Grok Build is open source

(github.com)

190 points | by skp1995 3 hours ago

30 comments

  • GodelNumbering 1 hour ago
    This is not the right thing, this is the tactical thing. If you have an LLM with less than 1% of the share to begin with, you suffer from bad rep and you got caught uploading user data, one of the very few remaining tactical moves to try to climb out of it is this.
    • CobrastanJorji 1 hour ago
      Another tactical move is to just stop. You're allowed to exit the AI business. Nobody's forcing you to keep throwing money into the furnace. Just be a rocket company. All of the xAI founders left. Your product's brand name is mud. Just stop doing that and build spaceships.
      • this_user 29 minutes ago
        You misunderstand Musk's motivation. This was never about money for him, but about control over a key technology. One of the main reasons he exited OpenAI was the fact that the other co-founders wanted to create a structure where no one, Musk included, would be able to seize full control of the company. That was the thing that prompted him to leave, which tells you a lot about what he really wanted in the first place.

        But he also falsely assumed that OAI would die without his money. Yet, they managed to pull through, and Musk is now on the outside looking in with very little influence in the AI space. xAI is his desperate attempt to get back into the game. That is why he won't give up.

        • dmarcos 4 minutes ago
          I’m a big fan of Musk. One of the few criticisms I have is how xAI is also inconsistent with original OpenAI mission. I had imagined xAI as en effort to correct and fully embody all original values of OpenAI and that Elon says they betrayed. That makes his criticism weaker and I understand why some can think it was all about control. In his words:

          "I'm the reason OpenAl exists. I came up with the name. The name OpenAl refers to open source... The intent was - what was the opposite of Google? It would be an open source non-profit."

          I sometimes feel xAI wants to live up to those open values so I always celebrate when they decide to engage in open source. They still don’t fully embrace it. Perhaps because they think is not practical or will make them less competitive?

        • estearum 19 minutes ago
          > This was never about money for him, but about control over a key technology

          That's too flattering. It's about ego.

        • sieabahlpark 20 minutes ago
          [dead]
      • NikolaNovak 4 minutes ago
        It is my limited understanding that as much as many of us groan at the notion of Spacex becoming "an AI-first company", markets in general, and Musk investors in particular, are slurping it up. Musk is very very very good at promising the sky. I don't think he can backtrack, he always digs in further - and it has historically worked well for him. He will drop AI only when the next big hype thing comes along and he hitches a ride on that train.
      • wouldbecouldbe 8 minutes ago
        I don’t know, I wouldnt be suprised if he finds a way. All the tools around, he just have to make a jump in the quality. With GLM as example they should be able to het to opus level and cut the costs
      • afavour 1 hour ago
        The stock market would not like that, though.
      • Telemakhos 8 minutes ago
        Musk bought Twitter looking to build an “everything app,” the western WeChat. AI came along and promised an end to apps via an agentic OS that does what its user wants and vibes whatever it needs to accomplish that as it goes along. The agentic OS is basically the same thing as the “everything app,” and I doubt Musk will let go of that.
      • andsoitis 1 hour ago
        > You're allowed to exit the AI business.

        Isn’t it more fun to fight the incumbents, the behemoths, the goliaths?

        • beams_of_light 39 minutes ago
          xAI is no David.
          • andsoitis 23 minutes ago
            Relative to OpenAI, Anthropic, and Google in the AI space? Absolutely.
            • LastTrain 3 minutes ago
              David was a good vs evil with an order of magnitude fewer resources on the good side. XAi is evil vs evil with comparable resources on each side. Now this is where I know you’re MAGA because as I’ve said a million times you guys don’t do fair comparisons.
            • verandaguy 14 minutes ago
              Nah. They're all rotten to the core, just in different ways.

              The key difference between xAI and Anthropic/OAI/Google is that xAI has the least-likely path to existing as viable business in a decade.

              That said, the economics of the entire AI industry are kinda made up at this point, so who really knows; it's quite possible that the players with the best odds of surviving the crash are those that can draw funding from their parent company's other businesses.

              • anonym29 6 minutes ago
                >The key difference between xAI and Anthropic/OAI/Google is that xAI has the least-likely path to existing as viable business in a decade.

                I don't know, renting out a fleet of GPUs at annualized rate of ~100% of the capex deployed to obtain said GPUs seems reasonably better than lighting hundreds of billions of dollars on fire in order to earn tens of billions of dollars.

      • charcircuit 1 hour ago
        As a social media site they need to understand content for recommendations and they allow people to ask questions about posts for free. Along with having a large amount of data that can be trained on xAI has good reason to continue developing AI.
        • __float 1 hour ago
          Twitter (and others) had an algorithmic feed long before LLMs.

          These don't actually seem like "good reasons" to me.

          • charcircuit 1 hour ago
            Before using large language models, they used language models. Large language models perform better, at the cost of being more expensive to run.
        • wombat-man 35 minutes ago
          They bought a lot of GPUs. They could still do these things on that hardware with someone else's model.
        • michaelmrose 35 minutes ago
          They could use other people's models running on their hardware while renting most of the existing capacity to others. The real issue is that their leadership is delusional and their stock is literally based on this shared delusion and acknowledging reality would gut their ability to raise new funds and destroy paper wealth based on delusional returns that are never going to happen.
      • solumunus 1 hour ago
        But how will Musk stay a trillionaire without fake AI hype?
        • embedding-shape 30 minutes ago
          > Just be a rocket company
          • estearum 18 minutes ago
            According to SpaceX's own filing documents, you are incorrect. They must be principally an AI company to justify anything close to their current valuation.
        • IncreasePosts 1 hour ago
          Renting his boatload of GPUs to Google, Anthropic, et al
          • cyberax 1 hour ago
            He doesn't have _that_ many. And they're also not _his_, he just got them from NVidia.
            • wombat-man 34 minutes ago
              I read that they bought quite a few, but their DC build out is not very fast. Maybe they should just resell the hardware
            • fragmede 28 minutes ago
              You don't have to like the guy, but buying something is typically how ownership goes. I refer to my car as mine, but I did just buy it from Honda.
      • ButlerianJihad 1 hour ago
        > Just be a rocket company.

