unintended; going to take us some time to get all of this right (and it still requires more research). generally speaking, within very wide bounds we want to enable people get the behavior they want when using AI. will talk more about it in january!
ETHICS
-
Current LLM Architecture Safety and Future Advancement Risks
By
–
I *guess* that the pure, unmodified Stack More Layers architecture used in *current* LLMs is stupid enough to be safe. Later advancements might break that capability barrier; the results would also be called 'LLMs'. The *task* of text prediction is general, and not safe.
-
Maintaining Regulatory Compliance with AI Implementation
By
–
How to maintain regulatory compliance with #AI in use?
By @ingliguori http://
bit.ly/3bfsSmH
#Healthtech #Health #DataScience #Bigdata #Analytics #Robotics #MachineLearning #Python #GDPR #Fintech #IoT #TensorFlow #MEDTECH #Programming #5G #VR #Innovation #privacy #Da… -
Contrast between fictional and actual artificial intelligence capabilities
By
–
Sci-fi AI: cold, rational, infinitely powerful
Real AI: randomly making stuff up -
Clarifying AI Text Prediction Beyond Simple Pattern Matching
By
–
Somebody needs to explain this to the people who are like "It's safe to predict human text – all the AI is doing is predicting what humans say, not [various aspects of thinking that humans do while they themselves decide what to say, which is what the AI is predicting]."
-
Opinion Polarization in AI Art Debate and Copyright Concerns
By
–
I think the worst part is that this opinion seems to be the root cause of the polarization in the AI Art debate. If he came out and said the same thing as for their music models ("they are infringing so we're being careful") then the debate would become reasonable in comparison.
-
ChatGPT Doxxing Concerns: Privacy Violations and Platform Rules
By
–
.
@stefanfis
's right, #ChatGPT could get banned for this! Giving the exact position of a person is against @Twitter
's rules! #doxxing -
Claim of ChatGPT political shift criticized for weak evidence
By
–
There's a popular thread claiming ChatGPT's political opinions shifted centrist after its Dec. 15 update. Evidence given for this is weak (4 online political tests given only once in each period), and the prompting method inappropriately forces in-test consistency of answers:
-
Galactica less popular, more hallucinatory than ChatGPT
By
–
Less, I think. Galactica didn't land with the consumer-facing splash that ChatGPT did. I only used it briefly, but it was also more hallucinatory, e.g. never refusing absurd requests. I think hallucination can be worse when infrequent, because we trust the model more.
-
AGI Debate Recording and Discussion
By
–
AGI Debate Recording : https://
youtu.be/JGiLz_Jx9uI #AGI #AGIDebate