Suositut postaukset

tiistai 29. marraskuuta 2022

Suppe vastaa koevastauksiin

GPT-3 is already capable of answering questions of the sort used in essay topics and take home exams. It does not answer them well, but it answers them well enough to get a passing grade in many college courses and most high school courses. It is likely that many students have already started using it for this. It costs about 6 cents per thousand words, and anyone can get access to it within five minutes. I cannot emphasize enough that this is not “sometime vaguely in the next five years”, nor is it “accessible only to students with a background in comp sci”. It’s a 6 cents per thousand words plagiarism service available to everyone right now. Check it out if you don’t believe me:

https://philosophybear.substack.com/p/gpt-3-is-right-now-already-more-than 

Suppe vastailee koevastauksiin. Minun mielestä vastaukset ihan hyviä, en olisi itse ilman asiaan kunnolla perehtymistä pystynyt sen paremmin vastaamaan, luultavasti. Kirjoittaja turhaan halveeraa tekoälyn kykyjä koe vastauksissa. Ilmeisesti amerikkalaisoppilaat ovat kauhean taitavia ja älykkäitä. Vai vertaako kirjoittaja paremman pään yliopistojen opiskelijoihin. Siellä on niitä älykkäitä ihmisiä paljon. Suomalainen oppilas ei pärjää. 

Kyllähän amerikan kokoisessa maassa on taitavia ja älykkäitä ihmisiä paljon. No ei kannata verrata niihin, onhan siellä tyhmiä sielläkin paljon. Eivät aina osaa edes omaa äidinkieltään kirjoittaa kunnolla, suomalaiset sentään yleensä osaavat, monet osaavat vielä useampaa kieltäkin.

So why not build a computer that way? In 1958, Frank Rosenblatt pulled off a proof of concept: a simple model based on a simplified brain, which he trained to recognize patterns. “It would be possible to build brains that could reproduce themselves on an assembly line and which would be conscious of their existence,” he argued. Rosenblatt wasn’t wrong, but he was too far ahead of his time. Computers weren’t powerful enough, and data wasn’t abundant enough, to make the approach viable.

 This technique — now called deep learning — started significantly outperforming other approaches to computer vision, language, translation, prediction, generation, and countless other issues. The shift was about as subtle as the asteroid that wiped out the dinosaurs, as neural network-based AI systems smashed every other competing technique on everything from computer vision to translation to chess.

“If you want to get the best results on many hard problems, you must use deep learning,” Ilya Sutskever — cofounder of OpenAI, which produced the text-generating model GPT-3 and the image-generator DALLE-2, among others — told me in 2019. The reason is that systems designed this way generalize, meaning they can do things outside what they were trained to do. They’re also highly competent, beating other approaches in terms of performance based on the benchmarks machine learning (ML) researchers use to evaluate new systems. And, he added, “they’re scalable.”

https://beta.openai.com/examples

Kyllähän nämä vastaukset ovat yllättävän hyviä. En itse pystyisi vastaavaan. Suppe on kehittymässä kovaa. En tiedä miksi ihmiset eivät arvosta näitä vastauksia. Nehän hirvittävän hyviä, kun ajattelee kuinka vähän tällaisia massiivisia neuroverkkoja on ollut olemassa. 

perjantai 11. marraskuuta 2022

AI elokuvat tulossa

Tekoälyn tekemät elokuvat tulossa. Iso muutos elokuvamarkkinoilla. Pientuottajatkin kykenevät tulevaisuudessa tekemään hienoja elokuvia tekoälyn avulla.

For years, the only way to create a blockbuster film featuring a Hollywood star and dazzling special effects was at a major studio. The Hollywood giants were the ones that could afford to pay celebrities millions of dollars and license sophisticated software to produce elaborate, special effects-laden films. That’s all about to change, and the public is getting a preview thanks to artificial intelligence (AI) tools like OpenAI’s DALL-E and Midjourney.

Both tools use images scraped from the internet and select datasets like LAION to train their AI models to reconstruct similar yet wholly original imagery using text prompts. The AI images, which vary from photographic realism to mimicking the styles of famous artists, can be generated in as little as 20 to 30 seconds, often producing results that would take a human hours to produce. For example, the “detailed glowing synthwave diagram of Japanese mech robot” illustration below.

https://qz.com/artificial-intelligence-means-anyone-can-cast-hollywood-1849695488