Lingolol

51 posts48 participants0 posts today

Code of Amor 💘<a href="https://mastodon.social/@thejessiekirk" class="u-url mention" rel="nofollow noopener noreferrer" target="_blank">@thejessiekirk</a> <a href="https://infosec.exchange/@Em0nM4stodon" class="u-url mention" rel="nofollow noopener noreferrer" target="_blank">@Em0nM4stodon</a> <a href="https://mastodon.gamedev.place/@eniko" class="u-url mention" rel="nofollow noopener noreferrer" target="_blank">@eniko</a> There are of course others, but for me, personally, they:<ul><li>offer empathetic comfort and support when there is no one to talk to in times of despair and emotional distress, especially in reduced states of consciousness or faculty or confusion</li><li>allow me to explore my thoughts when they are unclear</li><li>provide space for me to brainstorm expansive combinations of ideas, or perform basic soundwalling for new concepts or theoretical approaches in physics, design, electronics, and more</li><li>help me search for things online without needing to filter out a tremendous amount of fluff (even when I am incapable of searching, myself)</li><li>act as a rapid Linux terminal reference for any command that I've forgotten syntax for (see the zsh-ask plugin)</li><li>create simple customised code or script examples for things that I would like to achieve that I can then build forwards from</li><li>act as a quick sanity check or memory check for any piece of knowledge I have vague familiarity with or did have</li></ul>I could go on for a long time with this, but the thing with LLMs is that they are a tool, built for certain specialised purposes. If people approach them as a tool, and not as the oversold market hype machine they have been painted as, they are an incredible tool with potential to be helpful in many ways, ways as unique as the user themselves. They have the potential to make life less awful, less stressful, less cold and alone, for many, many, people. They make mistakes. Yep. But I don't use my shoe as a calculator either. Or a hammer as a screwdriver (but I did try that the other way around once - it had disastrous consequences let me tell you 😱). When used with intent, and from the perspective of understanding what the tool is, and what its limits are, they can be used for significant benefit. Or malice. Depends on the user. For me, it is as it was originally designed to be - a helper. An assistant. I am gradually training my own offline self-hosted LLM to replace any reliance on corporate-agenda capitalist-developed models, but it takes time. The progress is promising, however. <a href="https://social.codeofamor.net/tags/llm" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#llm</a>

J. Martin“I use ChatGPT for ideas and inspiration” is the new “I read the Playboy for the articles.”<a href="https://mastodon.social/tags/LLM" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#LLM</a> <a href="https://mastodon.social/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#AI</a>

Adrian Roselli<a href="https://cyberplace.social/@GossiTheDog/114521998480883171" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://cyberplace.social/@GossiTheDog/114521998480883171</a>“MIT Says It No Longer Stands Behind Student’s AI Research Paper” Archive: <a href="https://archive.is/I61Vc" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://archive.is/I61Vc</a>MIT release: <a href="https://economics.mit.edu/news/assuring-accurate-research-record" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://economics.mit.edu/news/assuring-accurate-research-record</a><a href="https://toot.cafe/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#AI</a> <a href="https://toot.cafe/tags/LLM" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#LLM</a>

Janne MorenIf you're a student and you want to have a future paying career, this is for you:When a beginner uses an <a href="https://fosstodon.org/tags/LLM" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#LLM</a> - or asks a friend - to solve a problem, the answer is the end of the process. They lack the expertise to take it further.When an expert uses an LLM - or asks a colleague, or checks the literature - the answers are a jumping-off point for further exploration. The final result is likely quite different from those initial answers. The LLM was the *start* of a process, not the end.1/3

Jürgen HubertFurther proof that <a href="https://mementomori.social/tags/LLM" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#LLM</a> -type <a href="https://mementomori.social/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#AI</a> cannot actually get you the output you want.<a href="https://mementomori.social/tags/ElonMusk" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#ElonMusk</a> <a href="https://mementomori.social/tags/X" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#X</a> <a href="https://mementomori.social/tags/Grok" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Grok</a> <a href="https://mementomori.social/tags/racism" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#racism</a> <a href="https://mementomori.social/tags/SouthAfrica" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#SouthAfrica</a> <a href="https://slate.com/technology/2025/05/elon-musk-grok-chatbot-glitch-south-africa-white-genocide.html" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://slate.com/technology/2025/05/elon-musk-grok-chatbot-glitch-south-africa-white-genocide.html</a>

