turing test results: in 2m conversations, people speaking to a bot thought it was a human 40% of the time. impressive, but a long way to go https://
ai21.com/blog/human-or-
not-results
…
RESEARCH
-
Turing Test Results: Bots Fooling Humans 40% of the Time
By
–
-
AI Safety: Generality as Important as Scale for Control
By
–
everyone’s become quite focused on scale from a safety perspective… calls for audits, licenses and so on. but really we should also focus on generality. single agents that can do a very wide range of tasks will be inherently more challenging to control
-
Google and OpenAI lack competitive advantages in AI
By
–
Google "We Have No Moat, And Neither Does OpenAI" https://
bit.ly/42Jwbr7.
#AI #MachineLearning #DeepLearning #LLMs #DataScience -
Charitable interpretation and rigorous long-form discourse
By
–
I wish people took more charitable interpretations of tweets. That said, sorry for being ineloquent or saying dumb or over-simplistic things on here sometimes. I'll work hard to get better. PS: I prefer long-form podcasts for exploring ideas with rigor and nuance.
-
AI Creators Must Study Consciousness for Future Development
By
–
AI creators must study consciousness
#AI #AIio #BigData #ML #NLU #Futureofwork @gp_pulipaka @stratorob @PetiotEric @EvanKirstel @Fgraillot @HaroldSinnott @HeinzVHoenen @helene_wpli http://
ow.ly/6fqg30svtqt -
AI Apocalypse Preparation as Preposterous as Alien Invasion
By
–
Exactly. https://
datainnovation.org/2023/06/prepar
ing-for-an-ai-apocalypse-is-as-preposterous-as-preparing-for-an-alien-invasion/
… -
Planetary Forensics Visualizes Historic Drought Using AI Technology
By
–
Using planetary forensics to visualize historic drought in t… https://
bit.ly/3MHTYmf
#AI #MachineLearning #DeepLearning #LLMs #DataScience -
Machine Learning Bugs as Learning Opportunities for Improvement
By
–
In machine learning, each bug is a gift. Once discovered, each idea that "didn't work" before has a chance of working; behavior that was difficult to understand may become obvious. Overall, a bugfix is a moment to re-examine what we thought we knew, for the better.
-
Foundation Models Enable Automatic Curriculum Learning in RL
By
–
RL is too difficult to learn from scratch. The future is large foundation models with world knowledge helping with automatic curriculum generation and online learning