{"id":72951,"date":"2023-06-08T08:41:45","date_gmt":"2023-06-08T08:41:45","guid":{"rendered":"https:\/\/alecmuffett.com\/?p=72951"},"modified":"2023-06-08T08:41:50","modified_gmt":"2023-06-08T08:41:50","slug":"a-short-note-for-civil-society-on-the-inadvisability-of-government-regulation-restriction-of-llm-and-ai-opensource-freespeech-codeisspeech","status":"publish","type":"post","link":"https:\/\/alecmuffett.com\/article\/72951","title":{"rendered":"A short note for Civil Society on the inadvisability of Government regulation & restriction of #LLM and #AI \u2014\u00a0#OpenSource #FreeSpeech #CodeIsSpeech"},"content":{"rendered":"\n

So there’s an interesting Reddit post<\/a> which leads to an article on VentureBeat<\/a> and a letter from a Senator to Mark Zuckerberg<\/a>, all attached below.<\/p>\n\n\n\n

There’s a fairly straightforward point which needs to be made:<\/p>\n\n\n\n

\"\"
indicative comment from the Reddit thread<\/figcaption><\/figure>\n\n\n\n
    \n
  1. Like it or not, Code<\/em> and Data<\/em> are forms of speech; yes they can be made to “do things” or they can “describe things” but that doesn’t make them any less “speech”, instead it makes them more-something-else-too.<\/li>\n\n\n\n
  2. LLM data, is data like everything else. Just as Firefox will not function without configuration files, so goes LLaMa.<\/li>\n\n\n\n
  3. If the Government is encouraged to suppress general access to code, it will attempt to do so; and then the suppressions will fail several years, perhaps a few decades, later. This is what happened with Cryptographic Export Controls, too.<\/li>\n\n\n\n
  4. In the face of suppression, open-source developers will hoard the extant code<\/a>, and build upon it, thereby sedimenting all the bias and other undesirable attributes which the extant code holds.<\/li>\n\n\n\n
  5. So if you want a kinder, gentler AI in deployment<\/em>, you need to encourage open development and collaboration so that hoarding is undesirable, and so that developers can freely track all the developments which make AI better.<\/strong><\/li>\n\n\n\n
  6. In short: involving the Government will be counterproductive.<\/li>\n<\/ol>\n\n\n\n
    \n\n\n\n
    \n

    > The open dissemination of LLaMA represents a significant increase in the sophistication of <code> available to the general public, and raises serious questions about the potential for misuse or abuse.

    God forbid the general public get access to "code."    https:\/\/t.co\/T163cZa1mC<\/a> pic.twitter.com\/9dzEt9tS7U<\/a><\/p>— Alec Muffett (@AlecMuffett) June 8, 2023<\/a><\/blockquote>