About Database Apply for funding For AI safety funders

We're building a public coordination layer for AI safety funding — so donors can move faster, grantees can stay visible, and the ecosystem can share more of its knowledge.

Questions? Takes? Suggestions?

Email us, we want to hear from you

hi@grantmaking.ai

© 2026 This site is released under a CC BY-SA license

Marius Hobbhahn → Demonstration of LLMs deceiving and getting out of a sandbox

Amount: $3,000Date Awarded: Jun 18, 2025

Details

Funder: Marius Hobbhahn
Recipient: -
Purpose: -
Source: https://manifund.org/projects/salient-demonstration-of-llms-deceiving-and-getting-out-of-a-sandbox

Metadata

Created: Apr 8, 2026, 5:20 PM UTC

Marius Hobbhahn → Demonstration of LLMs deceiving and getting out of a sandbox | grantmaking.ai