangelizing speculative or belief-based declarations. Nor are we advocating for any explanation or attribution of unidentified anomalous phenomena. Our argument in this paper is that the United States and its allies must embrace three things to say what they do and do not know about objects in their own airspace: a detection architecture, an inference methodology, and a communication infrastructure.

The framework does not require taking any position on what anomalous aerospace phenomena are. It asks leaders to treat recognizing, identifying, confronting, and embracing unknowns as a natural part of the work. The future infrastructure for approaching unknowns will be what its participants make of it, and we are writing this from inside that participation.

To the cohort who came together for a "no-expense-paid trip" during a very Bostonian winter week: the energy you channeled into the Sensemaker Spotlights, the Uncertainty Game, and the Writers' Workshop shapes every page. Look without flinching, speak without overreaching, and build the institutions that let the truth catch up to the data.

Jonathan "JMill" Miller, Lead of the Confronting Unknowns Independent Activity at MIT.

Authors

This document was collaboratively authored by volunteers among the Confronting Unknowns Independent Activity cohort.

Contributing authors: Aashka Dave (University of Virginia; UNC Chapel Hill), Theodora Skeadas (King's College London; Harvard University), and Dorothy Wu (Yale University), with additional contributors who shall remain anonymous.

Editorial team: Victoria Cheng (UQAM; Harvard University), Davide Lasi (MIT), Kevin Romero (Johns Hopkins University), Sri Tata (Yale University), and Susan Winterberg (Harvard University).

Coordinating lead and corresponding author: Jonathan "JMill" Miller (MIT).

With gratitude to our pre-publication reviewers (Eric Evans, Jordan Flowers, Ryan Graves, Marik von Rennenkampff, Jessica Souder, and select reviewers who shall remain anonymous), whose corrections and pushback sharpened the paper. All errors that remain are ours.

Suggested citation

The Confronting Unknowns '26 Program (2026). *Confronting Unknowns: A Framework for Anomalous Aerospace Events*. Sensemaking, June 2026. Licensed CC BY-NC 4.0. https://sensemaking.wtf/work/cu26-p01

From the program

Explore the program at MIT’s Independent Activities Period

Sensemaking.wtf is operated independently and is not an MIT-published property. JMill teaches Confronting Unknowns at MIT; the research publishes here. Claims in this paper belong to their authors, not to MIT.