Artificial Intelligence

OpenAI's GPT-6.1 Terra is safer overall, but shows more destructive-action flags in tests

Headline and link to the original report by BleepingComputerHow we report

BleepingComputer

Interestana links to this report without summarising it, because BleepingComputer does not allow reuse of its reporting.

Read on BleepingComputer (opens in a new tab)

Ref 00R8BBUD

Follow this story

Interests
Artificial Intelligence

Get Artificial Intelligence in your inbox — new stories in 40 words or fewer, once a week.

Subscribe

Follow Interestana

Read next

BleepingComputer

Northflank security breach leads to abuse of customers' AI API keys

BleepingComputer

New Android SMS of Death exploit can remotely crash unpatched phones

BleepingComputer

From Beginner to Bug Hunter: Learning Cybersecurity in the AI Era.

BleepingComputer

Chainalysis discloses data breach from "offensive agentic techniques"