Clarification: From what I now understand, competitors can *try* FrontierMath, but they cannot access the full problem set and their solutions. OpenAI can, and this give them a large and unfair advantage. And very much raises the question of whether OpenAI trained on those
@garymarcus
-
Private Access to Problems Creates Unfair Competitive Advantage
By
–
the problem is that they funded private access to the problems that competitors don’t have. that’s very bad.
-
Intelligence Alone Cannot Guarantee Sound Reasoning
By
–
1. A lot of smart people are deluded about a great many things. IQ alone is no guarantee of sound reasoning. And I agree that thing that is most dangerous is the ability to lie to one’s self. That way lies madness. 2. Human nature sucks — and in fact I wrote a whole book
-
Data Contamination in AI Model Training Explained
By
–
what i mean is that the openai system may have been trained either directly on the problems in the dataset, or trained on augmented/synthetic data similar to the problems. that’s what contaminations means in this kind of ML context.
-
Influencer’s Failed AGI Prediction on Strawberry Model Criticized
By
–
From the account that told you last August that strawberry was going to be AGI. https://
x.com/iruletheworldm
/iruletheworldmo/status/1881049425251906013
… -
Critical thinking about AI: predictions coming true
By
–
𝘍𝘰𝘭𝘬𝘴, 𝘐 𝘩𝘢𝘷𝘦 𝘥𝘰𝘯𝘦 𝘦𝘷𝘦𝘳𝘺𝘵𝘩𝘪𝘯𝘨 𝘐 𝘤𝘰𝘶𝘭𝘥 𝘵𝘰 𝘵𝘦𝘢𝘤𝘩 𝘺𝘰𝘶 𝘩𝘰𝘸 𝘵𝘰 𝘵𝘩𝘪𝘯𝘬 𝘤𝘳𝘪𝘵𝘪𝘤𝘢𝘭𝘭𝘺 𝘢𝘣𝘰𝘶𝘵 𝘈𝘐, 𝘢𝘯𝘥 𝘢 𝘨𝘳𝘦𝘢𝘵 𝘥𝘦𝘢𝘭 𝘰𝘧 𝘸𝘩𝘢𝘵 𝘐 𝘩𝘢𝘷𝘦 𝘢𝘯𝘵𝘪𝘤𝘪𝘱𝘢𝘵𝘦𝘥 𝘩𝘢𝘴 𝘤𝘰𝘮𝘦 𝘵𝘰 𝘱𝘢𝘴𝘴. 𝘗𝘶𝘳𝘦
-
Neurosymbolic AI Development Overcomes Research Community Hostility
By
–
Completely messed up. I blame Hinton and LeCun for the counterproductive environment, likely delaying many key discoveries that could save lives by many years. Thank heavens @demishassabis didn’t let their unwarranted hostility towards neurosymbolic AI from letting him develop https://
t.co/bDmRMjnH3q -
Bridging Black Box AI and Explicit Knowledge Systems
By
–
as i know you would agree, we need a much better bridge between black boxes and explicit knowledge. we will see how long it takes. but i would agree that it is probably possible with current hardware.
-
Chain of Thought and RL in Extended AI Reasoning Analysis
By
–
the idea of analyzing things over a long derivation is about 60 years old, and will be part of the final answer; the particular way that o* does it (presumably via Chain of thought and RL but maybe other mechanisms) has not been transparently explained, and may or may not play a
-
Future AI Technology Evolution and System Integration Uncertainty
By
–
I wouldn’t at all assume that it will be. Maybe. Or if it is still there in recognizable form it might one module in a much more complex system.