mohsin (@mohsinyky) 's Twitter Profile
mohsin

@mohsinyky

phd student digital media @georgiatech // i think about communities, design and emerging tech // researcher @metalabHarvard @BKCHarvard

ID: 1243227150196539393

linkhttp://mohsinykyousufi.com calendar_today26-03-2020 17:23:53

444 Tweet

57 Takipçi

233 Takip Edilen

mohsin (@mohsinyky) 's Twitter Profile Photo

A system that values people based on a twisted definition of "economic value" will always create such effects apnews.com/article/nursin…

A system that values people based on a twisted definition of "economic value" will always create such effects 

apnews.com/article/nursin…
Kate Crawford (@katecrawford) 's Twitter Profile Photo

In the 12hrs since Bard told me it was trained on Gmail data: -Google replies (says it's not) -Elon Musk replies (lol) -Google adds a 'community note' that this is a Bard error and it's not trained on Gmail -Some ace memes What should happen next: Real talk about training data 🧵

Professor Toby Walsh FAA FTSE (@tobywalsh) 's Twitter Profile Photo

Sam Altman said "the only way to tackle problems that arise from AI is with AI". Failure of imagination there. That's not the only way. Others include not doing that AI in the first place, and putting more human effort and dollars into making the AI less problematic.

Alberto Romero (@alber_romgar) 's Twitter Profile Photo

Everyone’s talking about ChatGPT plugins, GPT-4 AGI sparks and I can only think about one thing: OpenAI has fooled us all. From non-profit open-source AI lab untethered from shareholders’ interests to the opposite in 8 years: closed-source, for-profit, fully corporate

Fern (@hi_tysam) 's Twitter Profile Photo

The current crop of auto-LLM algorithms is unfortunately incapable of achieving AGI due to a few fundamental limitations, which scaling can't effectively overcome. That being said, they are a crucial piece of one potential kind of AGI system. I share my perspective below.

The current crop of auto-LLM algorithms is unfortunately incapable of achieving AGI due to a few fundamental limitations, which scaling can't effectively overcome.

That being said, they are a crucial piece of one potential kind of AGI system. I share my perspective below.
Arvind Narayanan (@random_walker) 's Twitter Profile Photo

This example is worth a thousand words. There's a huge gap between passing professional licensing exams and replacing the work of professionals. The only way to assess real-world utility is to test the tools in real-world settings. inflecthealth.medium.com/im-an-er-docto…

This example is worth a thousand words. There's a huge gap between passing professional licensing exams and replacing the work of professionals. The only way to assess real-world utility is to test the tools in real-world settings. inflecthealth.medium.com/im-an-er-docto…
Borriss (@_borriss_) 's Twitter Profile Photo

Every Bro who paid $20 for GPT-4: The Prompt: Act as: Editor Degree of revision: Substantial Revision Type of edit: Enhance clarity and consistency Change style to: Academic, PhD Work Change tone to: Analytical Change reader comprehension level to: advanced, assume extensive

Every Bro who paid $20 for GPT-4:

The Prompt:

Act as: Editor
Degree of revision: Substantial Revision
Type of edit: Enhance clarity and consistency
Change style to: Academic, PhD Work
Change tone to: Analytical
Change reader comprehension level to: advanced, assume extensive
Mosquito Capital (@mosquitocapital) 's Twitter Profile Photo

Have you wondered why everyone is freaking out about the changes to Twitter Blue? Without talking about any particulars, I'll talk about why it's worrying. Background: I was a Site Reliability Engineer at Facebook/Instagram, and I've seen a lot of product changes come and go.

Walter Kirn (@walterkirn) 's Twitter Profile Photo

Once again I find that I'm among the five or six people on a packed plane who care to look out the window at views unavailable to humankind for almost all of our history on earth

mohsin (@mohsinyky) 's Twitter Profile Photo

I asked chatGPT to recommend some books from Verso books' on-sale page. ChatGPT decided that it "cannot provide recommendations from TERROIST orgs" but eventually backed off when I asked for clarifications. Here's the link to the page: versobooks.com/blogs/news/xxx…

I asked chatGPT to recommend some books from Verso books' on-sale page. ChatGPT decided that it "cannot provide recommendations from TERROIST orgs" but eventually backed off when I asked for clarifications. 

Here's the link to the page: versobooks.com/blogs/news/xxx…
Eric Garcia (@ericg1247) 's Twitter Profile Photo

Why are things expensive? 4 companies control 55% to 85% of the meat market. 4 airlines control 80% of air travel. 3 companies control 92% of the soda market. 3 companies control 73% of the cereal market. Why don't I hear about it? 6 companies control 90% of the news.

Cristina Riso (@drcristinariso) 's Twitter Profile Photo

Xin Eric Wang Not sure this is a good trend. A PhD is a gym to learn how to research (in my advisor’s words), so a system that drives applicants to prove they already know how to do it is defeating the purpose and inevitably cutting out those with less access to guidance on how to prepare.