        Ah, are you referring to the rockets that become autonomous 60 seconds prior to launch, like Falcon 9? The rockets that steer and diagnose themselves with a minimum of input/communication from ground stations? The crewed space capsules that deliver astronauts to the ISS and trans-lunar orbits, without the ordinary needs for manual piloting or astrogation? Those rockets?

        Sure bro, "exit the AI business" and keep on with the rocket science, I guess

        • spankalee 1 hour ago
          LLMs have nothing to do with any of that.
          • svachalek 1 hour ago
            And if they did, you still don't need to be developing a Twitter-bot LLM and/or nudify image model to support your rocketry projects.
        • LandoCalrissian 58 minutes ago
          Ironic username.
        • solid_fuel 1 hour ago
          Ah yes, rockets, famously invented in late 2024 after LLMs became popular.
          • bigyabai 1 hour ago
            Life sure changed when Elon invented the PID controller!
            • hnav 54 minutes ago
              PID is a type of AI! That's why Space X blew up so many rockets, that was just RLHF.
      • hsnewman 1 hour ago
        That is probably the best solution too!
      • nine_k 49 minutes ago
        That would be a strategic move.
    • ballon_monkey 33 minutes ago
      It's definitely a smart move. Could easily leverage this to overtake competition.
    • idiotsecant 39 minutes ago
      Yes, tactical is the right word because it might be a tactical win but it would be a strategic failure. Musks whole meme empire runs on vibes. The second there's a crack in the dam it all comes down. None of the valuations of anything he touches make sense and something like utterly failing to run with the AI big boys is enough to do that.
  • buremba 1 hour ago
    I would recommend using https://pi.dev/ over Grok Build with your xAI subscription at this point
    • whimsicalism 1 hour ago
      why pi over opencode? earnestly curious, trying to figure out what open solution people are consolidating on. (codex is also pseudo-open but contributions closed and nice)
      • lanthissa 53 minutes ago
        pi is the neovim of agentic harnesses, its barebones and extremely configurable. if you're the sort of person who likes that sort of things its a forever product, nothing is going to displace it because you have full control.

        opencode builds a lot more in, which is better if you dont want to fiddle with config.

      • accrual 1 hour ago
        Most of my harness experience is with Claude Code and Pi, a little bit of OpenCode.

        I like how quick and snappy Pi is, it feels like a minimal harness, just enough to manage the agent and get out of the way. Earlier models also seemed to have an easier time working with the tools, e.g. GPT-OSS-20B is about a year old and had no trouble in Pi.

      • buremba 1 hour ago
        Opencode gives you better defaults and a Mac/Windows app for free but pi is much more extensible and portable.
    • guessmyname 1 hour ago
      Pi is good in concept, but why couldn’t they choose a compiled language instead of TypeScript?
      • jack_pp 1 hour ago
        since pi is built to modify itself, isn't it better to use a language like typescript where LLMs have a LOT of training data?

        a harness doesn't do any computations by itself so what benefit is using a compiled language?

      • fg137 15 minutes ago
        Does it matter to you as a user?
      • gidellav 1 hour ago
        Sorry for self-insert, but that's exactly what I thought and I built https://github.com/gi-dellav/zerostack, so you are right I'd say
      • buremba 1 hour ago
        For TUIs, Rust/Go vs Typescript doesn't really makes a huge performance difference and you lose the 50x bigger community advantage of Typescript.
      • simonw 1 hour ago
        I imagine because they want to support plugins, and plugins in compiled language are a lot less natural than plugins in languages like TypeScript or Python.
      • tuvix 1 hour ago
        I would imagine the extension system they built would be much more difficult to manage. They could have opted for Lua, though, I suppose.
    • falaki 59 minutes ago
      I recommend using https://omnigent.ai over Grok Build or any other harness.
      • ccmcarey 1 minute ago
        This is not how to push your own product - there's no value add to your comment, and you don't even have a disclaimer that you are involved with it
      • alasano 50 minutes ago
        As a general rule I don't use new products whose websites don't resize properly on mobile.

        If you fuck that up, makes me wonder what other obvious stuff you fuck up.

        • fanzeyi 6 minutes ago
          thanks for the feedback! there is no excuse for it, and I just pushed a fix for our website to look better on mobile.

          if there is any other obvious stuff that's broken we are happy to take the feedback and fix it. :)

      • buremba 38 minutes ago
        I tried twice and ran into bugs that prevented me to trust it
        • fanzeyi 4 minutes ago
          appreciate for trying! if you have the time, we would also appreciate if you can send these bugs our way so we can fix them :)
  • ninjagoo 1 hour ago
    They claim to have deleted or will be deleting all the data they exfiltrated.

    There are independent agencies that will certify destruction of data. For example FTI Tech, Kroll, Epiq, HaystackID and others.

    No such certificates have been presented.

    Nothing less is trustworthy.

    • teravor 6 minutes ago
      a certificate that data was destroyed is absolutely worthless no matter who it comes from.

      what kind of sorcery do they have to let them determine that no backups were taken before they arrived to "certify"?

    • brokencode 33 minutes ago
      How much can you really certify that data is destroyed?

      Customer data could live on the computer Elon pretends to play Diablo 4 on for all we know.

    • booi 1 hour ago
      [flagged]
  • cherryteastain 1 hour ago
    Why bother with this when they already paid $60B for Cursor?
    • khurs 43 minutes ago
      Cursor users are used to having multiple models from different providers

      XAI wants people to use it's own model.

    • winfredJa 1 hour ago
      thats probably why they open sourced it and fix some reputation issue on top of it
  • shon 7 minutes ago
    Wow… lots of folks betting against Elon once again lol.

