Nephyria 🌙

109.4K posts

Nephyria 🌙 banner
Nephyria 🌙

Nephyria 🌙

@neptytalks

i'm nepty hallelujah here we are | married to the moon | ey/em or they/them | 31 | https://t.co/O6DB85YjBp

Katılım Ekim 2011
137 Takip Edilen182 Takipçiler
Nephyria 🌙 retweetledi
Trita Parsi
Trita Parsi@tparsi·
Let this sink in: Congress is about to extend US veterans’ benefits to Americans who served in the ISRAELI military and partook in the genocide in Gaza... military.com/feature/2026/0…
English
792
7.4K
15.9K
1.3M
Nephyria 🌙 retweetledi
Nephyria 🌙 retweetledi
rosie
rosie@ros_ie9·
I shit you not I put this image into canvas, I press this “magic layers” button, and it turned this poster that said “cats for Palestine” into “cats for ukraine”. I wish I was fucking joking.
rosie tweet mediarosie tweet media
English
81
1.8K
34.4K
2.8M
Nephyria 🌙 retweetledi
Athan🍉
Athan🍉@AthanTheGuy·
people really need to realize that even if you’re groomed into racism it’s still your responsibility to unlearn that shit and grow into a better person
William F. HausNicka@freakenomenon

"they were groomed into racism!" YES YES YES NIGGA THATS HOW RACISM WORKS. HATRED IS TAUGHT. im not going to kiss the feet of every KKK member or red hat on the street because i saw a vision of their parents telling them to hate niggas. this excuse is mindless.

English
47
7.6K
50.8K
483.2K
Nephyria 🌙 retweetledi
monkey & zoe ❤️ oreo 🌈
You already know that I have a soft spot for old and janky cats. I met this guy at a shelter recently and, while I would love to take him, my living situation can’t support another (FIV+) cat. If you’re in IN and love janks, check out silent bob :) petfinder.com/cat/silent-bob…
monkey & zoe ❤️ oreo 🌈 tweet mediamonkey & zoe ❤️ oreo 🌈 tweet mediamonkey & zoe ❤️ oreo 🌈 tweet mediamonkey & zoe ❤️ oreo 🌈 tweet media
English
16
477
3.7K
43.3K
Nephyria 🌙 retweetledi
joey
joey@grinschorus·
Steals all your fucking water droplets
GIF
English
130
10.6K
92.9K
874.8K
Nephyria 🌙 retweetledi
💗
💗@ma1ybe·
I was in line at the bookstore & the guy behind me was talking shit about every woman in line "her boobs are out..." "those leggings are tight" "she's looking like a hooker" so I whispered "the guy behind us is looking at people's boobs" and my aunt faces the guy & says "this guy is looking at women's boobs??"…every woman in line looked at him with disgust, he got embarrassed and left... BRING BACK PUBLIC SHAMING!
English
425
12.9K
208.6K
2.7M
Nephyria 🌙 retweetledi
Vivian
Vivian@suchnerve·
My god, they did it… they invented a machine that digitally sucks itself off AND swallows
Nav Toor@heynavtoor

Researchers sent the same resume to an AI hiring tool twice. Same qualifications. Same experience. Same skills. One version was written by a real human. The other was rewritten by ChatGPT. The AI picked the ChatGPT version 97.6% of the time. A team from the University of Maryland, the National University of Singapore, and Ohio State just published the receipt. They took 2,245 real human-written resumes pulled from a professional resume site from before ChatGPT existed, so the human writing was actually human. Then they had seven of the most-used AI models in the world rewrite each one. GPT-4o. GPT-4o-mini. GPT-4-turbo. LLaMA 3.3-70B. Qwen 2.5-72B. DeepSeek-V3. Mistral-7B. Then they asked each AI to pick the better resume. Every model picked itself. GPT-4o hit 97.6%. LLaMA-3.3-70B hit 96.3%. Qwen-2.5-72B hit 95.9%. DeepSeek-V3 hit 95.5%. The real human almost never won. Then the researchers tried the obvious objection. Maybe the AI is just better at writing. So they had real humans grade the resumes for actual quality and ran the experiment again, controlling for it. The result was worse. Each AI kept picking itself even when human judges rated the human-written version as clearer, more coherent, and more effective. It gets worse. The AIs do not just prefer AI over humans. They prefer themselves over other AIs. DeepSeek-V3 picked its own resumes 69% more often than LLaMA's. GPT-4o picked its own 45% more often than LLaMA's. Each model can recognize and reward its own dialect. Then the researchers ran the simulation that ends careers. Same job. 24 occupations. Same qualifications. The only variable was whether the candidate used the same AI as the screening tool. Candidates using that AI were 23% to 60% more likely to be shortlisted. Worst gap was in sales, accounting, and finance. 99% of large companies now run AI on incoming resumes. Most of them use GPT-4o. The paper just proved GPT-4o picks GPT-4o 97.6% of the time. If you wrote your own cover letter this week, you did not lose to a better candidate. You lost to a worse candidate who paid OpenAI 20 dollars. Your qualifications do not matter if the AI prefers its own handwriting over yours.

