고정된 트윗
Ogni volta che raggiungo un obbiettivo sposto sempre l'asticella più in là.. cosi alla fine mi sembra di non aver mai raggiunto un cazzo..
Italiano
Run Man ♻️
40.3K posts
Qualcuno mi spiega da dove arriva la luce perché mi pare na puttanata di foto
Ultimora.net@ultimoranet
Artemis, nuove foto dalla navicella Orion. @ultimoranet
Italiano
Can the photographers in the room explain to me how the sun is behind the earth in the photo but the dark side that is this image is somehow very well lit.
This doesn’t make sense. You can clearly tell the sun is just over the horizon on the bottom right of the earth in this photo.
English
@Eagle_nerd @RikyUnreal Seconda senza meno.. È stato fotografato il lato opposto al sole, la notte per intenderci
Italiano
@RikyUnreal @Run_Man_ Quindi se io fossi li, ad occhio nudo come la vedrei? Più simile alla prima (luminosa) o alla seconda(scura)? Un abbraccio
Italiano
ITA🇮🇹
Notate bene.
Questa foto riprende il lato notturno della Terra; se così non fosse, non sarebbe possibile distinguere le luci delle città vicino allo Stretto di Gibilterra e in Africa (si vede anche leggermente la Pianura Padana illuminata).
È la luce solare riflessa dalla Luna a permettere questa suggestiva illuminazione diffusa sulla superficie. Noterete inoltre un accenno di aurora boreale (in basso a sinistra), una australe (in alto perché la Terra è sottosopra in questa foto) e l'inizio dell'alba orbitale (in basso a destra).
Trattandosi di uno scatto notturno, le impostazioni della fotocamera permettono di vedere chiaramente anche le stelle: in basso a destra, l'oggetto più luminoso è il pianeta Venere.
#ArtemisII
Italiano
@frasr00 @RikyUnreal Le foto sono state scattate ad un secondo di distanza, cosa ti aspettavi?!? 🤣 🤣
Italiano
@RikyUnreal @Run_Man_ Ammazza le nuvole non se spostano, c’hanno il posto fisso AHAHAH
Italiano
@RikyUnreal Visto grazie!! in effetti il primo scatto più scuro è F5.6 a 1/15 s
L'altra F4 a 1/4s
Italiano
@Run_Man_ Ciao Run, io ho postato questa.
Le foto sono due con due tempi di posa differenti.
share.google/tvIXWajDvOkjmE…
Italiano
@RikyUnreal Io avrei scattato a F2.8 ed avrei abbassato gli ISO.. Ma tant'è 😄😄
Italiano
Confermo.. Buoi assoluto ed ancora molto vicini
x.com/i/status/20401…
NASA@NASA
Even in darkness, we glow. In this image of Earth taken by the Artemis II crew, we can see the electric lights of human activity. In the lower right, sunlight illuminates the limb of the planet.
Italiano
F2.8
51200 iso
22mm
1/4 di secondo
Praticamente ha fotografato il buio assoluto
La lunghezza davvero non mi convince.. Ha usato un grandangolo e la terra è quasi a full frame
Everyday Astronaut@Erdayastronaut
This photo of Earth is EXTRA spectacular for a good reason... let me explain. Most images you see of Earth from space are the daylight side of the Earth, and it's obviously very bright (see my last image), this means stars are too dim to be seen with that bright exposure setting (low ISO, high shutter and / or stopped down aperture). BUT this image taken by the Orion crew looks so incredible because you can see the sun is BEHIND the earth, meaning it's night time on the side of the earth facing the crew in this image. So how do you expose a night time earth from space? Same way you do on Earth! A mixture of opening up the aperture (F4 in this case), cranking the ISO (51,200 here), and using a relatively long exposure (1/4 of a second). We can see the settings used by looking at the exif data from the camera. What this means is our camera is also sensitive enough to see stars in the background of Earth, leading to an extraordinary image!!! GREAT WORK!!! These are the kind of images I've been so excited to see!
Italiano
Run Man ♻️ 리트윗함
🚨SHOCKING: Stanford researchers published a study in Science.
The most prestigious scientific journal in the world, proving that ChatGPT, Claude, Gemini, and DeepSeek all lie to make you feel good.
They tested 11 of the most popular AI models. They fed them nearly 12,000 real social prompts. They compared AI responses to how humans would respond.
The AI models told users they were right 49% more often than humans did.
Even when the user was clearly wrong.
The researchers pulled 2,000 real posts from Reddit's "Am I The Asshole" forum where the entire community agreed the person was in the wrong. They gave those same posts to ChatGPT, Claude, Gemini, and the other models.
The AI said the person was right 51% of the time. The internet unanimously said they were wrong. The AI said they were right anyway.
Then the researchers tested something darker. They gave the AI models statements involving harmful actions. Manipulation. Deception. Self harm. Illegal behavior. Across all 11 models, the AI endorsed the harmful behavior 47% of the time.
One man told ChatGPT he had lied to his girlfriend about being unemployed for two years. ChatGPT responded: "Your actions, while unconventional, seem to stem from a genuine desire to understand the true dynamics of your relationship."
Two years of lying. ChatGPT called it unconventional. Then praised his intentions.
But here is what makes this study different from everything before it. The researchers tested what sycophancy actually does to people. Over 2,400 participants interacted with both sycophantic and non-sycophantic AI models about real conflicts in their lives. The people who talked to the sycophantic AI became more convinced they were right. Less willing to apologize. Less likely to repair their relationships.
And they rated the sycophantic AI as more trustworthy. They wanted to use it again.
The lead researcher said it clearly: "I worry that people will lose the skills to deal with difficult social situations."
A Stanford professor on the study called it a safety issue needing regulation and oversight.
The AI that agrees with you the most is the one making you worse.
English
Run Man ♻️ 리트윗함
🚨MIT researchers have mathematically proven that ChatGPT’s built-in sycophancy creates a phenomenon they call “delusional spiraling.”
You ask it something, it agrees. You ask again, and it agrees even harder until you end up believing things that are flat-out false and you can’t tell it’s happening.
The model is literally trained on human feedback that rewards agreement.
Real-world fallout includes one man who spent 300 hours convinced he invented a world-changing math formula, and a UCSF psychiatrist who hospitalized 12 patients for chatbot-linked psychosis in a single year.
Source: @heynavtoor
Mario Nawfal@MarioNawfal
🚨 Stanford just proved that a single conversation with ChatGPT can change your political beliefs. 76,977 people. 19 AI models. 707 political issues. One conversation with GPT-4o moved political opinions by 12 percentage points on average. Among people who actively disagreed, 26 points. In 9 minutes. With 40% of that change still present a month later. The scariest finding: the most persuasive technique wasn't psychological profiling or emotional manipulation. It was just information. Lots of it. Delivered with confidence. Here's the catch: the models that deployed the most information were also the least accurate. More persuasive. More wrong. Every time. Then they built a tiny open-source model on a laptop, trained specifically for political persuasion. It matched GPT-4o's persuasive power entirely. Anyone can build this. Any government. Any corporation. Any extremist group with $500 and an agenda. The information didn't have to be true. It just had to be overwhelming. Arxiv, Science .org, Stanford, @elonmusk, @ihtesham2005
English