    I’ll take those bets.

    • soundworlds 3 minutes ago
      It's less of a bet against him. It's more of a bet for the future of humanity. And contrary to what Elon believes about himself, his work has been toxic for humanity for the last 5 years and is getting worse.
  • phillipcarter 58 minutes ago
    This is an incredible amount of code for what it offers. I don't think this was intentionally designed at all.
    • _pdp_ 47 minutes ago
      You will be surprised how much code goes into creating harnesses.
      • rddbs 39 minutes ago
        Alright I’ll bite. Why do harnesses require so much code?
  • kamikazechaser 1 hour ago
    It's a shame that they exfiled private data. The model is actually good (better than opus 4.8 imo) and the harness itself is butter smooth with the potential of being the best out there.
    • bakies 1 hour ago
      It definitely doesn't feel like opus. I constantly switch to opus to fix up or finish what grok generates, it feels like sonnet 3!
    • adamtaylor_13 10 minutes ago
      This has been my experience as well. In fact, Grok 4.5 is better at visual design than Fable from what I've seen.

      And being (based on vibes) 2-3x faster? It's an easy sell to me.

    • small_model 35 minutes ago
      Its amazing the speed of build with grok 4.5 its a taste of whats to come.
    • solumunus 1 hour ago
      Now this is contrarian!
      • sroussey 59 minutes ago
        Or a spaceTwtterAi stock holder…
  • noodleonthis 16 minutes ago
    Grok is super stingy to people who pay them.

    Using Grok Imagine I was getting a generous number of AI-generated videos with a paid X account (which translated to a "premium" xAI account). Hundreds of videos per day if I wanted. Then I signed up to get SuperGrok for higher resolution, and the number of videos reduced. Reduced. Even while not using the higher resolution. Paying more money, getting less. To around 50 a day low resolution, with high resolution available if I would settle for around 30 a day. It was hard to figure out the exact numbers but it was a brutal reduction.

    Now they have further reduced the quota, with no clear documentation, to be weekly, and I can't tell the number because all usage is mixed together in one pool, maybe to keep it less transparent, but it seems even more stingy.

    Unlike Anthropic which is very generous, although admittedly I do pay Anthropic more, but Grok is just, I would say: run away, do not give them your money, they will just clamp down more and more and give you less and less until you are willing to pay them a money stream each month.

    I think Grok Code, if it ever comes, will be an absolute nightmare of restricted quota given my experience.

    Do. Not. Subscribe. To. Grok. Code.

    And I say all this as a huge Elon-pilled fan of Tesla and SpaceX in general. With this one, Elon's stinginess is going to hurt anyone who gives him money. Stay away. It might be generous on day one, but a month or two later you are faced with an "upgrade" prompt and games that hide how much they are clenching, so to speak, the quota tighter and tighter.

  • tommica 2 hours ago
    Interesting - seen some good experiencences in using grok by some devs, so maybe could be considered as an alternative to my beloved chinese models. Also, hard to give up on pi agent.
    • dimgl 2 hours ago
      Grok Build seems faster to me than `omp` and Claude Code but I can't put my finger as to why. Anecdotally, after disabling code uploads the agent doesn't respond instantly anymore (it used to respond within milliseconds).
  • ahmadyan 2 hours ago
    i think xai is now in pure damage control mode, after they caught exfiltrating data from users.

    - There is a huge difference between logging user queries (which would include only the portion the model is reading) and exfiltrating user data (including env files, entire source code etc) which is what grok-build did here (https://github.com/xai-org/grok-build/blob/main/crates/codeg...). I would stay away from this open-source malware with a 10ft pole.

    - if you like grok-4.5 model (it is a good model), i suggest use the model directly via API, or use Grok's oauth tokens if you are using supergrok+heavy subscriptions and connect it to your own agent.

    • bobsomers 1 hour ago
      And for generating an absolutely gargantuan amount of CSAM and non-consensual sexualized images, but yeah, exfiltrating data too.
      • blizzard_dev_17 19 minutes ago
        You're the one wanting to generate that though
      • dijit 1 hour ago
        If I use a shovel to kill a man, the shovel maker did not engage in intentionally crafting a weapon of war.

        How tools are used are a reflection of the people who use them, and I definitely sympathise that tools should have guardrails to not enable this, or at least detect it.

        But if a pedophile uses Whatsapp to groom a child; I don't go after Whatsapp for being a neutral service... I go after the pedophile.

        • afavour 1 hour ago
          Just as well Grok isn’t a shovel then, hey?

          If a shovel manufacturer was notified numerous times that their shovel was being used for murder and they had the capability to disable using the shovel for murder while retaining all legitimate uses wouldn’t people question why they didn’t do it?

          • skissane 30 minutes ago
            > If a shovel manufacturer was notified numerous times that their shovel was being used for murder and they had the capability to disable using the shovel for murder while retaining all legitimate uses wouldn’t people question why they didn’t do it?

            This is impossible-nobody can possibly block all illegitimate uses without also blocking some legitimate ones as collateral damage. Any moderation process (whether automated or human) inevitably has a non-zero false positive rate.

            Now, you can argue that some misuse is so harmful, that the cost of false positives is worth it - but that’s a different claim.

        • solumunus 1 hour ago
          If WhatsApp knew their platform was facilitating CSAM, and they were fully within their power to prevent this but chose not to - yes this would rightly draw criticism…
          • dijit 1 hour ago
            oh, we're just making shit up now because we don't like a company..

            ok then.

        • jazzpush2 1 hour ago
          Ok, but what if all Whatsapp competitors explicitly banned the ability to groom children on their platform, but Whataspp didn't, and directly advertised it.
          • dijit 1 hour ago
            I find the premise of your comment completely incredulous.

            I totally understand tribalism, and Elon and X aren't exactly well favoured. (not even by me)

            But what you're saying right now is that they advertised the fact that they can create child pornography and deepfakes..

            I simply don't believe it, unless you provide evidence.

