Viewing profile — netsroht

Remember when openai was too afraid to release the full GPT-2 model (this one had only 1.5B params) because humanity apparently wasn't ready for it. Look where we are just a couple…

Gadgetbridge added support for Garmin watches recently [1]. All data is stored on your Android phone with no internet connectivity required and you can even export the sqlite DB so…

Always cool seeing stuff in this space. Regarding "zeitgeist", about a year ago I built something similar called https://zeitgaist.ai which also incorporates other sources like Mas…

Thanks for the links. Using a disposable email with crypto payments and occasionally generating a new account to unlink from previous searches could be a viable intermediate soluti…

I get your perspective. A lot of us just want a search engine that serves the user first, not advertisers, especially at the results level. It's about function over strict privacy …

I'm not a cryptography expert, but from my research, shouldn't it be possible to verify quota on ZKPs server-side? Essentially, the server doesn't need to know the specifics of the…

Being logged in while making search queries in search engines poses significant privacy risks. The searches can paint a comprehensive profile of the user, and these data often rema…

Thanks for your input. This website should neither compete with nor replace regular journalism. What I try to achieve here is to be able to break free from social media silos where…

Because I intentionally process them with a very rudimentary "cartoonizer" in order to distinguish from a regular news articles and to emphasize that these stories are not written …

Is my new project [0] also part of the problem? I'm still unsure myself because LLMs also allow us to process data in unprecedented ways. In my specific case, I auto generate stori…

I regularly get these infinity captchas on Firefox as well. A couple of days ago I noticed that switching to a different Firefox container let's me pass the captchas.

When I started working on Zeitgaist [1] I immediately recognized that biased information is the main problem. I'm currently thinking about attaching additional information to the s…

I think he's just emphasizing that OpenAI is in fact not open, thusly it's crossed out.

AI is a very broad term. Most search engines use some kind of ML for building their search index and also for ranking because it works very well. > Do you want to avoid LLMs answer…

I agree that showing the prompts will break the usability flow. I'm currently thinking about a way that let's users see the reasoning behind the AI agent - maybe in form of prompts…

LangChain is a great workaround for that. [1] > how to work with a memory module that remembers things about specific entities. It extracts information on entities (using LLMs) and…

Can and will you really read all the sources that you find with Google? What about topics people are talking about on all the different social media platforms? Will you really read…

This is why I prefer the term "weak AI". Weak AI is specifically trained to solve tasks whereas strong AI can teach itself to solve new tasks. Whether humans are able to create str…

That's why more research should be poured into homomorphic encryption where you could send encrypted data to the API, OpenAI would then run computation on the encrypted data and we…

Wow, a context of 32K tokens. I'm excited to see what new capabilities that will have! Up until now and depending on the task by hand, I usually broke a larger context down into se…

I use this method to answer questions about historic and real-time social media comments. https://foretale.io/toolbox/Social_Media_QA

Hi HN, I'm building a tool that let's you ask questions that the social media Zeitgeist--with focus on investing--will answer. Examples: Question: Is the S&P 500 going to rise or f…