Sampsa Samila

68 posts

Sampsa Samila banner
Sampsa Samila

Sampsa Samila

@SampsaSamila

Prof at IESE Business School. Artificial intelligence, innovation, entrepreneurship, networks.

Barcelona, Spain Katılım Ağustos 2015
223 Takip Edilen357 Takipçiler
Sampsa Samila
Sampsa Samila@SampsaSamila·
@Afinetheorem @lugaricano Doesn't that prove François's point? China's biotech leap came from removing regulatory bottlenecks, not adding brains.
English
0
0
0
5
Kevin A. Bryan
Kevin A. Bryan@Afinetheorem·
@lugaricano I do not think this is consistent with available evidence from econ of innovation. E.g., China entering biotech development by "adding a billion brains" does appear to come up with a lot of new ideas and recombinations! The exact parameter here is empirical, of course.
English
1
0
0
212
Luis Garicano 🇪🇺🇺🇦
Luis Garicano 🇪🇺🇺🇦@lugaricano·
Fascinating hypothesis.
François Chollet@fchollet

Basically, consider a high-profile scientific problem (one that is getting enough attention from smart humans and their externalized cognitive infrastructure). How good is human Science at converting the information available about the problem into generalizable models of the phenomenon at hand? I'd argue it is very good, already near-optimal. Our inability to solve certain hard problems stems more from lack of information than low ability to operationalize available information. If the solution is already "within" the information we have, it's only a matter of time until Science figures it out. Also do note that this is inclusive of current AI, because current AI is an externalized cognitive tool leveraged by humans. "AI for science" is making us smarter as a collective.

English
2
2
14
7.4K
Kevin A. Bryan
Kevin A. Bryan@Afinetheorem·
"Beautiful classic watch set to 3:41". The watch is great (only minor error was IIII instead of IV), but so many watches in training data from ads set to 10:10 that we can't push the posterior of the model off this with a prompt.
Kevin A. Bryan tweet media
English
2
0
5
309
Kevin A. Bryan
Kevin A. Bryan@Afinetheorem·
"Nano Banana" aka Gemini 2.5 Flash Image still shows your best intuition for what's hard for transformers is not "things outside the training set" but the opposite: "things where epsilon different is all over the training set". Two classic examples:
English
1
1
15
3.1K
Sampsa Samila
Sampsa Samila@SampsaSamila·
Ultimately, LLMs reorganize work around the Generality-Accuracy frontier. Our paper offers a framework for leaders to manage these trade-offs. Read it on SSRN: papers.ssrn.com/sol3/papers.cf…
English
0
1
2
80
Sampsa Samila
Sampsa Samila@SampsaSamila·
This shifts the value of skills. As AI handles routine work, conceptual reasoning and critical judgment as well as domain expertise in areas where I fall short become the scarce human complements. #FutureOfWork
English
1
0
1
88
Sampsa Samila
Sampsa Samila@SampsaSamila·
New paper! We introduce the Generality-Accuracy-Simplicity (GAS) framework to analyze how LLMs are reshaping organizations and strategy. #AI #Strategy #OrgDesign
English
1
1
2
149
Douglas Carswell🇬🇧🇺🇸
Douglas Carswell🇬🇧🇺🇸@DouglasCarswell·
Home Depot is worth more than all the European start ups created in the past 50 years combined. Home Depot.
Douglas Carswell🇬🇧🇺🇸 tweet media
English
1K
4K
45.4K
3.4M
Amanda Sharkey
Amanda Sharkey@amandajsharkey·
Today I joined the @WPCareySchool at @ASU as an associate professor. The energy, momentum, and sense of purpose here is palpable. Excited for this next chapter!
English
9
0
49
0
Sampsa Samila retweetledi
Franz Heukamp
Franz Heukamp@FranzHeukamp·
Congratulations to @SampsaSamila, Chris Zott, Jordi Canals and Adrian Caldart for winning the Alumni Association’s Research Excellence Awards for their ideas on gender and innovation, business model innovation and a company's strategic shift. bit.ly/ieseawards
English
1
2
10
0

Keşfet

@Afinetheorem @lugaricano @kelvinltse @DouglasCarswell @iesebschool @Matera_Michal @AllenOvery @PGdanski