Esc
EthicsCase Closed

3M expert used ChatGPT to draft defense in fatal explosion trial

Is this a scandal?

No longer — the story has resolved. Noise 27/100, cooling down, across 1 source.

SCAND-202656as of Methodology
Cite this incident"3M expert used ChatGPT to draft defense in fatal explosion trial." SCAND.Ai incident SCAND-202656, noise 27/100 as of September 11, 2026. https://scand.ai/scandal/3m-expert-chatgpt-defense-report-fatal-explosion-trial
FORECASTForecast, not fact

Courts will likely mandate explicit disclosure of AI assistance in expert reports because this verdict signals judicial skepticism toward undisclosed generative workflows in high-stakes litigation.

27

Noise 27/100 — louder than 98% of tracked AI controversies.

AI-assisted analysis · How we work

Why it matters

This case establishes legal precedent regarding AI-generated expert testimony admissibility and highlights liability risks when professionals outsource critical analysis to LLMs.

Key points

  1. Harris County jury awarded $61 million to plaintiffs after learning 3M's expert relied heavily on ChatGPT for his defense report.
  2. Expert Josh Autenrieth billed $475 per hour while using prompts asking AI to show 3M was 0% at fault for a fatal explosion.
  3. Discovery revealed 350 pages of prompts and testimony indicating 85-90% of the expert report was AI-generated output.
  4. Autenrieth defended his methodology by claiming AI served only as a straw man draft for his own verified professional opinions.
  5. The January 2020 Watson Grinding explosion killed three people and destroyed approximately 200 homes in Houston.

The story

A Harris County jury awarded $61 million to plaintiffs suing 3M over a fatal 2020 explosion after discovering the company’s expert witness generated most of his defense report using ChatGPT. Court records reviewed by 404 Media reveal expert Josh Autenrieth billed $475 hourly while prompting the AI to show 3M was zero percent at fault for the Watson Grinding blast that killed three people. Plaintiffs’ attorney Will Moye obtained 350 pages of prompts showing Autenrieth asked the chatbot to create an exceptional defense report, with trial testimony estimating 85 to 90 percent of the document was AI-generated. Autenrieth testified he used the tool only as a drafting aid and stood by his opinions based on twenty years of experience. The jury ultimately assigned 30 percent fault to 3M and 70 percent to Watson Grinding.

Who's involved

Critic
Will Moye

Plaintiffs' attorney argued the expert relied exclusively on ChatGPT to form opinions rather than using it merely as an assistive device.

Defender
Josh Autenrieth

Knighthawk Engineering expert testified he used AI only to draft a straw man and modified outputs based on his 20+ years of industry experience.

Neutral
Harris County Jury

Awarded $61 million to plaintiffs and assigned 30% fault to 3M despite the AI-generated defense report controversy.

How the conversation shifted

the split has narrowed

Polarity (0–100) from the noise pipeline, sampled over time.

Join the Discussion

Discuss this story

Community comments coming in a future update

Be the first to share your perspective. Subscribe to comment.

Noise Level

Murmur27?Noise Score (0–100): how loud a controversy is. Composite of reach, engagement, star power, cross-platform spread, polarity, duration, and industry impact — with 7-day decay.
Decay: 67%
Reach
38
Engagement
35
Star Power
15
Duration
100
Cross-Platform
20
Polarity
50
Industry Impact
50

The timeline

  1. Reddit post highlights 404 Media investigation

    User shares report detailing how discovery uncovered 350 pages of ChatGPT prompts used by 3M's expert witness.

  2. Jury delivers $61 million verdict against 3M

    Harris County jury assigns 30% fault to 3M after reviewing evidence of AI-generated expert testimony.

  3. Watson Grinding explosion occurs in Houston

    Industrial blast kills three people and destroys roughly 200 homes, triggering subsequent litigation against 3M.

The full record

Sources & methodology

Every claim above traces to these primary items. How we score →

The forecast

Courts will likely mandate explicit disclosure of AI assistance in expert reports because this verdict signals judicial skepticism toward undisclosed generative workflows in high-stakes litigation.

Forecast, not fact — an editorial estimate we score when this resolves.

You're up to date

That's the complete picture as of — nothing more to know right now. We'll update this page the moment it changes.