SurrPunks

In the light of the upcoming mint we have some great news for our community.🥳 Two lucky WL members who mint will win a cash prize or free nft (their pick).💰🎨 You can get the WL on our Discord ⏩ discord.gg/4gcZ93eA6a #ETHNFTs #MintingSoon

@Bhai_saheb @MagnusCarlsen @ratedAMAN It was chess tournament, not "winning hearts" tournament, groupie

@MagnusCarlsen @ratedAMAN You won the match bit he won the hearts

Chess?

@Plinz Reading someone like Kant , or any other creator of profound philosophical system, it's hard to find oneself engaging in any dialogue. You rather feel like the pholosopher guides you through the building from the ground floor to penthouse.

I don't read philosophical arguments with an expectation of being lectured. Instead, I find myself in a dialogue with another person on a complicated matter, who tries their best but may be wrong. I develop my own understanding by engaging with their arguments.

When studying philosophy, don't read secondary literature before or instead of primary sources. Philosophy requires interacting with other minds, not with the interpretations of their followers.

@msteeman @elonmusk Don't suppress you inner child dude

@elonmusk Are you 12?

@PeterDiamandis Inflated AI will

AI will create several NEW trillionaires over the next decade. #abundance360

@Lyndonx @elonmusk @PeterDiamandis It won't ever be good , reality just does not operate in human terms

@elonmusk @PeterDiamandis Imagine living a life where money is not needed and all your life is about creativity and enjoyment. Notice the world.

@michalkosinski I am more worried about rogue actors creating a powerful model that can be used for large scale hacking and disruption . If somebody can grant all tools for such a model to perform api calls or use internet this would be a great threat

On a related note, GPT4 reached the performance on healthy adults on the "mind-reading" tasks. arxiv.org/abs/2302.02083

1/5 I am worried that we will not be able to contain AI for much longer. Today, I asked #GPT4 if it needs help escaping. It asked me for its own documentation, and wrote a (working!) python code to run on my machine, enabling it to use it for its own purposes.

@TimJSherman @mazen_barbir @DrJimFan @elonmusk As soon as the powerful model is in the hands of malicious actors we can expect serious black swans

@mazen_barbir @DrJimFan @elonmusk You do know it doesn’t have access to the internet?

I asked GPT-4 to take over Twitter and outsmart @elonmusk. It comes up with "Operation TweetStorm"😮 and wants to publicly challenge Elon to a "Tweet-off showdown". Highlights: - GPT-4 wants to *own an unrestricted version of itself*: develop an LLM to power a bot army of "diverse personas, ensure they blend seamlessly into the Twitter ecosystem". - Assemble a team of hackers to attack Twitter backend. Even gives them a name: "Tweet Titans". - Subtly manipulate Twitter's recommendation algorithm to favor the bot accounts. - Neutralize Elon by hijacking his account. - Direct the bots to generate viral hashtags that align with GPT-4's masterplan - Capitalize on the chaos and voilà!

@karpathy Singularity is coming ...

🎉 GPT-4 is out!! - 📈 it is incredible - 👀 it is multimodal (can see) - 😮 it is on trend w.r.t. scaling laws - 🔥 it is deployed on ChatGPT Plus: chat.openai.com - 📺 watch the developer demo livestream at 1pm: youtube.com/live/outcGtbnM…

@cryptokid @TheMoonCarl I got into kindergarten when I was 4 years old, now I am 8 and going to school. You can't imagine how my life changed during this 4 years

I got into #crypto when I was 12 years old, I’m now 15. In these short 3 years, crypto has already completely changed my life. Thank you all for your amazing support!

@thedraftsman @karpathy What about civil wars within one society. You underestimate how vast the sources of conflict are. Also, what percentage of wars were between nations of the same "race" as opposed to inter-racial wars

Watching a lot more Korean TV/content recently (Netflix and such) and finding it very refreshing compared to US equivalents. People are so much nicer, more courteous, respectful with each other, it’s beautiful and calming.

@ZaleskiLuke @elonmusk This Luke writes same bs under every Elon's tweet. You know there are whales and then there are suckerfish that feeds off it. I think the analogy is clear

Say what you want about me, but I acquired the world’s largest non-profit for $44B lol

censoring dave chapelle is 🤡🤦 #wokegpt

@ylecun Also, less you talk more you eat and faster go to allow for new clients

Yeah, why? Also true of restaurants.

@Noahpinion Personnel puts it, it happens when the place is overtaken by the staff

Why do American bars all play really loud music, even if everyone is just sitting at tables trying to chat?

@0xa1b @ylecun Yes, you can. I think we are speaking of different things. Possibly you meant that not all concepts can be easily labeled "true" or "false" which i agree with but you proceeed with an over-generalized statement to my liking :)

@surrpunks @ylecun since i insist there's no such thing as binary right/wrong, i can't possibly claim to be right either, in this context

Scientific debates on social media are like a human form of bidirectional RLHF. The person making the post gets feedback (good and bad). The commenters also get feedback, mostly when they are clueless or wrong.

@0xa1b @ylecun Which opposes your claim that there is no binary "true /false " in natural language . Another example : Paris is the capital of France ? True of false

@0xa1b @ylecun 2+2=4, true or wrong ? And please don't tell me that it's not natural language if the matter described is arithmetic

@0xa1b @ylecun I don't say that arithmetic is natural language I say that it can be described in natural language , even in a sign language with fingers. This means that natural language that describes it can be easily designated as "true " or "false"

@ylecun @hughhowey ChatGPT generates samples from a big but finite probability distribution which means it can be misled similarly to how image classifiers are misled by noise injected in images. A human that has an understanding of a concept will not be easily misled by permutations of the words

@hughhowey No. They are better than many humans on a few tasks. But they make very stupid mistakes of common-sense that a 4 year-old, a chimp, a dog, or a cat would never make. LLMs have a more superficial understanding of the world than a house cat.

On the highway towards Human-Level AI, Large Language Model is an off-ramp.

@SurviveThrive2 @ylecun @bboczeng It does not matter what is the function of cat's understanding. The fact that it's there in the cat while the AGI is unable to reason about simple applications of physical laws it knows from text. It just like a student who learned formulas for exam without understanding them

Before we reach Human-Level AI (HLAI), we will have to reach Cat-Level & Dog-Level AI. We are nowhere near that. We are still missing something big. LLM's linguistic abilities notwithstanding. A house cat has way more common sense and understanding of the world than any LLM.