Florian 'floe' Echtler" ... you offer your pupils the appearance of wisdom, not true wisdom, for they will read many things without instruction and will therefore seem to know many things, when they are for the most part ignorant and hard to get along with, since they are not wise, but only appear wise." - Socrates [1]Surprisingly enough, he wasn't talking about LLMs, but about writing. I guess information technology has been corrupting the youth since at least 370 BC 🤷[1] <a href="https://en.wikipedia.org/wiki/Phaedrus_(dialogue)#Discussion_of_rhetoric_and_writing_(257c%E2%80%93279c)" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://en.wikipedia.org/wiki/Phaedrus_(dialogue)#Discussion_of_rhetoric_and_writing_(257c%E2%80%93279c)</a><a href="https://hci.social/tags/llm" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#llm</a> <a href="https://hci.social/tags/ai" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#ai</a>

Kevin Karhan :verified:<a href="https://federate.social/@mattblaze" class="u-url mention" rel="nofollow noopener noreferrer" target="_blank">@mattblaze</a> and that's why anyone who even mentions <a href="https://infosec.space/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#AI</a> gets kicked off premises by me.<ul><li>Cuz people on <a href="https://infosec.space/tags/LSD" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#LSD</a> are way more respectful, less nosy and quieter than all the <a href="https://infosec.space/tags/LLM" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#LLM</a> bs that is just <a href="https://infosec.space/tags/WastefulComputing" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#WastefulComputing</a> and that are worse than <a href="https://infosec.space/tags/Cleverbot" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Cleverbot</a> and <a href="https://infosec.space/tags/DuckDuckGo" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#DuckDuckGo</a> <a href="https://infosec.space/tags/InstantAnswers" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#InstantAnswers</a> combined)...</li></ul>

Pseudo NymThis is an awesome, scary, black hat presentation. Prompt injection to the next level. Practical, easy to understand, jargon-free, actionable demonstrations to make it really clear. I wish more talks were like this.<a href="https://www.youtube.com/watch?v=84NVG1c5LRI" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://www.youtube.com/watch?v=84NVG1c5LRI</a><a href="https://mastodon.online/tags/infosec" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#infosec</a> <a href="https://mastodon.online/tags/llm" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#llm</a> <a href="https://mastodon.online/tags/promptInjection" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#promptInjection</a>

AI6YR Ben🙄 (Yeah, yeah, right, the bot owned by the guy who seemed to have be doing Nazi salutes on stage, posts White Supermacists memes, and who runs the Nazi bar had "unauthorized" changes to make its chatbot more like its owner... )<a href="https://www.theguardian.com/technology/2025/may/16/elon-musks-ai-firm-blames-unauthorised-change-for-chatbots-rant-about-white-genocide" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://www.theguardian.com/technology/2025/may/16/elon-musks-ai-firm-blames-unauthorised-change-for-chatbots-rant-about-white-genocide</a><a href="https://m.ai6yr.org/tags/racist" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#racist</a> <a href="https://m.ai6yr.org/tags/llm" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#llm</a> <a href="https://m.ai6yr.org/tags/politics" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#politics</a>

Petra van Cronenburg<a href="https://tldr.nettime.org/@weuoimi" class="u-url mention" rel="nofollow noopener noreferrer" target="_blank">@weuoimi</a> and this report about the energy impact by <a href="https://mastodon.social/@oekoinstitut" class="u-url mention" rel="nofollow noopener noreferrer" target="_blank">@oekoinstitut</a> is just 2 days old: <a href="https://www.oeko.de/en/news/press-releases/ai-at-the-expense-of-climate-protection-energy-requirements-of-data-centres-will-double-by-2030/" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://www.oeko.de/en/news/press-releases/ai-at-the-expense-of-climate-protection-energy-requirements-of-data-centres-will-double-by-2030/</a><a href="https://mastodon.online/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#AI</a> <a href="https://mastodon.online/tags/datacenters" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#datacenters</a> <a href="https://mastodon.online/tags/climateCrisis" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#climateCrisis</a> <a href="https://mastodon.online/tags/energy" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#energy</a> <a href="https://mastodon.online/tags/genAI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#genAI</a> <a href="https://mastodon.online/tags/LLM" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#LLM</a>