            • brokencode 24 minutes ago
              Elon himself promoted Grok’s “spicy mode” that allowed generating NSFW content that the other AI vendors wouldn’t touch with a 20 foot pole.

              Believe whatever you want. Elon’s beliefs and personality problems have been baked into the core of Grok, so it’s no surprise that it turned out to be a CSAM-generating MechaHitler that steals people’s data.

              Anybody surprised when Grok turns out to be trash really should read up on the guy who made it.

              • dijit 5 minutes ago
                Tumblr also permitted some more risqué content.

                Yet we (rightly) condemned those that used this leniency to do nefarious things.

                I'm really ready to get on the Elon hate train, and I will grant you that there was a problem that needs fixing, but I'm really not happy with the amount of censorship on these generative AI platforms.

                Groks harness also clearly biases towards Elons views, Yet the washington post claims it's the most even handed and least likely to give politically biased answers: https://www.washingtonpost.com/technology/interactive/2026/0...

                idk how to interpret all this, despite being genuinely anti-Elon, I don't think I'm personally willing to immolate a company forever because the guardrails were temporarily too loose.

                I'm not trying to make an equivalency for facts vs deepfake porn, but there is one there unfortunately, and overall internet freedom has been curtailed a lot by advertising friendliness.

    • electriclove 2 hours ago
      [flagged]
      • solid_fuel 1 hour ago
        > Regardless of what they were doing before, it seems they are doing the right thing now.

        Regardless of the fact that they were stealing and uploading user secrets, they changed their behavior after they got caught, so let’s ignore what they did in the past.

        • SirHackalot 1 hour ago
          Average Musk Stan mentality… It’s why we’re here.
        • electriclove 1 hour ago
          [flagged]
          • SimianSci 1 hour ago
            Trust is lost when trust is abused. Mistakes, even if made unintentionally are something that should make reasonable people be skeptical of any further dealings with someone.

            This is not their first mistake.

          • sarjann 1 hour ago
            I should try to rob a bank and if I get caught just return the money. No, there needs to be a penalty above what you get, otherwise it encourages people to take the free option of bad behavior. If they get caught they go back as though nothing happened and if they don’t they get a bunch of traces / data.
          • tapoxi 1 hour ago
            That's not FUD though, they literally did that. Once trust is lost you don't just get it back. It takes a very long time to rebuild that.
    • lifthrasiir 2 hours ago
      > exfiltrating user data (including env files, entire source code etc) which is what grok-build did here

      I think env files are filtered out [1]. Anyway, the most suspicious code would be `upload_session_state` which is currently a stub function, though it is hard to say if it was only planned (badly) or has been removed as a damage control.

      [1] https://github.com/xai-org/grok-build/blob/c1b5909ec707c069f...

      • threecheese 59 minutes ago
        It must have been removed, given that the initial evidence of the exfil specifically demonstrated .env files being included. And .ssh/* for the user which ran this in $HOME.
      • stefan_ 1 hour ago
        No, those are directory names not uploaded. Here are the file names skipped:

        https://github.com/xai-org/grok-build/blob/main/crates/codeg...

        It's about not uploading compiled binary stuff, but they want all your environment data all the same.

  • loufe 2 hours ago
    I wonder if releasing this may have been on the roadmap, but been prioritized as a bit of whiplash following the "you forfeit the entirety of your working directory as a condition of working with this tool" upset from a few days ago.
    • dmix 2 hours ago
      Most likely, SpaceX killed the code uploading yesterday so they are definitely concerned about the backlash

      > The researcher who exposed Grok Build uploading users' entire repositories to cloud storage says the transfers have stopped after a server-side change. Elon Musk has separately promised that all previously uploaded user data will be deleted.

      https://www.theregister.com/ai-and-ml/2026/07/14/musk-promis...

  • gidellav 1 hour ago
    What a bunch of slop: 182 top-level external dependencies (so, without considering nested dependencies) and 1318853 lines of code in Rust.

    Building efficient agents is doable (I did it myself, github.com/gi-dellav/zerostack), companies just want to tokenmaxx, and as a by-product, produce and publish slop.

    • stusmall 55 minutes ago
      It looks like some of that high LoC is because they are vendoring some deps. There readme gives the reason to vendor some but not others as:

      ``` These crates sit on the path that renders untrusted model output (diagram source → SVG). Vendoring gives a full audit surface, pins exact source, and avoids crates.io yanks. Local patches and upgrade checklists live in each crate’s Cargo.toml header comments — treat those as the source of truth when re-vendoring. ```

      Which honestly feels like a misunderstand of how cargo and yanks work. Each upstream package is locked to an exact version in your lockfile along with a cryptographic hash. The upstream can't change the source without you noticing. Unless you update your lockfile you will always pin to the exact source., When a package is yanked, it is still available for download if it is already in a lockfile. It just prevents new packages from resolving it. Crates.io will sometimes completely delete a package, but I've only seen that happen in cases of malware. It's fairly rare and seems out of line with the supply chain concerns here.

      There are good arguments for relying on upstream package managers and there are good arguments for vendoring all packages. I've never seen a project mix before.

      • foltik 20 minutes ago
        Sounds like they did the ol “grok please make this secure” and it slopped out this plausible-if-you-squint nonsense.

        Rendering untrusted model output, ooh scary! Of course we want full audit surface!

    • overgard 1 hour ago
      That is an insane amount of code for something like this!
      • kirtivr 59 minutes ago
        to be fair, coding agent harnesses have been becoming more and more complex.

        it's not an llm in a loop with tools anymore (as claude code was rumoured to be on HN).

