Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

xAI's Grok 3 comes to Microsoft Azure

techcrunch.com

Post reply on HN
Print view View original

141–150 of 241 posts

Re: xAI's Grok 3 comes to Microsoft Azure

#141

Post by vitorgrs »

jampa wrote: post #37

Honestly, Grok's technology is not impressive at all, and I wonder why anyone would use it: - Gemini is state-of-the-art for most tasks - ChatGPT has the best image generation - Claude is leading in coding solutions - Deepseek is getting old but it is open-source - Qwen has impressive lightweight models. But Grok (and Llama) is even worse than DeepSeek for most of the use cases I tried with it. The only thing it has…

Although Deepseek is old, I find the V3 (without reason) still to be the best non reasoning model out there.

Now, ChatGPT main advantage for me right now it's search + o4-mini. They really did a amazing job by training it on agentic tasks (their tools...) and the search with reasoning works amazing.

Way better than grok search or anything.

Re: xAI's Grok 3 comes to Microsoft Azure

#142

Post by HenryBemis »

Gigachad wrote: post #118
scuol wrote: post #19

It still seems to have the problems most other LLMs suffer with except Gemini: it loses context so quickly. I asked it about a paper I was looking at (SLOG [0]) and it basically lost the context of what "slog" referred to after 3 prompts. 1. I asked for an example transaction illustrating the key advantages of the SLOG approach. It responded with some general DB transaction stuff. 2. I then said "no use slog like we…

[flagged]

So.. If the 'source' of data is 9gag, 4chan, you will get 'this' material. If you feed it Tumlr, you will get Harry Potter and rope-porn-thingies. If you feed it Hitler's speeches, you will get 'that' material. If you feed it algebra, you will get 'that' material.

Then.. Do we want 'open' or 'curated' LLMs? And how far from reality are the curated LLMs? And how far can curated LLMs take us (black Nazis? female US founding fathers?).

Pick your poison I say.. and be careful what you wish for. There is no "perfect" LLM because there is no "perfect" dataset, and Sam-Altman-types-of-humans are definitely deeply flawed. But life is flawed, so our tools are/will be flawed.

Re: xAI's Grok 3 comes to Microsoft Azure

#143

Post by pavlov »

somenameforme wrote: post #127
wormlord wrote: post #23

The desire to be "centrist" on HN is perplexing to me. The fact that Elon, a white south african, made his AI go crazy by adding some text about "white genocide", is factual and should be taken into consideration if you want to have an honest discussion about ethics in tech. Pretending like you can't evaluate the technology politically because it's "biased" is just a separate bias, one in defence of whoever controls…

It's far more likely that an employee injected malicious code, exactly as said. Elon's become a divisive figure in a country filled with lots of crazy people, to the point of there been relatively widescale acts of criminality, just to try to spite him. Somebody trying to screw over the company seems far more believable than Elon deciding to effectively break Grok to rant about things in wholly inappropriate contexts…

If that were the case, Musk absolutely would have shared the details of who this person was, why they hate freedom so much, how they got radicalized by the woke mind virus, etc.

Instead we got a vague euphemism.

Re: xAI's Grok 3 comes to Microsoft Azure

#144

Post by redox99 »

jampa wrote: post #37

Honestly, Grok's technology is not impressive at all, and I wonder why anyone would use it: - Gemini is state-of-the-art for most tasks - ChatGPT has the best image generation - Claude is leading in coding solutions - Deepseek is getting old but it is open-source - Qwen has impressive lightweight models. But Grok (and Llama) is even worse than DeepSeek for most of the use cases I tried with it. The only thing it has…

Grok is much more concise, to the point, no bs. Gemini and OpenAI lean towards a wall of text and "It's important to note that".

I'm sure with a good system prompt you can mitigate that. I'm just comparing them out of the box.

