If you are here because of the @nytimes tweet reference on ChatGPT writing of scientific papers, you might find this 2.5 year old column interesting.

Wow. it must be HER after she left Twombly. .

The traditional Fourth of July barbecue. #Tempe

. and this .

🎉🥳Congratulations to Yochanite Atharva Gundawar (@_gundawar) who successfully defended his MS thesis on "Scaling LLMs with LLM-Modulo" today! He is off to The AGI Company (really!).

(It was fun chatting with Ms. Gina Kolata about the increasing role of LLMs in Science and AI--both in paper writing, and more alarmingly--reviewing!). Over the years I have read countless news articles by her on medical breakthroughs--large and small. 🤗) . Article Link:.

Woo hoo. #AIAphorisms breaks into @nytimes . 🤓

Hot take: The burgeoning cottage industry that insists on interpreting length of intermediate tokens output by LRMs as their "Thinking Effort" -- and churn out papers on under thinking/over thinking/efficient thinking etc. may be the new cargo cult of LRM reasoning. (c.f.

Link to the article .

Enjoyed talking to Troy Wolverton of @sfchronicle about our group's work on the planning/reasoning capabilities of LRMs.

SuperiorIntelligence. .

#Tempe is Seattle South atm.

What exactly is the alpha in suiting up to make yet another proprietary frontier model with the same architecture as everyone else? 🤔.

With the armies of beneficial and evil AI aligned neatly on each side and poised on the cusp of their epic battle, Lord Sama took a break with warrior Noam, and proceeded to teach him about the Meta unreality--aka Maya--of AGI.

RT @JadeCole2112: Sounds exactly like what a company on the cusp of AGI would do.

No Virginia, you don't need a PhD to help build Taj Mahal or even oversee its construction--but you probably can benefit from (the training of) one to start wondering/understanding what makes it last. .