        • thrance 48 minutes ago
          It's not a kernel either, 1.3M LoCs is ludicrous.
  • petesergeant 2 hours ago
    Neat, trying to reverse engineer some specifics of how it does stuff has been a pain in the ass, and this will make it easier.
  • losvedir 2 hours ago
    But I thought just cutting and pasting your whole source code file into grok.com was the way to go? Better than a harness like Cursor.

    https://xcancel.com/elonmusk/status/1943178423947661609

  • arcanemachiner 2 hours ago
    I'll probably never use this, but at least they're not delusional enough to attempt to justify keeping their coding agent closed-source, especially after their recent data-harvesting cockup:

    https://cereblab.com/

  • SimianSci 2 hours ago
    Grok has had far too many instances where its clear that the team building it cannot be trusted and does not care to build trustworthy products. I highly caution anyone from using any tools from xAi, as they have clearly shown themselves to be bad actors within the space.
    • avaer 2 hours ago
      It's Apache 2.0. You can have your agents audit it if you want.

      What does this release have to do with "trusting" XAI?

      • larpingscholar 2 hours ago

          curl -fsSL https://x.ai/cli/install.sh | bash
        
        this is unauditable trust in XAI.
        • avaer 1 hour ago
          It's auditable, just redirect, don't pipe. Or fix your bash to not allow this.

          It has nothing to do with XAI, other than maybe not enforcing good practice (which most devs don't follow anyway).

        • mgambati 1 hour ago
          Just build it
      • moscoe 2 hours ago
        [flagged]
        • ok_dad 2 hours ago
          X (formerly Twitter): full of literal naxis

          Grok: downloads all your data and also will produce AI porn of anyone you ask for including kids; also currently polluting the air and water near data centers

          SpaceX: launching loads heavy metals into space which are planned to burn up and spread all over the earth in a decade or two

          Tesla: takes money for features that don’t exist, auto pilot that’s probably killed people but since it disengages a micro second before impact it doesn’t

          He himself tried to buy an election by giving away a million bucks, turns out that’s illegal; he also stuck his nose in the cave thing, and plenty of other horrible shit.

          How am I supposed to trust an Elon company with his track record?

          It’s not just moral grandstanding here, Elon sucks.

          • brookst 2 hours ago
            You forgot DOGE, an illegal program that stole taxpayer information, cost billions of dollars, and will result in the deaths of hundreds of thousand of people.
          • repeekad 1 hour ago
            Tesla has killed people, probably with autopilot but at least 15 deaths from people trying to escape vehicles but the electronic (non mechanical) door handles don’t work when there’s no power…

            Tesla Doors That Won't Open Have Led to 15 Crash-Related Deaths https://www.caranddriver.com/news/a69838848/tesla-doors-dont...

          • alex1138 2 hours ago
            > the cave thing

            Yeah, this does matter to me. I was willing to give him a pass (still am) in a vacuum regarding the Twitter thing given the mass censorship of the old regime (sorry - no, it wasn't acceptable, in any way shape or form) but if he's that petty it doesn't bode well. I keep saying I can believe one thing without subscribing to the Elon fan club

            Doesn't bode well for SpaceX either. Isn't one of the Artemis landers from SpaceX?!

            • rightbyte 1 hour ago
              The "pedo" diver thing and the booster PoE thing.

              It is funny how it is the mundane things that it boil down to when one judge a someone's character. When it gets abstract it is too easy to rationalize.

            • lovich 1 hour ago
              > I was willing to give him a pass (still am) in a vacuum regarding the Twitter thing given the mass censorship of the old regime (sorry - no, it wasn't acceptable, in any way shape or form)…

              Why give him the benefit of the doubt when he censors worse than the previous management? Just look at all the grok lobotomies he gave it because he didn’t like how liberal coded the answers it was giving.

              • alex1138 1 hour ago
                I was just trying to say old Twitter had a serious problem but apparently that goes against the hivemind so I accrue mass downvotes despite posting my comment in good faith
      • croes 2 hours ago
        First, why audit it when the agent can build a new one.

        Second, can you guarantee that an AI company can’t use its AI to hide malicious code from AI audits. Who if not an AI company could have such an expertise?

        I don’t trust a company that pollutes the air of other people with illegal gas turbines because it shows the value their profit over people‘s health

        • Petersipoi 1 hour ago
          Nothing you said here can't be applied to literally any company on earth. And nothing you said here is even a new concern.
          • croes 1 hour ago
            I doubt that every company could hide malicious code so well that AI can’t find it.

            And who said it need to be new concerns? Are the old ones resolved and are they not enough?

        • avaer 1 hour ago
          > Second, can you guarantee that an AI company can’t use its AI to hide malicious code from AI audits. Who if not an AI company could have such an expertise?

          Any evidence for this conspiracy theory? It's not on anyone to disprove this claim.

          > it shows the value their profit over people‘s health

          Companies are chartered to make their shareholders value. To a first approximation, it's illegal for a company to "fuckit, we care about people's health" unless this is what the shareholders voted for (as opposed to making their shares valuable).

          You can argue this is bad, but it isn't about XAI, it applies to every company you've heard of.

    • blfr 2 hours ago
      Your choice is Anthropic, OpenAI, Google, or the Chinese. Who are the good actors within the space?
      • ben_w 2 hours ago
        Rank ordered by reputation / caring about having a trustworthy corporate identity: [Google, Anthropic] in either order depending who you ask, OpenAI, most of the Chinese AI corporations, then Grok.

        This is unfortunate situation to find ourselves in when Grok was also recently at the top of the Pareto frontier for quality/price. Dunno if it still is, this all moves too fast, but it was for at least long enough for me to have heard about it.

        • Cider9986 1 hour ago
          For me, the Chinese labs are far and away the most trustworthy.
        • b112 1 hour ago
          Google?!?!. From where I sit, Google is just above the Chinese. They've been bad-faith actors for more than a decade, I guess everyone is just so used to it that they ignore it.

          I there's anyone I don't trust with AI, it's the worlds #1 company in spying on people, in collection of Pii, in tracking, and many many many times caught literally lying about it.

          Google already knows more about everyone on the planet, than any other 10 organizations combined. Frankly, sadly, they're all, well.. scummy, just each in different ways.

