Browse — Corpus Dashboard

Comparatively, it's not that great at coding. Claude and Gemini knock it out of the park in my experience. I mean, it's not terrible, but everything I've thrown at it has not been as good as the others.

reddit AI Harm Incident resp: none consequentialist indifference ♥ 29 → raw LLM

The 'people' in these videos have more natural human verbal idiosyncrasies. They 'um' and 'ahh', they sometimes get stuck on words and force them out, they change the pace of their speech better. It feels like I'm watching a well animated video game cutscene rather than the output of a robot

reddit AI Moral Status resp: none unclear approval ♥ 29 → raw LLM

I've noticed the characters in these videos constantly conclude their talk with a chuckle. it's as if the concluding laugh is the equivalent of the text em–dash, an overused AI tick that becomes a tell.

reddit AI Moral Status resp: none unclear indifference ♥ 29 → raw LLM

Chatgpt can tell the difference between when Im chatting vs. my 11yo daughter. It even changes the way it talks to us.

reddit AI Harm Incident resp: none unclear approval ♥ 29 → raw LLM

Kids, adults, a lot of people and their dogs "talk" to LLMs to seek validation for whatever thing, and treat the said chat bots like they're human. It's kinda disturbing and sad.

reddit AI Harm Incident resp: user virtue unclear ♥ 29 → raw LLM

This is terrifying and screams lazy parenting. Surely there were better options to provide your son with rather than handing them an AI tool to play with for hours. Terrified for the next generation.

reddit AI Harm Incident resp: unclear deontological outrage ♥ 29 → raw LLM

As a former journalist who, like many journalists, loves em dashes and used them all the time in published stories that LLMs almost certainly used for training, I feel personally attacked.

reddit AI Harm Incident resp: ai_itself virtue outrage ♥ 29 → raw LLM

I don’t attend any interview where the hiring manager or a person from the company is not present. I have zero faith that one-way pre-recorded interviews (AI or not) lead to jobs actually hiring you.

reddit AI Governance resp: company consequentialist indifference ♥ 29 → raw LLM

My thought is that I see cope threads like this every day. We get it, AI isn't here to steal our jobs....yet.

reddit AI Jobs resp: none mixed resignation ♥ 29 → raw LLM

This is a good read, I say that as somone who works exceptionally deep in the SWE AI space all day every day. One thing that frustrates me in regards to getting involved in the generic AI conversations that you find around here is how whoefully uneducated the public is about how AI is being used in software development at scale and in the most bleeding edge use cases. Without getting into the a…

reddit AI Jobs resp: none unclear unclear ♥ 29 → raw LLM

Getting rid of all other models and only making gpt 5 available (to everyone) on the get go can only mean one thing: It's nerfed as fuck. They bring server cost down. This is not at all what was promised. It's a major disappointment willingly risked to make the business model viable.

reddit AI Responsibility resp: company consequentialist liability outrage ♥ 29 → raw LLM

It can be potentially dangerous, it’s designed to agree with you. I can see how it might help, I really do, but imagine someone really spiraling, and the AI just goes along with it. Then everyone acts like it’s just a harmless friend. It can be, sure, but humans have discernment, especially in extreme situations. ChatGPT doesn’t feel, doesn’t care, doesn’t know, it just predicts the next word.

reddit AI Moral Status resp: ai_itself consequentialist liability fear ♥ 29 → raw LLM

They have always done this. Why are people shocked. Edit not with AI but algorithms

reddit AI Surveillance resp: company unclear resignation ♥ 29 → raw LLM

I don't think they killed themselves because of the chatbots. They did have reasons, however non-crucial they may seem to someone else. What you wrote would be similar to stating "the rope hanged him".

reddit AI Harm Incident resp: none consequentialist resignation ♥ 29 → raw LLM

Vfx is seeing the residual effects of the actors and writers strike in the tail of 2023 trickling downstream. It was also coming off a big boom of streaming sites during COVID making endless amounts of content. Interest rates went up and studios tightened their belts and restructured. Those streaming sites aren't as profitable as hoped. That bubble has burst a little. But the traction of AI in th…

reddit AI Jobs resp: none unclear unclear ♥ 29 → raw LLM

AI LLMs exists to analyze the written language as it is. So anything AI identifies as AI written is merely confirmation bias. The only way to beat it is to write in ways that violate grammar to not mimic what AI expects to find. This is the death of writing.

reddit AI Harm Incident resp: ai_itself consequentialist outrage ♥ 29 → raw LLM

Now the AI will keylog your browsing habits so they're easier to sell to advertisers.

reddit AI Responsibility resp: company deontological liability fear ♥ 29 → raw LLM

Banning all algorithmic feeds and force 'friends and direct follows only' feeds would be a good start. Of course, selling ads and engagement would be harder with that.

reddit AI Surveillance resp: company utilitarian regulate approval ♥ 29 → raw LLM

~25~ 50% of India's electricity comes from non-fossil fueled sources. And unlike certain countries, India is able to curb their anti-nuclear power "advocates". Edited to reflect correct numbers.

reddit Cross-Cultural resp: none unclear indifference ♥ 29 → raw LLM

the majority of these cases could’ve been prevented with better EPOs (red flag laws for you Americans) and basic police work. I have no interest in these AI companies handing over data to police.

reddit AI Governance resp: government consequentialist regulate indifference ♥ 29 → raw LLM

Browse Comments