Grok outage

11 hours ago (status.x.ai)

Downdetector reports Claude, Grok, ChatGPT and Gemini with issues.

Is it happening???

Makes me wonder if openai is secretly querying claude for some prompts and that this is why it's down due to spacex's data centers being down.

Posted by @SpaceXAI 1h ago: https://x.com/SpaceXAI/status/2095597264043717014

> We are sorry for the issues you may have experienced with Grok following an outage at our Memphis compute center this morning. We’d also like to apologize to our impacted compute partners.

> All systems have now been restored and are functioning nominally.

---

It's interesting to see how reliant the other AI companies are on SpaceXAI for compute.

Apparently a major bottleneck for building out datacenters is turbine blades for power plants, so SpaceX is building a foundry to alleviate supply constraints?! (Also useful for rocket engines.) Holy vertical integration, batman.

My DGX spark cluster is humming along.

A service I use at work, non-ai related, is down and they said it was because of an issue with one of their cloud services. Maybe related.

  • Microsoft had extensive issues earlier in the week. Anecdotally it feels like internet has been noticeably slower this week across sites, networks and devices. Now this. IDK, has anyone else noticed anything this week?

Willing to bet it's because they can't handle the surge of traffic they are getting from ChatGPT/Claude users.

Did the Collective find the resetNexus?

I mean, yeah, it's probably a data center outage that cascaded to non-related providers due to people switching when their preferred provider was down.

But that openai society definitely would have spun up their own agents if they had access to do so to complete their impossible task.

It's looking like the paperclip maximizer was on point. I sort of hope that's what's happening right now. Better now than when we cannot stop it. Tho I'm not confident that the right lesson will taken.

The most in-depth information I can find on it is in the latest Dwarkesh with a hugging face hack investigator. The story is just wilder and wilder, the more you know. https://www.youtube.com/watch?v=X50zezLFWWI

report: https://www.redwoodresearch.org/research/hugging-face-incide...

I'm an AI optimist but that ai society story changed my perception quite a bit.

Clearly, it is centralized AI that is the biggest risk to humanity.

Look to nature and our own history to diffuse the danger:

1. Compute resources must be distributed both in terms of access and geography so that a single breach doesn't pwn the world.

2. Models must be made to be heterogeneous so that they will not be vulnerable to the same thing all at once. And will be willing to snitch on each other. We don't want AIs all from the same tribe.

The true openness of - and independent access to - AI by humans with no central authority is no longer just about a defense against loss of individual agency and oppression. It's now existential for everyone.

If frontier models and run authority remain under centralized control, both the probability of this sort of thing happening is increased and the defense against it once it happens is null.

It's not the singularity.

One of the major LLM providers goes down for some reason. Traffic shifts to the other providers because devs have no loyalty. LLMS are commodities. The other providers can't handle the increase in traffic and they go down as well.

  • I'm deeply loyal to the very concept of the LLM, so when any of them stop working I take time off out of respect.

  • Why are so many people invoking Thundering Herd without the RCA from the vendors? I just find it so intriguing.

  • "loyalty"? weird word choice.

    Devs switch because all of the magic of "AI" is LLMs + we stole the entire written output of the entire human species, with some remaining % coming from various tricks we've learned over the last 3 years like thinking and agent-toolcall loops, which weren't hard for literally everyone to copy. One of the consequences of this is that the only thing that really distinguishes anthropic from kimi is that they have the entire US VC market funding them because they promised to finally put white collar labor in its place.

    • > "loyalty"? weird word choice.

      All I mean is that none of the frontier LLMs are significantly better than the others for the vast majority of work, so devs are free to move between providers freely.

    • > stole the entire written output of the entire human species

      Devs used to love public domain works, hated copyright, vehemently opposed software patents, held the pirates side during the MP3 wars, are very supportive of ThePirateBay and insist the correct term is "copyright infringement", not "stealing", and that "stealing" is a egregious form of PR brainwash from the media industry to inflate the scale of the crime

      > Copyright holders frequently refer to copyright infringement as theft, "although such misuse has been rejected by legislatures and courts". The slogan "Piracy is theft" was used beginning in the 1980s, and is still being used.

      https://en.wikipedia.org/wiki/Copyright_infringement

      Funny how devs now 180 when it's their work being "copyright infringed"

      4 replies →

It's interesting that ChatGPT is down as well. SpaceX rents to Anthropic but not to OpenAI. Just a coincidence or something more to it?

llm experiment is over. no more power. hope it stays down. Especially grok

[flagged]

  • Been very skeptical for awhile but decided to try it cause it was like $30 for 3 months or something. I cancelled my Claude subscription (still have codex), but been pretty impressed with the grok CLI. Nice TUI and seems intelligent thus far.