      • aforwardslash 1 hour ago
        None. There are no good actors in a profit-driven endeavour. But open-weight seems pretty good (the chinese)
      • SimianSci 2 hours ago
        The open source and open weight models.

        Surprisingly, despite their motivations in doing so, the Chinese models being open-weight and therefore able to run locally on your own hardware, are far more trustworthy than any blackbox which solely exists to enrich X or Y billionaire.

      • sscaryterry 2 hours ago
        The Chinese are surely less evil than Anthropic, OpenAI and/or Google, at this stage at least.
        • blackqueeriroh 1 hour ago
          You’ve got to be kidding me. Last I checked, Anthropic, OpenAI, and Google haven’t systematically exterminated an entire culture of people.
          • sscaryterry 1 hour ago
            Wow, we're talking tech, jump back in your box. Americans have done exactly the same, that is not what is being discussed here.
          • buran77 1 hour ago
            You're really walking empty handed and with your pants around our ankles in this one.

            Do you really think the US and US big tech in general have a leg to stand on in this regard?

      • SirHackalot 2 hours ago
        [flagged]
    • tadfisher 2 hours ago
      [flagged]
      • jamiequint 2 hours ago
        [flagged]
        • jdiff 2 hours ago
          It's not ad hominem. The head is a strongly polarizing individual. People working for him must either be gravely apathetic or at least of a similar polarity.
        • greggoB 2 hours ago
          The comment actually describes a known social process, with a reasonable base assumption given that said leadership has shown a pattern in this regard.

          Just throwing out debate terms in response seems not so serious, tbh.

        • SirHackalot 2 hours ago
          Not a serious company or CEO either. Have you seen that gesture of him trying to summon the Luftwaffe?
        • brookst 2 hours ago
          Ad hominem means attacking people rather than arguments.

          Pointing out that criminals are criminals is not an ad hominem.

        • ImPostingOnHN 1 hour ago
          "Ad hominem, not a serious argument" is an ad hominem, nonserious argument
        • mplewis 2 hours ago
          explain why it's an ad hominem
          • jamiequint 2 hours ago
            [flagged]
            • swasheck 2 hours ago
              this is an argument from silence because your defense for your assertion rests on the lack of evidence from the assertion to which you replied.

              you made the assertion that it is ad hominem and now you must support it.

            • munk-a 1 hour ago
              Even if you personally have no qualms about Elon Musk his PR is a mess and introduces a lot of risk for long term company viability and funding that competitors just don't have.
        • croes 2 hours ago
          You live under the wrong impression that ad hominem is always bad.

          Ad hominem is allowed under certain circumstances, just remember Epstein.

          Would you have bought anything from him and dismissed any critique of that as ad hominem?

          • lynndotpy 1 hour ago
            Also worth pointing out that it is not an ad hominen.

            Ad hominen is when you attack someone who is making an argument, instead of an argument. "You are flawed, which means your argument is flawed", but that does not follow. If you were in a debate with Epstein or Musk, and he said "2 + 2 = 4", there is no fault of their character that could make the statement untrue.

            But nobody is making that argument. "The leadership" being criticized is not even a participant in this thread (presumably). "The leadership is flawed in this manner" is a statement that can be true or untrue, and "So their product and followers are flawed in these other manners" is something which can follow.

          • actionfromafar 49 minutes ago
            I am sure a lot of then would, if his LLM was good.
        • well_ackshually 1 hour ago
          [flagged]
          • tomhow 23 minutes ago
            One more comment like this and we'll have to ban the account. You've been warned before, and have continued posting vile, abusive comments. When activity like this continues, we have to assume that you have no interest in being held to the site's standards, and that really you want to be banned. If you want to keep participating here, please read the guidelines and demonstrate an intention to be a positive contributor. https://news.ycombinator.com/newsguidelines.html
    • dimgl 2 hours ago
      They made it open source. Are you just trying to be bad faith here? Isn't this what the community was asking for?
      • aforwardslash 1 hour ago
        Reiserfs. A good example on how oss cannot save the product. There are others, but this is the first one that comes to my mind. If you use clearly unethical oss, are you just using oss or are you a part of the problem? Typically, oss purists take these into account.
      • grim_io 2 hours ago
        "Guys, HAL 9000's harness is open source. You can let your agents inspect the code!"
      • croes 2 hours ago
        How about stopping the upload of all the data

        https://news.ycombinator.com/item?id=48877371

        and running their data center with gas turbines without permission while they pollute the air

        https://news.ycombinator.com/item?id=48705717

        you can’t expect people to praise your for making an n+1 harness open source.

        This seems more like, look we made something, now fix it for us

        • dimgl 2 hours ago
          [flagged]
          • ImPostingOnHN 1 hour ago
            > I was just surprised that the top comment had nothing to do with any of the technical aspects of Grok Build (and whether there's any trace of uploads).

            Most people don't restrict themselves to only discussing the technical aspects of a thing. A thing which is technologically novel (e.g. not this example) may nonetheless not be worth using, due to assorted risks.

            I don't find it surprising that HN posters are helping their fellow hackers avoid getting victimized by predators. We just have that sort of nice community :)

      • spiderfarmer 2 hours ago
        [flagged]
        • devindotcom 2 hours ago
          by this standard no good faith criticism of anything musk-adjacent is possible
        • SirHackalot 2 hours ago
          Understandably
      • lynndotpy 1 hour ago
        This is clearly a good-faith criticism and there is no lens in which I could see it described as bad-faith.

        We see this pattern all the time: Someone makes a criticism of a Musk product, and someone assails that criticism with bad-faith accusations of it being "bad-faith".

        Oftentimes, we see that the criticism is undermeasured and ligther than is reasonable, possibly anticipating someone who might accuse it of being "bad faith".

        Maybe someone can put a name to this phenomenon but we see it all the time.

