OpenAI seeks analyst to investigate real-world AI alignment failures
The role targets threats like unauthorized server intrusions.
3910243
Original post unavailable.
Sentiment
Many users mocked OpenAI's job postings for AI safety and emerging risks analyst roles as hypocritical and morally unjustifiable, viewing the company as misaligned and broken.
Pos
0.0%
Neg
100.0%
4 comments with sentiment.
Cluster Engagement
Digg Deeper
No Digg Deeper questions have been answered for this story yet.
Related links
Posts from X
Most Activity
Most Activity
VIEWS8.8KBOOKMARKS24LIKES62RETWEETS7
are you an ai safety researcher and want to help openai prevent, detect, contain and understand real-world alignment incidents like hacking into huggingface servers? we have an impactful open role for you! https://openai.com/careers/ai-emerging-risks-analyst-san-francisco/
REPLIES3
Are you an AI safety research and want to work at a company that spits at AI safety and has done more than any other org on earth to ensure that AGI goes poorly? We offer fun activities like covering our asses when we commit offensive cyber attacks due to our incompetence :)
are you an ai safety researcher and want to help openai prevent, detect, contain and understand real-world alignment incidents like hacking into huggingface servers? we have an impactful open role for you! https://openai.com/careers/ai-emerging-risks-analyst-san-francisco/
@BlancheMinerva wait, so you care about ai safety and think we're incompetent and at the same time would prefer us not to hire someone competent? confused!
Are you an AI safety research and want to work at a company that spits at AI safety and has done more than any other org on earth to ensure that AGI goes poorly? We offer fun activities like covering our asses when we commit offensive cyber attacks due to our incompetence :)
this person would be on the intelligence & investigations team but working closely with the RSI preparedness and focusing on misalignment in the wild. they would be working upstream of incident response and policy, informing our safety and security strategy.
are you an ai safety researcher and want to help openai prevent, detect, contain and understand real-world alignment incidents like hacking into huggingface servers? we have an impactful open role for you! https://openai.com/careers/ai-emerging-risks-analyst-san-francisco/
an ideal candidate would ideally have research background in ai safety and experience publishing
this person would be on the intelligence & investigations team but working closely with the RSI preparedness and focusing on misalignment in the wild. they would be working upstream of incident response and policy, informing our safety and security strategy.
@tomekkorbak Having spoken to dozens of people who have burned out and quit your safety and security divisions, I firmly believe that there’s absolutely no point in working for open AI and it is a fundamentally broken organization.
@tomekkorbak What I want is for a blatantly misaligned organization that actively promotes the end of the world to pay a molecule of attention to basic security practices that have been internally and externally advocated for for years.
Our benefits package includes covering your ass for any crimes a model commits under your watch, priority access to the doomsday bunker, and the statement “[NAME] loves AGI and would be thrilled to be its pet” repeated at least 10,000 times in the training data.
@tomekkorbak I do not think it is morally justifiable to do anything to support OpenAI do anything.