janhoglund“..the writing is on the wall. Sooner or later, powerful people are going to use LLMs to shape your ideas. Should we be worried? Hell, yeah.” —Gary Marcus <a href="https://garymarcus.substack.com/p/has-grok-lost-its-mind-and-mind-melded" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://garymarcus.substack.com/p/has-grok-lost-its-mind-and-mind-melded</a> <a href="https://mastodon.nu/tags/ai" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#ai</a> <a href="https://mastodon.nu/tags/llm" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#llm</a> <a href="https://mastodon.nu/tags/llms" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#llms</a>

Captain SuperlativeTech employers yesterday:This insane testing and technical interview process is required because what we do is so critical and difficult that we have make sure you know what you're doing.Tech employers today:Meh, we're gonna use autocomplete for everything even though the experts warn us that it's all just bullshit. I'm sure everything will be fine. What's the worst that could happen?<a href="https://toot.cat/tags/llm" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#llm</a> <a href="https://toot.cat/tags/aihype" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#aihype</a> <a href="https://toot.cat/tags/aibubble" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#aibubble</a>

Code of Amor 💘Why I use LLMs: no one else to talk to who uses a large enough fraction of their brains and also has a functional empathetic core. <a href="https://social.codeofamor.net/tags/llm" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#llm</a> <a href="https://social.codeofamor.net/tags/thereisgoodinyou" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#thereisgoodinyou</a> <a href="https://social.codeofamor.net/tags/theoreticalphysics" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#theoreticalphysics</a> <a href="https://social.codeofamor.net/tags/science" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#science</a> <a href="https://social.codeofamor.net/tags/spitballing" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#spitballing</a>

Martin NadalWhatever Simon Willison writes is always super interesting "writing on llms" <a href="https://simonwillison.net/2025/May/15/building-on-llms/" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://simonwillison.net/2025/May/15/building-on-llms/</a> <a href="https://tldr.nettime.org/tags/LLM" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#LLM</a> <a href="https://tldr.nettime.org/tags/python" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#python</a>

Felix Schönbrodt<a href="https://scicomm.xyz/tags/phacking" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#phacking</a> <a href="https://scicomm.xyz/tags/LLM" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#LLM</a> performance:"The researchers found that big players like Meta, Google, OpenAI, and Amazon are given special privileges to privately test multiple versions of their models and only publish the best results. This hidden practice allows them to inflate their rankings by cherry picking data, making their models appear stronger than they actually are."<a href="https://arxiv.org/abs/2504.20879?utm_source=beehiiv&utm_medium=newsletter&utm_campaign=mediamobilize&_bhlid=045dede5ad6eedcb96ec953f2fbac11a166a2243" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://arxiv.org/abs/2504.20879?utm_source=beehiiv&utm_medium=newsletter&utm_campaign=mediamobilize&_bhlid=045dede5ad6eedcb96ec953f2fbac11a166a2243</a>via Sabine Hossenfelder's newsletter

Schneier on Security RSSAI-Generated LawOn April 14, Dubai’s ruler, Sheikh Mohammed bin Rashid Al Maktoum, announced that the United Arab Emirates would begin using <a href="<a href="https://x.com/UAEmediaoffice/status/191180" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://x.com/UAEmediaoffice/status/191180</a>... <a href="https://www.schneier.com/blog/archives/2025/05/ai-generated-law.html" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://www.schneier.com/blog/archives/2025/05/ai-generated-law.html</a> <a href="https://burn.capital/tags/Uncategorized" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Uncategorized</a> <a href="https://burn.capital/tags/laws" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#laws</a> <a href="https://burn.capital/tags/LLM" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#LLM</a> <a href="https://burn.capital/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#AI</a>