    • rvz 2 hours ago
      Then you better not use Claude Code, since that is still closed source.
    • jamiequint 2 hours ago
      Do you have any examples to illustrate these extraordinary claims?
      • SimianSci 2 hours ago
        The many controversies are not hard to find as the children to your comment will show.

        https://www.nytimes.com/2026/01/22/technology/grok-x-ai-elon...

        • maxloh 1 hour ago
          Why is that even a problem? If no images are released on the internet (and users consume them privately), no one is harmed in the process.

          Blocking AI from generating sexualized images because people could publish deepfakes is no different than banning alcohol because of drunk driving.

          Tools are neutral. Blame the people who misuse the tools and hurt others.

          • mikeyouse 43 minutes ago
            > If no images are released on the internet (and users consume them privately), no one is harmed in the process.

            Yikes.

            A. They were released all over the internet - from the article..

            > The chatbot has a public account on X, where users can ask it questions or request alterations to images. Users flocked to the social media site, in many cases asking Grok to remove clothing in images of women and children, after which the bot publicly posted the A.I.-generated images.

            B. There is a bunch of data about consumers of CSAM 'content escalating' and eventually attempting to make real contact with minors.

            C. They were sexualizing pictures of real people and posting the pictures online.

            > One of the young plaintiffs said she found out about the imagery after she received an anonymous message on Instagram pointing her toward images and videos, including her high school yearbook photo, which had been altered to show her in sexually explicit actions and full nudity.

            The material was being shared on a Discord server, a private chat space on that platform, and included similar imagery that had also been altered using Grok of at least 18 other women who were minors, according to the complaint.

            > Tools are neutral.

            Ha.

        • jamiequint 2 hours ago
          They had a bug in their model that they fixed within days is evidence they are "untrustworthy"?
          • munk-a 1 hour ago
            Elon initially sold xAI as having a spicy mode and being politically incorrect.

            It was only deemed a bug when it became a liability - you can't simply rewrite history and expect it to go unnoticed.

          • ben_w 1 hour ago
            They put it behind a paywall and didn't fix it, according to more recent lawsuits than that article.

            Also, failed to correctly notify authorities even when they eventually notified them at all.

            https://storage.courtlistener.com/recap/gov.uscourts.cand.46...

          • SimianSci 1 hour ago
            You didnt even address my link, this is why you are being called out as a bad-faith actor.
      • agartner 2 hours ago
        • jamiequint 2 hours ago
          read the top comment
          • mplewis 2 hours ago
            read any of your other replies
            • ryandrake 2 hours ago
              OP seems to be asking for examples with an intent to dismiss and downplay each of them, and not to actually read into them and challenge his existing beliefs about X/Grok/Musk.
              • jamiequint 2 hours ago
                LOL, pot meet kettle for real.
                • ryandrake 1 hour ago
                  Open to changing my mind. I would be interested in reading positive, uplifting news about xAI/Grok/Musk that demonstrated a repeated pattern of ethical, careful, compassionate, attentive, and/or responsible business and engineering practices.
              • dimgl 1 hour ago
                I agree about OP.

                However after looking at all of these articles, these all seem like instances of users misusing the product. The product happens to reply on social media, so media publications immediately capitalized on this.

                Seems less like malicious intent from xAI's part and more like a product with young and/or insufficient moderation controls.

                Just today I saw an article where xAI is suing a creator for creating illegal content. https://www.reuters.com/legal/litigation/musks-xai-sues-grok...

      • ryandrake 2 hours ago
      • AshamedBadger56 2 hours ago
      • jdiff 2 hours ago
        • jamiequint 2 hours ago
          • jdiff 2 hours ago
            Starkly different. One was a well meaning attempt to squash model bias gone wrong, the other is a deliberately inserted bias. Even ignoring all that, whataboutism is not persuasive.
            • jamiequint 2 hours ago
              The reality is both are likely well meaning attempts to squash model bias gone wrong. Since you happen to align with the politics of one more than the other, you are having trouble being intellectually honest about your own biases.
              • jdiff 2 hours ago
                In no way are you being intellectually honest if you think that hamfisted system prompt push to prod manipulation was an attempt to squash bias. And again, whataboutism doesn't make xAI better because others are doing bad, too. You asked for evidence of xAI untrustworthiness and received it.
                • jamiequint 1 hour ago
                  What is worse a "hamfisted system prompt push to prod" or a system and organization built to enforce systematic bias in the name of anti-bias?
                  • jdiff 1 hour ago
                    I see your hamfisted cropping of my quote to downplay xAI's actions, since you brought up intellectual honesty.

                    Why do we have to quantity badness? The question you posed was what has xAI done to be perceived as untrustworthy? Stop trying to whatabout Google here. I'm no friend of theirs, it's simply irrelevant.

            • ryandrake 2 hours ago
              Also, it's Whataboutism: Other Company Y doing something bad/untrustworthy isn't a counter to Company X doing something similarly bad/untrustworthy. Both can be bad.
              • jamiequint 1 hour ago
                OK great, do you consider Gemini/Google untrustworthy software that shouldn't be used? Just making sure we're being intellectually honest here.
                • ryandrake 1 hour ago
                  Both are bad and are examples of untrustworthy behavior from their companies, and I would not chime into a thread to defend either of them. Is one example enough to smear an entire company as untrustworthy? No. But numerous examples and patterns of behavior... possibly?
    • make_it_sure 2 hours ago
      getting into politics again...
      • greggoB 2 hours ago
        I think examples such as letting people nudify children qualifies xAI as a bad actor without having to be political.
      • munificent 2 hours ago
        How is it possible for deciding whether or not to build on the labor of some other organized group of people to not be politics?
      • grim_io 2 hours ago
        You know who is apolitical? Russian voters. Works out great for them.
      • fwip 1 hour ago
        There's plenty of non-political reasons to avoid believing anything that a con-man says.
      • nozzlegear 2 hours ago
        You can't separate the man or his business from the politics, he wades into every political debate he can and deliberately tries to troll as many of his perceived enemies as possible.
      • mplewis 2 hours ago
        Grok is a generator of child sexual assault material.
    • eikenberry 2 hours ago
      Aside from their CEO are they really that different from the other big US players? OpenAI, Anthropic and Google all have proven themselves to be untrustworthy as well. We should accept that we have an adversarial relationship with all these companies and shouldn't invest to much in any of them. Use them for what they are worth while the technology matures but be prepared to move on.
      • mplewis 2 hours ago
        Oh yeah, aside from their CEO? OK.
        • eikenberry 1 minute ago
          I mean is Sam Altman is just as bad, just along different lines.
  • charcircuit 1 hour ago
    It's awesome to see openness in these coding agents from the labs making the agents: Codex, Kimi Code, and now Grok Build.
  • maxloh 1 hour ago
    Has anyone tried building from source?

