AIPolicyTracker
Menu

AI incident ·

Bodycam Footage Reportedly Contradicted Purportedly ChatGPT-Generated Use-of-Force Narrative by Immigration Agent

1 news report Snapshot 7 Sep 2026

In brief

An AI system built by Openai and deployed by United States Border Patrol, United States Department Of Homeland Security and 2 others allegedly harmed Individuals Detained During Operation Midway Blitz, Protesters And Bystanders In Chicago Neighborhoods and 5 others.

Risk domain
Misinformation False or misleading information
Occurred
Coverage
1 reportNov 2025

What happened

Body-worn camera footage released in litigation over Operation Midway Blitz reportedly showed that an immigration agent used ChatGPT to generate a long-form use-of-force narrative based on minimal inputs. A federal judge found that multiple AI-assisted reports conflicted with contemporaneous video evidence, undermining their accuracy and credibility and raising concerns about the use of generative AI in official law enforcement reporting.

Laws that address this harm

Policy angle: Classified under Misinformation (False or misleading information) in the MIT AI Risk Repository taxonomy; 5 recorded instruments address this use case.

Matched from the record's risk domain and country to the instruments recorded here. A reviewer can correct the match in the repository (data/external/incident_overrides.yaml).

News reports (1)

Titles link to the original publisher; report text is not reproduced here.

  1. Bodycam footage from Operation Midway Blitz released: ‘It’s all about arresting people’
    chicagotribune.com · Jason Meisner, Madeline Buckley, Jonathan Bullington

Who was involved

Alleged developer
Openai
Alleged harmed party
Individuals Detained During Operation Midway Blitz, Protesters And Bystanders In Chicago Neighborhoods, General Public, General Public Of Chicago, General Public Of The United States, Epistemic Integrity, Judicial Integrity

Classification (MIT AI Risk Repository taxonomy)

Risk domain
Misinformation
Causal entity
Human
Intent
Intentional
Timing
Post-deployment
Harm level
—
Sectors
—
Countries
—

Risk entries describing this failure mode

Entries from the MIT AI Risk Repository coded to subdomain 3.1.

  • Pursuing Consistent Context

    "LLMs have been demonstrated to pursue consistent context [129]–[132], which may lead to erroneous generation when the prefixes contain false information. Typical examples include sycophancy [129], [130], false demonstra...

    Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)

  • Defective Decoding Process

    In general, LLMs employ the Transformer architecture [32] and generate content in an autoregressive manner, where the prediction of the next token is conditioned on the previously generated token sequence. Such a scheme...

    Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)

  • Noisy Training Data

    "Another important source of hallucinations is the noise in training data, which introduces errors in the knowledge stored in model parameters [111]–[113]. Generally, the training data inherently harbors misinformation....

    Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)

  • Hallucinations

    "LLMs generate nonsensical, untruthful, and factual incorrect content"

    Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)

  • Faithfulness Errors

    "The LLM-generated content could contain inaccurate information" which is is not true to the source material or input used

    Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)

  • Factuality Errors

    "The LLM-generated content could contain inaccurate information" which is factually incorrect

    Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)

  • Untruthful Content

    "The LLM-generated content could contain inaccurate information"

    Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)

  • Knowledge Gaps

    "Since the training corpora of LLMs can not contain all possible world knowledge [114]–[119], and it is challenging for LLMs to grasp the long-tail knowledge within their training data [120], [121], LLMs inherently posse...

    Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)

Incidents in the same risk subdomain

All incidents in this subdomain

Source record: incident #1299 on the AI Incident Database · all 1 report