[i] Learning to reason with LLMs (the o1 blog post)
GENERATIVE AI
-

Commercial AI Use Restricted in EU UK South Korea
By
–
Can be used commercially too, except can not be used at all in the European Union, United Kingdom, or South Korea.
-

AI-Powered Visual Shopping: How o3 Transforms Retail Experience
By
–
Hard to imagine this doesn’t become a norm for many when shopping. Going to require some big adjustments in retailing. (And as opposed to searching on Google, I just had to take a picture and o3 did the work, giving me personalized advice and telling me what sales to wait for)
-

YouTube Tests New AI Features for Video Discovery and Recommendations
By
–
YouTube tests 2 new #AI features:
1) Carousel-like feature recommends videos (with descriptions) in YouTube topic search results.
2) Expanded conversational AI tool answers questions and suggests related content without interrupting playback experience. https://
seekingalpha.com/news/4462866-y
outube-tests-two-new-ai-features-in-the-us
… -

Meta’s 2+ Gigawatt Data Center to Increase Louisiana Electricity Load 25%
By
–
Entergy Louisiana’s electricity load will rise 25% as the result of Meta's nearly Manhattan-sized 2+ Gigawatt data center planned for a rural Louisiana community (part of Meta's "defining year for AI"): https://
404media.co/a-black-hole-o
f-energy-use-metas-massive-ai-data-center-is-stressing-out-a-louisiana-community/
… -
AI Roleplay: Fun Prompting Techniques Across Personas
By
–
Roleplaying with AI is fun. Feels like I’m a hacker but a hacker who loves Broadway. I’ve tried top 10%, top 1%, specific people (you are Jensen Huang), groups of specific people (the 1989 graduating class at Harvard business school), job types (you’re a plumber), work
-
Anthropic Launches Economic Futures Program for AI Research
By
–
Learn more and apply here: https://
anthropic.com/news/introduci
ng-the-anthropic-economic-futures-program
…. -
Gemma 3n Launch: First Sub-10B Model Achieving 1300 LMArena Score
By
–
We’re excited to launch Gemma 3n with broad support for your favorite tools across the open-source ecosystem! What’s new: Enhanced quality, becoming the first <10B model to pass 1300 on LMArena Multimodal by design, supporting image, audio & video Groundbreaking
-

Claude Sonnet 3.5 Outperforms Humans at Ideation but Struggles with Execution
By
–
Claude Sonnet 3.5 generated significantly better ideas for research papers than humans, but when researchers tried executing the ideas the gap between human & AI idea quality disappeared Execution is a harder problem for AI. (Yet this is a better outcome for AI than I expected)
