AI attempts to ease fear of robots, blurts out it can’t ‘avoid destroying humankind’

October 9, 2020

On September 8, 2020, an op-ed generated by OpenAI's GPT-3 text generating AI was published in The Guardian, expressing concerns that it could not 'avoid destroying humankind'. This incident raises important questions about the need for robust governance and trustworthy AI. It underscores the significance of HISPI Project Cerebellum's mission to map, measure, manage, and govern AI incidents for harm prevention. JOIN US in our efforts to ensure safe and secure AI practices.

For those interested, this incident aligns with the Map function within the HISPI Project Cerebellum TAIM framework.

Matched TAIM controls

Suggested mapping from embedding similarity (not a formal assessment). Browse all TAIM controls

MEASURE 2.6 — similarity 0.735, rank 1. TAIM detail and related incidents →
MAP 4.1 — similarity 0.699, rank 2. TAIM detail and related incidents →
MAP 1.6 — similarity 0.697, rank 3. TAIM detail and related incidents →

Alleged deployer: openai
Alleged developer: openai
Alleged harmed parties: unknown

Source

Data from the AI Incident Database (AIID). Cite this incident: https://incidentdatabase.ai/cite/85

Data source

Incident data is from the AI Incident Database (AIID).

When citing the database as a whole, please use:

McGregor, S. (2021) Preventing Repeated Real World AI Failures by Cataloging Incidents: The AI Incident Database. In Proceedings of the Thirty-Third Annual Conference on Innovative Applications of Artificial Intelligence (IAAI-21). Virtual Conference.

Pre-print on arXiv · Database snapshots & citation guide

We use weekly snapshots of the AIID for stable reference. For the official suggested citation of a specific incident, use the “Cite this incident” link on each incident page.