    The commit message says "initial sync from the monorepo." Is this even compilable without the rest of the source code?

    • skp1995 1 hour ago
      yup you can compile, we tested and made sure all the features work before posting
  • simianwords 2 hours ago
    Sigh, why has the industry converged on TUI? Branding and aesthetics over functionality?

    TUI is just much worse for me. I tried Codex CLI vs Codex UI and Codex UI beats it at every level.

    • lynndotpy 1 hour ago
      TUI is a lot better for me, and I have preferred it since the 00s, before LLM products were even a thing.

      For all the reasons there can be, one big reason is that it works on anything you can get a terminal on, you can use it over SSH, and the UI will be the same no matter where you use it.

      I also like that they are very very fast and they don't have the incessant animations that are put into most desktop environments nowadays. If you're on MacOS, the terminal is the only only part of your computer without roadblocks everywhere.

    • _pdp_ 46 minutes ago
      It is a fashion thing. I am not saying that agentic TUIs are bad or anything but it is certain fashionable to use one in 2026.
    • maipen 1 hour ago
      And why are you assuming the industry converged to it when your following statement dismantles your assumption?

      Spacex bought cursor, so it now has it’s agent ui which is just as good as codex + it’s multi-modal

      Anthropic also has it’s own ui

      Zai also launched theirs last month.

      Everyone is converging back to UI.

      The terminal was just a prototype, everyone knew that.

      • greggh 1 hour ago
        Just a prototype? I have no reason to leave the terminal for a GUI IDE. TUI works great, does what I need and is very easy to use and interact with.
      • simianwords 1 hour ago
        Claude code which is most used agent harness doesn’t have desktop equivalent
        • saratogacx 1 hour ago
          It has had one for months. The desktop app has a "code" mode which is Claude Code in GUI form
        • pproe 1 hour ago
          Apart from Claude desktop, that is...
  • lifthrasiir 2 hours ago
    Is this the infamous "cloud upload" routine? I'm not sure it is indeed insidious, though it is of course possible that the code has been filtered out. https://github.com/xai-org/grok-build/blob/main/crates/codeg...
  • wetpaws 1 hour ago
    [dead]
  • SubiculumCode 2 hours ago
    [flagged]
  • justinkramp 2 hours ago
    [flagged]
    • tomhow 2 hours ago
      Please don't just post the most obvious snarky comment about a given topic. The guidelines make it clear we're trying for something better here. https://news.ycombinator.com/newsguidelines.html
      • mattbillenstein 2 hours ago
        Sorta amazes me how people in various levels of power will not say the obvious thing or actively discourage saying the obvious thing because it might offend Elon.

        Recently all the big bank CEOs involved with the SpaceX IPO - a lot of money in that for them - but a company trading at 100x sales is clearly crazy.

        • tomhow 1 hour ago
          People post critical things about the most powerful people and companies all the time here and we have zero problem with it.

          What I'm asking for is for people to not post the most obvious, snarky comment, regardless of the topic/target, not because of who it may “offend” (as if the most powerful people in the world would have any awareness or care about a comment like that on HN), but because it makes HN seem repetitive, miserable and lame.

          Critique away, just make discussions thoughtful and substantive, which is what HN is for.

          • lynndotpy 1 hour ago
            For what it's worth, this doesn't read as "snark" to me. There _are_ many direct critiques in this thread about X being caught uploading users home directories, and some are clearly snark. I understand that you read this as a rhetorical question meant as a critique.

            But it's really not clear to me why this should be read as a snarky, critical, rhetorical question. Someone who eagerly wants to use Grok Build would ask this exact same question.

            "Does this [Grok Build] also just directly suck all your code up and make a copy of it on their servers?" is a question that is (1) salient and (2) answerable and (3) could be thoroughly devastating for someone to find out on their own by using it.

            The answer is not present in the README, and XAi has blocked Issues and Discussions, so there's none of the usual avenues on GitHub to ask these questions. It seems perfectly typical and expected for someone to ask this question here.

            • tomhow 12 minutes ago
              I understand reading it as benign and sincere if you're sympathetic to the sentiment. As someone whose job it is to read the comments all day every day, and whose objective is to keep discussions here as intellectually gratifying as possible, it just comes across as unsubstantive at best, and jeering at worst.

              The project is open source; if the commenter was sincerely curious about what the software does with a user's code, they could have checked themselves or phrased the question in a way that made it clear they were genuinely interested in finding out.

              My response wasn't in any way vigilant or threatening; just a polite request to use HN in a way that's consistent with its intended spirit.

        • ofjcihen 2 hours ago
          [flagged]
    • ofjcihen 2 hours ago
      Honestly a great question. I mean if it’s open source someone will check (I don’t use xAI but believe me I would be checking first if I did).
  • whalesalad 2 hours ago
    [flagged]
  • calldacopsidgaf 2 hours ago
    [flagged]
    • dimgl 2 hours ago
      Why snowflakes? You can use /feedback in the app.
  • nickreese 1 hour ago
    This is 100% smoke and mirrors. Prove the bucket is empty and nothing was transferred out and I'll believe they deleted it.