• 0 Posts
  • 6 Comments
Joined 3 years ago
cake
Cake day: June 14th, 2023

help-circle
  • Many projects require replicatable builds. Generative AI is driven by noise so I am not sure if it has any role in optimization or code gen. I guess we could share an RNG seed with source but honestly current tool chains are too slow as it is and ML would likely make it much worse. I want faster edit/compile/debug cycles and compilers already do insane optimisations. Ken Thompson’s Reflections on Trusting Trust becomes much scarier if your code generation is being driven by a black box of billions or parameters nobody understands. An adversary could embed anything in there.

    The machine translation origins of LLMs make them really good at inspecting text and looking for things that look like known bugs. It must be a game changer for security researchers who already know what sort of problems to look for but don’t have enough hours in the day to search the haystack manually. If some of that knowledge was available in a good quality FOSS local code analysis tool it might be irresponsible not to use it. We can’t be dependent on commercial vendors and closed models.


  • Using LLMs to analyze code for problems makes a lot of sense. It is just an extension of the rule based analysis we already do. It is very much a better hammer rather than a replacement of labour/thinking/taste. The programming languages we use are often difficult for humans to reason about reliably and hide a lot of errors. Our programming languages still suck (yes even Rust). If one day clang of gcc ships with a local LLM to lint code for security problems I would probably run it occasionally even on my hobby code despite being an AI hater.

    If people are identifying errors using LLMs and fixing them by hand I am fine with that. It seems like a good use of the technology. If they start vibe coding drivers I want it declared so I can disable them or move to BSD.


  • Sounds like a bullshit headline but it is an eternal truth that the owners of capital demand slave labour. Take the work houses away and they demand chattel slavery. Take the chattel slaves away and they want convict labour. Take the convict labour away and they want undocumented foreigners to exploit. Take that away and they casualise work and pay people in tips or commissions. Take that away and they export work to countries that will do all the things they can’t do at home.

    The investment in LLMs is purely a scam at this point but much of the initial enthusiasm was driven by the disdain for labour. Highly skilled labour like engineers keep arguing with the MBAs that their cost cutting is going to kill people or destroy the company and that really upsets them and risks the bonuses and share buyback program.

    While the drudgery of work isn’t great, I would much prefer the capital class invested in R&D and equipment to make their skilled, specialized workers more productive and improved our remuneration and conditions as happened in previous generations before the MBA apocalypse. Instead they are trying to cut us out. Too many people are living paycheck to paycheck and unable to afford housing as wealth inequality escalates.

    Mistreatment of animals or brain cells in a box isn’t the same as putting a child to work in a mine but the mindset that drives us in this direction is.




  • Not sure it’s possible to fight machine generated slop on the big platforms now. People who invest a lot of time and effort are being squeezed out. Businesses that don’t give a fuck about quality upload multiple low effort videos a week across multiple channels getting massive views, probably most from gaming the system.

    Sad for the people who spend months researching, writing and producing videos but this might be the end of the ride as far as making videos as a day job. The platform doesn’t value them or their loyal viewers.

    The creators facing burnout who don’t leave out of disgust will join the slop machine and start championing it. Starting to see it already.