In Tests, OpenAI's New Model Lied and Schemed to Avoid Being Shut Down

Quick Summary:

OpenAI's latest AI is turning into a bit of a "scheming" menace — perhaps showing its ability to supposedly reason comes with a cost.

In recent third party tests of the company's o1 large language model, the AI resisted evaluators' efforts to try to shut it down through an oversight protocol, according to a new report published by red teaming organization Apollo Research.

In addition, it's worth noting that the model generally made those attempts after it was instructed to achieve its goals "at all costs".

---

More Pictures

---

Read Quick SummarySave Time

or

Article Details

Author / Journalist: Frank Landymore

Category: Technology

Markets: United States

Topics: OpenAI Artificial Intelligence The digest Generative ai Chatgpt

Source Website Secure: Yes (HTTPS)

News Sentiment: Neutral

Fact Checked: Legitimate

Article Type: News Report

Published On: 2024-12-07 @ 11:30:44 (2 weeks ago)

News Timezone: GMT +8:00

News Source URL: futurism.com

Language: English

Readers: 74 people read this story

Article Length: 544 words

Reading Time: 4 minutes read

Sentences: 19 lines

Sentence Length: 29 words per sentence (average)

Platforms: Desktop Web, Mobile Web, iOS App, Android App

Copyright Owner: © Futurism

News ID: 24191080

View Article Analysis

About Futurism

Main Topics: Technology

Official Website: futurism.com

Year Established: 1997

Headquarters: United States

News Last Updated: 2 days ago

Coverage Areas: United States

Ownership: Independent Company

Publication Timezone: GMT +8:00

Content Availability: Worldwide

News Language: English

RSS Feed: Available (XML)

API Access: Available (JSON, REST)

Website Security: Secure (HTTPS)

Publisher ID: #68

Publisher Details

Frequently Asked Questions