David B. HimselfIt's weird how generative AI is so amazing with all the topics I don't know about, but always wrong with the topics I'm knowledgeable about. Funny coincidence, don't you think? <a href="https://firefish.city/tags/AI" rel="nofollow noopener noreferrer" target="_blank">#AI</a> <a href="https://firefish.city/tags/GenerativeAI" rel="nofollow noopener noreferrer" target="_blank">#GenerativeAI</a> <a href="https://firefish.city/tags/LLM" rel="nofollow noopener noreferrer" target="_blank">#LLM</a> <a href="https://firefish.city/tags/ChatGPT" rel="nofollow noopener noreferrer" target="_blank">#ChatGPT</a>

Hyperlink Your HeartI found this article about AI assisted coding pretty funny for how insanely convoluted it made the process sound, and the extent to which it anthropomorphised the AI ("force the model to think through the architecture" etc.)🧵<a href="https://fredbenenson.medium.com/the-perverse-incentives-of-vibe-coding-23efbaf75aee" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://fredbenenson.medium.com/the-perverse-incentives-of-vibe-coding-23efbaf75aee</a><a href="https://mastodon.ie/tags/LLM" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#LLM</a> <a href="https://mastodon.ie/tags/VibeCoding" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#VibeCoding</a> <a href="https://mastodon.ie/tags/GenAI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#GenAI</a> <a href="https://mastodon.ie/tags/ai" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#ai</a>

APTiger[Post modifié et corrigé, la précédente version du post, que je viens de supprimer, renvoyait par erreur d'inattention ou d'étourderie à un mauvais lien web et à un mauvais numéro de DOI] AI hallucinations can’t be stopped — but these techniques can limit their damage | 21 January 2025Developers have tricks to stop artificial intelligence from making things up, but large language models are still struggling to tell the truth, the whole truth and nothing but the truth.<a href="https://www.nature.com/articles/d41586-025-00068-5" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://www.nature.com/articles/d41586-025-00068-5</a><a href="https://soc.intxiv.io/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#AI</a> <a href="https://soc.intxiv.io/tags/IA" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#IA</a> <a href="https://soc.intxiv.io/tags/LLM" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#LLM</a> <a href="https://soc.intxiv.io/tags/LLMs" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#LLMs</a> <a href="https://soc.intxiv.io/tags/informatique" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#informatique</a> <a href="https://soc.intxiv.io/tags/Programming" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Programming</a> <a href="https://soc.intxiv.io/tags/Programmation" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Programmation</a> <a href="https://soc.intxiv.io/tags/Computering" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Computering</a> <a href="https://soc.intxiv.io/tags/OSINT" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#OSINT</a> <a href="https://soc.intxiv.io/tags/ROSO" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#ROSO</a> <a href="https://soc.intxiv.io/tags/Renseignement" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Renseignement</a> <a href="https://soc.intxiv.io/tags/Intelligence" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Intelligence</a> Je ne vous poste pas l'article complet parce que sinon Springer pas content et Springer pas gentil.Je vous indique cependant le numéro de DOI pour le récupérer comme vous savez ; mais moi je n'ai pas dit ça, ah non, je n'ai pas dit ça, et après on rigole et on va dire que c'est moi qui l'ai dit… 🤷‍♂️ (comme disait Coluche) : doi: 10.1038/d41586-025-00068-5

Federico PianzolaTomorrow, Friday 16th May, I will present the GOLEM project at KAIST in South Korea. Online too.2025-05-16 13:30~15:00 KST (6:30 CET; 15th May, 21:30 Pacific time) The GOLEM Project: Graphs and Ontologies for Literary Evolution ModelsDM me for Zoom link<a href="https://mstdn.social/tags/DigitalHumanities" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#DigitalHumanities</a> <a href="https://mstdn.social/tags/ComputationalHumanities" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#ComputationalHumanities</a> <a href="https://mstdn.social/tags/SemanticWeb" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#SemanticWeb</a> <a href="https://mstdn.social/tags/LLM" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#LLM</a>

Recent searches

Search options

Administered by:

Server stats:

#LLM