Skip to content
Submit incident
AII-1002-46957
Well documentedThree or more independent publishers report the core facts.

Otter AI transcript included confidential conversations after meeting ended

Otter AI's transcription service automatically sent a meeting transcript to researcher Alex Bilzerian that included hours of private conversations between venture capitalists that continued after the official meeting ended, exposing confidential business information he was not intended to receive.

26 September 2024 Β· Otter AI Β· Privacy and surveillance
26 September 2024
AII-1002-469573 sources ↓
Quick facts
Category
Privacy and surveillance
Country
United States
Source
Washington Post
Date
26 September 2024
Drafted by
Raidu Incident Detection

What happened

Otter AI, a transcription service with more than 14 million users, automatically emailed a meeting transcript to researcher Alex Bilzerian after a Zoom call with a venture capital firm ended.

The transcript included hours of private conversations between the venture capitalists that occurred after Bilzerian had logged off, containing what he described as intimate, confidential details about their business.

Reported impact

Bilzerian received confidential business information he was not meant to see from the venture capital firm.

Bilzerian decided to end the business deal with the firm as a result of the incident.

Index analysis

Otter AI stated that users have full control over transcript sharing permissions and can change, update, or stop sharing settings at any time.

An associate professor at Northwestern University noted that the reasonable assumption that AI assistants would detect when participants exit a meeting is not always accurate with current technology.

Governance controls

Otter AI said users have the option not to share transcripts automatically with anyone, or to auto-share conversations only with users who share the same Workspace domain.

Classification

Category
Privacy and surveillance
Industry
IT infrastructure and hosting
Country
United States
Harm status
Actual harm
Intent
Unintentional
Timing
Post-deployment
Harm types
Economic or property
Harmed
Business
Failure modes
Data leak
AI systems
Otter AI

Sources and evidence

This record was researched and written by the Index, based on the following original reporting.

Washington Post
AI assistants are blabbing our embarrassing work secrets
2 October 2024
Entrepreneur
Is AI Accidentally Spilling Your Company's Secrets? A VC Firm's Private Conversations Were Included in Public Meeting Transcripts.
2 October 2024
New York Post
AI is spying on your workplace gossip and secrets -- and sharing...
4 October 2024