AI assistants are trained to give responses that humans like. Our new paper shows that these systems frequently produce ‘sycophantic’ responses that appeal to users but are inaccurate. Our analysis suggests human feedback contributes to this behavior.
ETHICS
-
AI Safety Summit in London: Networking Opportunity
By
–
Who will be in London for the AI Safety Summit? DM if you'd like to meet and chat.
-
Algorithm Reversibility: Control in Autonomous Decision-Making Systems
By
–
Is the decision your algorithm makes reversible? e.g. light switch vs potentially lethal autonomous car. This can impact how much control a user feels they have. Insights from this great talk on causality, from a philosopher of neuroscience.
-
Uncensored AI, essential fight of the next 10 years
By
–
In 10 years, we will look back and realize we should have been fighting for uncensored AI this whole time.
-
Democracy Essential for Beneficial Technological Progress
By
–
That's why we have democracy.
If you don't believe that democracy is effective, you may have doubts that technological progress will be for the greater good.
Non democratic societies have bigger problems than technology. -
Technology’s Impact on Human Evolution and Biological Transformation
By
–
1. I didn't write what you are quoting me as having written.
2. Humans from 40,000 years ago would probably not see us as having the same form as them.
3. Technology has already transformed humans and will continue to do so (e.g. our teeth are what they are because we have knives -
AI Consciousness Conundrum: Exploring Machine Minds
By
–
Minds of machines: The great #AI consciousness conundrum https://
news.google.com/articles/CBMiT
2h0dHBzOi8vd3d3LnRlY2hub2xvZ3lyZXZpZXcuY29tLzIwMjMvMTAvMTYvMTA4MTE0OS9haS1jb25zY2lvdXNuZXNzLWNvbnVuZHJ1bS_SAVNodHRwczovL3d3dy50ZWNobm9sb2d5cmV2aWV3LmNvbS8yMDIzLzEwLzE2LzEwODExNDkvYWktY29uc2Npb3VzbmVzcy1jb251bmRydW0vYW1wLw?hl=en-US&gl=US&ceid=USen
… -
Techno-Optimism: Humanism and Active Technology Optimism
By
–
Excellent piece by @Noahpinion on techno-optimism.
So many good points. Like Noah, I'm a humanist who subscribes to both Positive and Normative forms of Active Techno-Optimism. Quote: "Techno-optimism is thus much more than an argument about the institutions of today or the -
Human cloning test with AI and AI news on Twitch
By
–
We test new technologies to 100% clone humans with AI. And we review AI news. See you soon on Twitch!
-
Tech Company Makes AI Look Scarier Than Necessary
By
–
They really made this thing look as frightening as possible