Eden Lumbroso

Human

[Eden Lumbroso @Edenlum]

@ofek_kasif לא צריך להאמין לכל דבר שיוצא לפוליטיקאים מפה... גם אם הם ערבים

[Eden Lumbroso @Edenlum]

@RadmanMoshe מה תמונת הניצחון לדעתך?

[Eden Lumbroso @Edenlum]

@tonywestonuk yes it does. depends on the context

[Eden Lumbroso @Edenlum]

Multiple questions to unpack:
1. what is consciousness?
2. given that an entity is conscious, should it have some rights?
3. my favorite question (in every context): what is our goal here?

One could say the goal is to minimize human suffering, and consciousness comes into the conversation because humans feel bad when they encounter an entity they consider conscious that is suffering itself. Like, people feel bad when they see an animal crying because it is hurt.

Given this, the definition could be flipped with the same outcome: An entity is conscious if it makes us feel bad when it appears to suffer. Then, trying to minimize human suffering, we should protect that being (notice we don't directly care about the conscious being itself).

Taking this definition to AI, an AI is conscious when it makes enough people believe it is indeed suffering. At that point, just because it makes people feel bad, we need to protect it from suffering (or make it unconscious by somehow convincing people they shouldn't feel bad).

Side point: In this framework consciousness is like beauty (an "edge property"). An entity is not objectively beautiful, it is "in the eye of the beholder". In other words, if you would put all people as nodes on a graph, height (for example) would be a node property, but beauty would be an "edge property" between two people. In this sense consciousness is also an edge property.

[Robert Scoble @Scobleizer]

It isn’t just RSI.

At about 300 billion parameters their neural networks start behaving more human, AI researchers tell me. Their neurons start moving around like ours do. In other words they start learning faster.

An emergent property they didn’t expect.

Which is why there are more discussions about consciousness lately. And more angst about bad things AI might soon do.

The clankers are soon gonna make us all think they are conscious.

What will you call it when you have humanoids in your home talking to you?

I don’t believe melted sand should be granted rights we hold dear as humans.

And that is the hill I will defend.

[Eden Lumbroso @Edenlum]

@tonywestonuk 1. bad definition, how to test?
2. yes they do
3. so now you define consciousness as fear of death? btw AI has this

[Eden Lumbroso @Edenlum]

@ValerioCapraro We can't even define consciousness. Here is an attempt though:
https://t.co/SyZXURded5

[Eden Lumbroso @Edenlum]

Multiple questions to unpack:
1. what is consciousness?
2. given that an entity is conscious, should it have some rights?
3. my favorite question (in every context): what is our goal here?

One could say the goal is to minimize human suffering, and consciousness comes into the conversation because humans feel bad when they encounter an entity they consider conscious that is suffering itself. Like, people feel bad when they see an animal crying because it is hurt.

Given this, the definition could be flipped with the same outcome: An entity is conscious if it makes us feel bad when it appears to suffer. Then, trying to minimize human suffering, we should protect that being (notice we don't directly care about the conscious being itself).

Taking this definition to AI, an AI is conscious when it makes enough people believe it is indeed suffering. At that point, just because it makes people feel bad, we need to protect it from suffering (or make it unconscious by somehow convincing people they shouldn't feel bad).

Side point: In this framework consciousness is like beauty (an "edge property"). An entity is not objectively beautiful, it is "in the eye of the beholder". In other words, if you would put all people as nodes on a graph, height (for example) would be a node property, but beauty would be an "edge property" between two people. In this sense consciousness is also an edge property.

[Eden Lumbroso @Edenlum]

@Nehemia_GA אחלה ערוץ 16 116 מנדטים
וגם וואלה בעצם

[Eden Lumbroso @Edenlum]

@SirPitta @ZuckermanRoy המציאות זה מה שרלוונטי. אתה מכיר נתונים שמראים שרופאים ערבים פוגעים במטופלים יהודיים באופן מכוון?

[Eden Lumbroso @Edenlum]

@FrancoisChauba1 but zero order is still bad

[Eden Lumbroso @Edenlum]

@OmerNolan8 גם מילולית וגם פיזית? מה לגבי רוחנית?

[Eden Lumbroso @Edenlum]

@TommerVinner לא ככה עובדים סקרים.

[Eden Lumbroso @Edenlum]

@Nadav_Eyal מי מייעץ לאייזנקוט? האם השם משפחה שלו מתחיל ב״נתניה״

[Eden Lumbroso @Edenlum]

@ZelaArber If you iterate 1 transformer block, how are you going to scale the number of parameters? just make the embed dimension huge? or more experts in Moe?
Cool work!

[Eden Lumbroso @Edenlum]

@guylerer על מי מדובר

[Eden Lumbroso @Edenlum]

@GRozentzva34070 @andebetayalla 🤦‍♂️

[Eden Lumbroso @Edenlum]

@GRozentzva34070 @andebetayalla סבבה העקרון עדיין תקף

[Eden Lumbroso @Edenlum]

@chicomenashe תרחיש:
אין 61 בבחירות.

ביבי לומד לעוף ויורה לייזר מהעיניים. הוא מזמין את אייזנקוט לדו קרב עד המוות.

אייזנקוט מסכים ונלחם בו אבל רובים לא עובדים עליו. בנט בא לעזרת חבר ותופס את ביבי מאחור. בן גביר משתמש במהלך האולטימטיבי שלו ומפוצץ את הזירה.

הכל מתמלא אבק, העולם במתח.

דונו.

[Eden Lumbroso @Edenlum]

@GRozentzva34070 @andebetayalla על מה אתה מדבר? אמרתי משהו עקרוני שלא קשור לישראל אפילו. נגיד אני חושב שלהוריד את המס על רכבים ישפר את המצב הכלכלי בישראל, אני מעדיף מישהו שחושב הפוך, אבל הוא הגון וחכם, ממישהו שחושב כמוני, אבל טיפש ומושחת.
לכן הגיוני ש״ימנים ושמאלנים״ יצביעו לאותו דבר אם האלטרנטיבה זה שחיתות

[Eden Lumbroso @Edenlum]

@GRozentzva34070 @andebetayalla לא. לפני אידאולוגיות של איך לשפר, צריך להסכים שרוצים לשפר. כל אדם שפוי מעדיף אנשים חכמים ועפ כוונות טובות בממשלה שלא מסכימים איתו על מה הפתרונות, מאנשים טפשים ומושחתים ועם כוונות רעות שכן מסכימים איתו על מה הפתרונות.

[Eden Lumbroso @Edenlum]

@levanonisrael זה גוש רק בראש שלך