AI Safety efforts need to focus on solutions sooner rather than later. Greater online privacy, control, experimental transparency, and use of trust signals will improve the safe use of AI systems.
Hi Ravi! Thank you for this incredibly insightful piece. The problem-space feels so big that it's overwhelming (for me) to know which solutions to prototype and test. I currently lead design for Moderator Tools and Trust and Safety Tools at The Wikimedia Foundation (i.e. for Wikipedia) so I'm hoping to more intentionally define the hypotheses we're testing when we release new tools, just so that we're tracking what we learn and can chip away at the problems.
I learned a lot via the hyperlinks you've included as well and will keep digging :) Thanks again!
Hi Ravi! Thank you for this incredibly insightful piece. The problem-space feels so big that it's overwhelming (for me) to know which solutions to prototype and test. I currently lead design for Moderator Tools and Trust and Safety Tools at The Wikimedia Foundation (i.e. for Wikipedia) so I'm hoping to more intentionally define the hypotheses we're testing when we release new tools, just so that we're tracking what we learn and can chip away at the problems.
I learned a lot via the hyperlinks you've included as well and will keep digging :) Thanks again!