HN user

pps

3,238 karma
Posts203
Comments239
View on HN
arxiv.org 1y ago

Arecibo Wow I: An Astrophysical Explanation for the Wow Signal

pps
1pts0
annas-archive.se 1y ago

The critical window of shadow libraries

pps
5pts0
embracethered.com 1y ago

Breaking Instruction Hierarchy in OpenAI's GPT-4o-mini

pps
1pts0
www.notion.so 2y ago

Why AI gets better when we let it explore

pps
3pts0
willdarwin.com 2y ago

Note Taking in Anki

pps
8pts1
www.theguardian.com 2y ago

February on course to break unprecedented number of heat records

pps
2pts0
latentdictionary.com 2y ago

Latent Dictionary: 3D map of Oxford3000+search words via DistilBert embeddings

pps
89pts42
wattenberger.com 2y ago

Evolving the Infinite Canvas

pps
94pts34
www.youtube.com 2y ago

SmartGPT: Major Benchmark Broken – 89.0% on MMLU and Exam Errors [video]

pps
4pts5
arc.net 2y ago

Arc Browser 1.0

pps
186pts255
www.youtube.com 3y ago

Automata: The Extraordinary “Robots” Designed Hundreds of Years Ago [video]

pps
4pts0
www.sciencedaily.com 3y ago

New research puts age of universe at 26.7 – not 13.7 – billion years

pps
44pts12
www.collider.space 3y ago

Effective Note-Taking: Avoid Common Traps

pps
2pts0
www.psypost.org 3y ago

We are wrong to think that morality is declining over time

pps
2pts0
www.openforum.com.au 3y ago

Geometric constraints on human brain function

pps
2pts0
www.reproof.app 3y ago

Notes apps are where ideas go to die (2022)

pps
694pts223
fivebooks.com 3y ago

Five Books: The best books on everything

pps
243pts85
www.youtube.com 3y ago

John Vervaeke – AI: The Coming Thresholds and the Path We Must Take

pps
2pts0
github.com 3y ago

Grobid – ML software for extracting information from scholarly documents

pps
4pts0
www.youtube.com 3y ago

Cal Newport’s Thoughts on ChatGPT

pps
2pts0
www.ykilcher.com 3y ago

OpenAssistant Conversations – Democratizing Large Language Model Alignment [pdf]

pps
251pts61
www.abc.net.au 3y ago

How do we cultivate deep reading processes in a digital age?

pps
2pts0
every.to 3y ago

AI Doomsday for People Who Don’t (Yet) Wear Fedoras

pps
2pts0
www.youtube.com 3y ago

Demo of the first OpenAssistant models

pps
1pts0
robotic.substack.com 3y ago

What it feels like to work in AI right now

pps
350pts737
openai.com 3y ago

Our Approach to AI Safety

pps
76pts108
noahpinion.substack.com 3y ago

Why Americans fear the AI future

pps
3pts0
cube.fyi 3y ago

Cube CSS – Simple, consistent, and pragmatic CSS methodology

pps
2pts0
every.to 3y ago

A Short History of Artificial Intelligence

pps
1pts0
www.oliverburkeman.com 3y ago

Treat your to-read pile like a river, not a bucket

pps
404pts186
Kagi Assistant 2 years ago

In my opinion, the results from Kagi Assistants are significantly inferior. I'm subscribed to both services and tried the Ultimate tier for a few days two weeks ago to test the new assistants. If they add the ability to search with multiple steps and allow the AI to digest more results, then they could be on par. It's more similar to ChatGPT at this stage, IMO.

Kagi Assistant 2 years ago

There's no noticeable difference in search speed between Kagi and Google for me in Poland.

Bitwarden isn't pinned anywhere so to log in I need to click "Extensions > Bitwarden"

I also open Bitwarden often, so I just use the keyboard shortcut shift+cmd+b, not sure if that's the default, you can change it here arc://extensions/shortcuts

Another method I use for other extensions is cmd+t, then type the first letters of the extension to open it.

How they intend to do that without any staff is something I would like to know about

There are great note-taking apps that only a few people work on. Tana, Obsidian, Bear, UpNote come to mind, but there are many more. So it's definitely not a problem. It sucks for the old Evernote staff, but it doesn't have to suck for users, it might even be what Evernote needs, since it's pretty ancient in its feature set compared to the competition.

You can give it the article in parts. Here is the summary by GPT-4:

In a kingdom plagued by a dragon that demands a daily human sacrifice, the people have come to accept the beast as an inevitable fact of life. The dragon represents human aging and the sacrifices symbolize the lives lost to senescence. Eventually, a boy challenges the kingdom's resignation to the dragon's tyranny, igniting a movement to find a way to kill the dragon.

Technological advancements lead to the development of a missile capable of killing the dragon. Despite initial resistance from the king and his advisors, who worry about the social ramifications of killing the dragon, the beast is eventually slain, and the people celebrate their newfound freedom.

The fable teaches several moral lessons:

Acceptance of the "inevitable" can blind us to the suffering it causes. A static view of technology may cause us to underestimate the possibility of finding solutions to seemingly insurmountable problems. Society may prioritize damage control over addressing the root cause of an issue. Societal good must ultimately benefit individuals. A sense of proportion and priority is essential. Rhetoric can obscure the truth and hinder progress. Addressing urgent problems, like aging, should be a top priority. Societal reorganization and adaptation will be necessary after overcoming long-standing challenges. In conclusion, the fable urges society to invest in research to extend the human health-span, combat aging, and improve the quality of life, rather than merely accepting it as an inescapable fact.

AdGuard (the full, paid version for MacOS) is the way. I don't use Safari anymore, but I'm still happy enough with it to not install anything else for other browsers. Unfortunately, the iOS version only works as an extension for Safari.

ChatGPT is not 4 years old, it's a newcomer, give it some more time, ok? It's still developing, it's not a final product. There is already a huge difference between GPT-3.5 and GPT-4. Next iterations will be able to do more and more, especially if they get tools, self-reflection and vision.

ChatGPT Plugins 3 years ago

What's to stop you from asking it to give you a list of recommendations to choose from, based on your current preferences? The idea is that you ask what you want and you get it, without clicking and manually solving a task like checking website X, website Y, website Z, comparing all the different options, etc. They just want to show the basics of what's going on with these plugins, and then you can expand on it however you want.