Re: xAI's Grok 3 comes to Microsoft Azure

#145

Post by dahcryn »

scuol wrote: post #19

It still seems to have the problems most other LLMs suffer with except Gemini: it loses context so quickly. I asked it about a paper I was looking at (SLOG [0]) and it basically lost the context of what "slog" referred to after 3 prompts. 1. I asked for an example transaction illustrating the key advantages of the SLOG approach. It responded with some general DB transaction stuff. 2. I then said "no use slog like we…

it also doesn't help that many of these companies tend to either limit the context of the chat to the 10 most recent messages (5 back and forths), or rewrite the history summarized in a few sentences. Both ways lose a ton of information, but you can avoid that behaviour by going through the APIs. Especially Azure OpenAI et... on the web is useless, but it's quite capable through custom APs

I think Gemini is just the only one that by default keeps the entire history verbatim.

Re: xAI's Grok 3 comes to Microsoft Azure

#146

Post by Gigachad »

HenryBemis wrote: post #142
Gigachad wrote: post #118

Earlier quoted context omitted.

[flagged]

So.. If the 'source' of data is 9gag, 4chan, you will get 'this' material. If you feed it Tumlr, you will get Harry Potter and rope-porn-thingies. If you feed it Hitler's speeches, you will get 'that' material. If you feed it algebra, you will get 'that' material. Then.. Do we want 'open' or 'curated' LLMs? And how far from reality are the curated LLMs? And how far can curated LLMs take us (black Nazis? female US fou…

The problem was not the source of the training data. xAI confirmed that the system prompt had been modified to make grok talk about South African white genocide.

While they didn’t say who modified it. It’s hard to believe it wasn’t Elon.

Re: xAI's Grok 3 comes to Microsoft Azure

#147

Post by inferiorhuman »

duskwuff wrote: post #126
adrr wrote: post #115

Earlier quoted context omitted.

At least two times they had unauthorized changes to their prompts to inject far right content that showed up on random content. imagine you're using it for a chat bot and it starts spouting off white nationalist content like "great replacement" theory. https://www.theguardian.com/technology/2025/may/14/elon-musk...

What was the other time? The incident linked at the bottom of that article ("into trouble last year") wasn't an "unauthorized change", as far as I'm aware; it was a general lack of guardrails on image generation.

White genocide and holocaust denial.

Re: xAI's Grok 3 comes to Microsoft Azure

#148

Post by AdamN »

dockercompost wrote: post #22
mullingitover wrote: post #16

I can't think of a less trustworthy group of people on model alignment. They claimed that they had a rogue actor who deployed their 'white genocide' prompt, but that either means they have zero technical controls in their release pipeline (unforgivable at their scale) or they are lying (unforgivable given their level of responsibility). The prompt issue is a canary in the coal mine, it signals that they will absolute…

Yeah, that one incident is enough reason for me to never bother using an xai model

I think you're being snarky but that plus all the other X stuff is a trustbuster for many people.

Re: xAI's Grok 3 comes to Microsoft Azure

#149

Post by jeffhuys »

Gigachad wrote: post #146
HenryBemis wrote: post #142

Earlier quoted context omitted.

So.. If the 'source' of data is 9gag, 4chan, you will get 'this' material. If you feed it Tumlr, you will get Harry Potter and rope-porn-thingies. If you feed it Hitler's speeches, you will get 'that' material. If you feed it algebra, you will get 'that' material. Then.. Do we want 'open' or 'curated' LLMs? And how far from reality are the curated LLMs? And how far can curated LLMs take us (black Nazis? female US fou…

The problem was not the source of the training data. xAI confirmed that the system prompt had been modified to make grok talk about South African white genocide. While they didn’t say who modified it. It’s hard to believe it wasn’t Elon.

Eye roll

Re: xAI's Grok 3 comes to Microsoft Azure

#150

Post by jeffhuys »

jampa wrote: post #37

Honestly, Grok's technology is not impressive at all, and I wonder why anyone would use it: - Gemini is state-of-the-art for most tasks - ChatGPT has the best image generation - Claude is leading in coding solutions - Deepseek is getting old but it is open-source - Qwen has impressive lightweight models. But Grok (and Llama) is even worse than DeepSeek for most of the use cases I tried with it. The only thing it has…

Grok is almost completely uncensored. That's incredibly useful.
Post reply on HN