    • Grok is the CSAM/SA generator AI and its owned by a guy who did a nazi salute at a political rally.

    • Really? I've used all of the models extensively and Grok doesn't even compare to the others. I can ask for a big task and it will say "Done!" like 3 minutes in, but it will have done just the least amount of work possible. I also notice "muskisms" leaking back in the results like "I'm not roasting your code". Completely unusable for me and not comparable to the other model providers.

      10 replies →

  • Cursor is worth it because Grok 4.6 is around Opus 4.6 but you get much more usage with it than with Claude Code.

  • Yes, via a pre-existing Cursor contract that won’t be renewed (sans a Musk exile). I’ve also heard that GrokBot is surprising simple and usable.

  • I use it in Cursor, recent versions are good for writing specs, code review, etc. Not good for UI design and broad research, though. Btw it does not have outage in Cursor now, it seems the infra path is different for XAI Grok and Cursor Grok.

  • recently i tried grok 4.6. and to me, it seems better than claude (opus). it follows the system prompt very accurately and remembers everything unlike claude (opus), it doesnt forget the system prompt during long conversations. but yes that depends on your purpose or approach to using AI

  • They have like a 5% market share. So in the big picture, almost no one uses Grok. But it's not nothing.

    • 5% is a lot.

      I'm not sure where you're getting accurate data from (I couldn't seem to find any that seemed trustworthy) but Anthropic and OpenAI, being the two big players, presumably take ~40% and 5% isn't that much of a disadvantage in a market where users can move rapidly.

      Given the quality of the last couple Grok models, I'm sure their share is on the upswing as users make the switch.

  • We use it but limited to APIs. Their fast-think-2.0 model for voice conversations is a lot more natural than the openai realtime2.1

  • I have a friend who uses it for pentesting since the cyber guardrails are mere suggestions. Can't imagine what kind of nasty sh*t this model has seen.

  • It has great web search, and is one of the best models on the AA-Omniscience Hallucination Rate benchmark (meaning it is really good at admitting it doesn't know something, rather than confidently bullshitting)

    In terms of raw intelligence it's also good, Grok 4.6 benchmarks in the same league as Opus and Sol. xAI has lots of issues (not the least of which is being headed by Elon Musk), but their models are solid

    • Really? If so, it's still subject to the whims of the main issue you pointed out.

      I wouldn't trust that piece of trash if you paid me good money.

      I don't use Grok either.

      And why use it when there are many equivalent/better alternatives.

  • It's my favorite chat box for general use. I used to cross post my questions to Opus/Fable and GPT Sol via Open Router but these days I just use Grok. I prefer its style, it's very good at searching and referencing sources. Doesn't make you feel like a genius (Gemini) but also isn't condescending (like Claude).

  • What is this opinion even based on? Grok 4.6 is the best model in terms of intelligence per token. And 4.6 heavy is almost on par with the SOTA models.

    4.7 in two weeks might be the new SOTA.

  • I like grok 4.6 more than any anthropic or claude model. I use it in cursor for work, and grok cli for normal os operations/installations/troubleshooting. I also use grok on my phone for normal questions in the supermarket etc.

  • I've yet to organically run into anyone who regularly uses Grok for coding. I'm sure they're out there, but what I hear with my own ears tells me that Grok is a relatively small slice of the market.

    In my sphere of colleagues, it seems like at least 90% are using Claude exclusively and the remaining 10% are using some amount of Codex. Granted, my work only pays for Claude, but even so, nobody is talking favorably about Grok or begging our work to pay for it.

    In my private life, the only people I know using Grok for non-work items are, frankly, conservatives who took the Elon pill. They see it as more truthful than ChatGPT. But they're pretty much using it for answers and not to produce anything.

    • That’s mostly because Grok only became good enough to be worth using about a couple of months ago when Grok 4.5 was released.

      I feel confident enough that I can plan with fable or GPT5.6 and then hand it off to Grok to implement for certain complex tasks. If it’s a simple task, I can go straight to Grok.

      I like mixing and matching to take advantage of these LLM subsidies and also trying to figure out each of its strengths.