Comment by foo12bar 2 days ago Because they are testing it and are expected to erect guardrails before releasing. 2 comments foo12bar Reply no-name-here 2 days ago What are the specific guardrails implemented after the verification/testing phase of development?Is it safe to release such software if it has only been tested in environments where certain major risk areas do not exist? Maxious 2 days ago https://openai.com/index/updating-our-preparedness-framework... goes through the process. Anthropic has a similar framework, that's why Mythos was never publically accessible once initial tests like the one in OP revealed it's capabilities.Appendix C Illustrative safeguards, controls, and efficacy assessments has specific examples like:- Agent actions are all logged in an uneditable database, and asynchronous monitoring routines review those actions for evidence of harm- Limiting internet access and other tool access- Limiting credentials- Limiting access to system resources or filesystem (e.g., sandboxing)- Limiting persistence or state
no-name-here 2 days ago What are the specific guardrails implemented after the verification/testing phase of development?Is it safe to release such software if it has only been tested in environments where certain major risk areas do not exist? Maxious 2 days ago https://openai.com/index/updating-our-preparedness-framework... goes through the process. Anthropic has a similar framework, that's why Mythos was never publically accessible once initial tests like the one in OP revealed it's capabilities.Appendix C Illustrative safeguards, controls, and efficacy assessments has specific examples like:- Agent actions are all logged in an uneditable database, and asynchronous monitoring routines review those actions for evidence of harm- Limiting internet access and other tool access- Limiting credentials- Limiting access to system resources or filesystem (e.g., sandboxing)- Limiting persistence or state
Maxious 2 days ago https://openai.com/index/updating-our-preparedness-framework... goes through the process. Anthropic has a similar framework, that's why Mythos was never publically accessible once initial tests like the one in OP revealed it's capabilities.Appendix C Illustrative safeguards, controls, and efficacy assessments has specific examples like:- Agent actions are all logged in an uneditable database, and asynchronous monitoring routines review those actions for evidence of harm- Limiting internet access and other tool access- Limiting credentials- Limiting access to system resources or filesystem (e.g., sandboxing)- Limiting persistence or state
What are the specific guardrails implemented after the verification/testing phase of development?
Is it safe to release such software if it has only been tested in environments where certain major risk areas do not exist?
https://openai.com/index/updating-our-preparedness-framework... goes through the process. Anthropic has a similar framework, that's why Mythos was never publically accessible once initial tests like the one in OP revealed it's capabilities.
Appendix C Illustrative safeguards, controls, and efficacy assessments has specific examples like:
- Agent actions are all logged in an uneditable database, and asynchronous monitoring routines review those actions for evidence of harm
- Limiting internet access and other tool access
- Limiting credentials
- Limiting access to system resources or filesystem (e.g., sandboxing)
- Limiting persistence or state