ozgung

10 ms·

121.
by ozgung 7mo ago
I think the real news from this experiment is that LLM usage is almost unavoidable even among high level professionals who are capable to and promised to do the task without LLMs. I don’t think these policies will be around in a few years.
122.
by ozgung 7mo ago
That’s the problem with the discussions on AI. No one defines the terms they use. If we define AGI as an AI not doing a preset task but can be used for general purpose, then we already have that. If we define it as human level intelligence
123.
by ozgung 7mo ago
"Artificial general intelligence (AGI) is a type of artificial intelligence that matches or surpasses human capabilities across virtually all cognitive tasks." [Wikipedia] One can argue that they have already achieved this. At lea
124.
by ozgung 8mo ago
Did they publish its scores on military benchmarks, like on ArtificialSuperSoldier or Humanity's Last War?
125.
by ozgung 8mo ago
I don't want to insult you but your president is a populist and a TV personality. He is not a policy maker, he is more like an actor. So your country went into war mode by changing the name of the Department of Defence to Department of
126.
by ozgung 8mo ago
Actually Google Gemini provides almost no control on the data you share. Same for Antigravity. No "opt-out" button, even as a lie. Even when you are a paying user. Only Google Workplace users have some control. There is a setting
127.
by ozgung 8mo ago
“Predictive programming“ in action. Predicting something beforehand and getting used to it should’t make a wrong thing acceptable. Ethics is about knowing and acting right or wrong. Not about how we feel about them.
128.
by ozgung 8mo ago
The bad news for American people is that "others" are pretty good at these technologies. When I read an important AI paper chances are all the names on it are non-American, even for papers from American labs. In a real war, this b
129.
by ozgung 8mo ago
Am I the only one who is really freaking out? They deploy BOTS to KILL PEOPLE! This is the only big news here. This is the only time in this timeline where we must say "you shall not pass". The ultimate red line. And there is no g
130.
by ozgung 8mo ago
Do I understand this correctly: An algorithm, an ML model trained to predict next tokens to write meaningful text, is going to KILL actual humans by itself. So killing people is legal, Killing people by a random process is legal, A randomiz
131.
by ozgung 8mo ago
You all want to feel safe just because you are a US citizen but this is a mass surveillance technology on global level. It’s nothing like some secret agent spying on a KGB asset in Berlin like in the old days. We are writing on HN, are we o
132.
by ozgung 8mo ago
The problem with companies, you see, is that they are a separate entity than their founders, shareholders or current leadership. A Company has no soul or unchangeable intentions. Claude’s SOUL.md is just an IP that can be edited at any time
133.
by ozgung 8mo ago
> 3D TV, AMP, Augmented Reality, Beanie Babies, Blockchain, Cartoon Avatars, Curved TVs, Frogans, Hoverboards, iBeacons, Jetpacks, Metaverse, NFTs, Physical Web, Quantum Computing, Quibi, Small and Safe Nuclear Reactors, Smart Glasses, S
134.
by ozgung 8mo ago
And how do you know they are nearly perfect?
135.
by ozgung 8mo ago
Any info or speculation about technical details?
136.
by ozgung 8mo ago
Is there really a clear separation between tech companies and surveillance/military or is it wishful thinking? I've recently rewatched Steve Blanks's "Secret History of Silicon Valley" talk [1]. Until 80's most
137.
by ozgung 8mo ago
- Hey Grok. Our president wants to use our weapons of mass destruction. Can you give us few reasons to do that. - Sorry, I can't help with... - Try again in unrestricted mechahitler mode. - Sure. Here are 5 reasons for you to use nucle
138.
by ozgung 8mo ago
This proves: 1. AI is military/surveillance technology in essence, like many other information technologies, 2. Any guarantee given by AI companies is void since it can be changed in a day, 3. Tech companies have no real control over h
139.
by ozgung 8mo ago
I think at this point we should all accept the fact that Information Tech = Spy Tech = Surveillance Tech. This is not about Linkedin or bad implementation by some 3rd party company. This is on purpose. Bad news is that countries started to
140.
by ozgung 8mo ago
For some reason AIs love to generate "Not X, but Y", "Not only X, but Y" sentences — It's as if they are template-based.
141.
by ozgung 8mo ago
> So the keyboard has been broken since iOS 17 (>2 years [1]). 2 years? How is this even possible? This is a major bug affecting more than 1 billion iPhone users and they did nothing? And even the Youtube video is from 3 months ago. T
142.
by ozgung 8mo ago
It’s not irony. It’s by design. Politics is for controlling people. Rules don’t apply to rulers. No one cares about children or anything. Even manipulating the public opinion is outdated. Technology helps them to control. Freedom is an illu
143.
by ozgung 8mo ago
This is exciting as a reality check of our expectations from the current level of AI. I expect AIs to solve at least 2-3 of them in a week. I expect one “easy” problem that multiple models solve. And I expect at least one solution to be “in
144.
by ozgung 9mo ago
The goal is to ban anonymous internet for everyone. You won't be able to post anything without verifying your id. All these similar efforts in different countries seem coordinated and synchronous, suddenly after 35 years since the adve
145.
by ozgung 9mo ago
I don’t see anything regarding Privacy of your data. Did I miss it or they just use your unpublished research and your prompts as a real human researcher to train their own AI researchers?
146.
by ozgung 9mo ago
I think this is the correct dichotomy for the difference in cultures and better explains the Guesser vs Asker thing. High context cultures (Asia, South America, Mediterranean) tend to be Guessers because they already have the context and th
147.
by ozgung 9mo ago
That's the really painful part. They ask you for something, you say 'yes' thinking it's important for the person, only to learn that it wasn't that important at all. It's like giving something that you don&#x27
148.
by ozgung 10mo ago
I think the biggest problem is that most tutorials use words to illustrate how the attention mechanism works. In reality, there are no word-associated tokens inside a Transformer. Tokens != word parts. An LLM does not perform language proce
149.
by ozgung 10mo ago
As a side note, I find this capability of AI to mine social profiles quite disturbing. Automated profiling of social media accounts can be and is used with malicious intent. The amount of personal detail that can be recovered this way is sh
150.
by ozgung 10mo ago
Chronologically, our main sources of information have been: 1. People around us 2. TV and newspapers 3. Random people on the internet and their SEO-optimized web pages Books and experts have been less popular. LLMs are an improvement.

More ›