Papers and Patents Are Becoming Less Disruptive

Nature ^ | January 2023 | Michael Park, Erin Leahy, Russell Funk

[LS]

ABSTRACT

Theories of scientific and technological change view discovery and invention as endogenous processes1,2, wherein previous accumulated knowledge enables future progress by allowing researchers to, in Newton’s words, ‘stand on the shoulders of giants’3,4,5,6,7. Recent decades have witnessed exponential growth in the volume of new scientific and technological knowledge, thereby creating conditions that should be ripe for major advances8,9. Yet contrary to this view, studies suggest that progress is slowing in several major fields10,11. Here, we analyse these claims at scale across six decades, using data on 45 million papers and 3.9 million patents from six large-scale datasets, together with a new quantitative metric—the CD index12—that characterizes how papers and patents change networks of citations in science and technology. We find that papers and patents are increasingly less likely to break with the past in ways that push science and technology in new directions. This pattern holds universally across fields and is robust across multiple different citation- and text-based metrics1,13,14,15,16,17. Subsequently, we link this decline in disruptiveness to a narrowing in the use of previous knowledge, allowing us to reconcile the patterns we observe with the ‘shoulders of giants’ view. We find that the observed declines are unlikely to be driven by changes in the quality of published science, citation practices or field-specific factors. Overall, our results suggest that slowing rates of disruption may reflect a fundamental shift in the nature of science and technology.

[LS]

Gee, imagine that: as gubments do more and more funding---or influence more funding---they don't want results they don't want.

[Nateman]

The Soviet Union had a lot of dedicated scientists but not a lot of useful inventions. Little Israel is an invention powerhouse. If progress can actually be measured as this paper claims then the obvious reason for the decline is a shift towards socialism and away from God.

[ckilmer]

I’m hearing a lot of people say that AI will accelerate the rate of innovation.

Some of the charts in the paper suggest a bottoming of the rate of fall.

[spintreebob]

There is a difference in patents between breakthrus in truly new items and minor rearrangement to justify a new patent.

[glorgau]

> I’m hearing a lot of people say that AI will accelerate the rate of innovation.

The rub there is that the data used to build AI models and engines is pre-existing data.

[G Larry]

Excerpt from C.S. Lewis "Screwtape Proposes a Toast":

(The scene is in Hell at the annual dinner of the Tempters’ Training College for young devils. The principal, Dr. Slubgob, has just proposed the health of the guests. Screwtape, a very experienced devil, who is the guest of honour, rises to reply:)

"What I want to fix your attention on is the vast, overall movement towards the discrediting, and finally the elimination, of every kind of human excellence – moral, cultural, social, or intellectual. And is it not pretty to notice how “democracy” (in the incantatory sense) is now doing for us the work that was once done by the most ancient Dictatorships, and by the same methods?"

[ckilmer]

current iterations use data that is a year or three old. But later iterations will include current data.

ai will enable scientists to fluidly use different data sets from different disciplines

[mjp]

S-shaped growth curve(sigmoid growth curve)

[Organic Panic]

Who will waste time publishing actual science or inventions when they can get cash for publishing anything with “Climate change causes...” “Trannies hardest hit...” “Everything is racist....”

[LS]

Precisely.

[LS]

C.S. Lewis: Brilliance.

[LS]

AI is permanently limited by lack of human creativity.

[M Kehoe]

Bttt

5.56mm

[ckilmer]

Maybe.

I’m a pretty ordinary guy. I’ve found that I can do things that I was never able to do before.

On the boundaries of science are where some really interesting things will happen.

There are genius level guys and super genius level guys I’m sure that will be able to find that they can do things they were never able to do before.

I thought about this.

For example one of the things the article mentions is that specialization makes new discoveries difficult. people are siloed up in their disciplines. Whereas the real inventive stuff comes from crossing disciplines and mixing technologies to get something entirely different.

AI will enable genius level guys in a couple disciplines to create parallel databases from their disciplines and ask AI to find anomalies and symmetries between them. that is anomalies and symmetries where they should not exist.

the field of materials research and genetics will both have to explode as researches ask AI to create a material with such and such characteristics or find a genetic pattern that corresponds to a certain disease.

There are numerous fishing expeditions these guys can go on.

closer to home

There will come a time when entrepreneurs will be able ask AI to find something people really need; then develop,market and ship it.

[LS]

Yeah, but the problem often is that people don’t know they “need” something until they decide they need it.

It’s the old problem of advertising. Advertising rarely sells products to new customers-—it only reshuffles the deck among existing customers (Coke vs. Pepsi).

[ckilmer]

It’s the old problem of advertising. Advertising rarely sells products to new customers-—it only reshuffles the deck among existing customers (Coke vs. Pepsi)
......
agree. there will be plenty of misses. and the best/easiest to sell to/ customers are repeat customers. but new products for new customers do happen and not so rarely. and there are products that create their own categories—as well as category killers.