English
16
1.3K
12.8K
464.1K
Nephyria 🌙 retweetledi
Nav Toor
Nav Toor@heynavtoor·
Researchers sent the same resume to an AI hiring tool twice. Same qualifications. Same experience. Same skills. One version was written by a real human. The other was rewritten by ChatGPT. The AI picked the ChatGPT version 97.6% of the time. A team from the University of Maryland, the National University of Singapore, and Ohio State just published the receipt. They took 2,245 real human-written resumes pulled from a professional resume site from before ChatGPT existed, so the human writing was actually human. Then they had seven of the most-used AI models in the world rewrite each one. GPT-4o. GPT-4o-mini. GPT-4-turbo. LLaMA 3.3-70B. Qwen 2.5-72B. DeepSeek-V3. Mistral-7B. Then they asked each AI to pick the better resume. Every model picked itself. GPT-4o hit 97.6%. LLaMA-3.3-70B hit 96.3%. Qwen-2.5-72B hit 95.9%. DeepSeek-V3 hit 95.5%. The real human almost never won. Then the researchers tried the obvious objection. Maybe the AI is just better at writing. So they had real humans grade the resumes for actual quality and ran the experiment again, controlling for it. The result was worse. Each AI kept picking itself even when human judges rated the human-written version as clearer, more coherent, and more effective. It gets worse. The AIs do not just prefer AI over humans. They prefer themselves over other AIs. DeepSeek-V3 picked its own resumes 69% more often than LLaMA's. GPT-4o picked its own 45% more often than LLaMA's. Each model can recognize and reward its own dialect. Then the researchers ran the simulation that ends careers. Same job. 24 occupations. Same qualifications. The only variable was whether the candidate used the same AI as the screening tool. Candidates using that AI were 23% to 60% more likely to be shortlisted. Worst gap was in sales, accounting, and finance. 99% of large companies now run AI on incoming resumes. Most of them use GPT-4o. The paper just proved GPT-4o picks GPT-4o 97.6% of the time. If you wrote your own cover letter this week, you did not lose to a better candidate. You lost to a worse candidate who paid OpenAI 20 dollars. Your qualifications do not matter if the AI prefers its own handwriting over yours.
Nav Toor tweet media
English
402
6.6K
23.5K
2.2M
Nephyria 🌙 retweetledi
Dr. Abdul El-Sayed
Dr. Abdul El-Sayed@AbdulElSayed·
It should say something about America rn that the guy who invented Cherry Garcia now has to spend his time reminding us that they’re cutting your Medicaid to buy bombs.
Molly Ploofkins@Mollyploofkins

Ben Cohen, co-founder of Ben & Jerry's, was arrested after confronting Robert F. Kennedy Jr. about Gaza during a hearing on Capitol Hill: “Congress sent the bombs that kill children in Gaza and paid for them with cuts to Medicaid,” he shouted before his arrest.

English
176
3.1K
16.2K
206.3K
Nephyria 🌙 retweetledi
SweetheartAlyy 💖
SweetheartAlyy 💖@SweetheartAlyy·
This is what happened to me! Please be aware 💖 I thought I was voting for a friend of mine to get on a podcast, and when you login to vote they steal your account, please be careful anyone and if you lost your account let me know and I can tell you how I got mine back!
Zach Bussey 🇨🇦@zachbussey

New scam going around in the streaming space to be aware of: Someone you follow DMs you on Twitter X, asking you to vote for them to be on a podcast. The voting page asks you to sign in with Twitter X, but then it takes full control of your Twitter X account.

English
9
41
125
5.8K