Recent searches

No recent searches

Search options

Only available when logged in.
mastodon.ie is one of the many independent Mastodon servers you can use to participate in the fediverse.
Irish Mastodon - run from Ireland, we welcome all who respect the community rules and members.

Administered by:

Server stats:

1.5K
active users

mastodon.ie: AboutProfiles directoryPrivacy policy

Mastodon: AboutGet the appKeyboard shortcutsView source codev4.4.3

#llm

220 posts192 participants0 posts today
infoek.cz :archlinux:<p>😎 HTC představuje lehoučké AI brýle VIVE Eagle </p><p>ℹ️ PR článek</p><p>🔗 <a href="https://infoek.cz/htc-predstavuje-lehoucke-ai-bryle-vive-eagle-2025/" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">infoek.cz/htc-predstavuje-leho</span><span class="invisible">ucke-ai-bryle-vive-eagle-2025/</span></a></p><p><a href="https://mamutovo.cz/tags/HTC" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>HTC</span></a> <a href="https://mamutovo.cz/tags/VIVEagle" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>VIVEagle</span></a> <a href="https://mamutovo.cz/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LLM</span></a> <a href="https://mamutovo.cz/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a></p>
kasdeyaAI hot take. pretty scary hypothetical
KinkKong<p><span class="h-card" translate="no"><a href="https://infosec.exchange/@sortius" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>sortius</span></a></span> </p><p><a href="https://kinkycats.org/tags/TheGuardian" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>TheGuardian</span></a> itself partners with OpenAI and sacks human journalists. Maybe the article has been written by ChatGPT?</p><p><a href="https://www.theguardian.com/gnm-press-office/2025/feb/14/guardian-media-group-announces-strategic-partnership-with-openai" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="ellipsis">theguardian.com/gnm-press-offi</span><span class="invisible">ce/2025/feb/14/guardian-media-group-announces-strategic-partnership-with-openai</span></a></p><p><a href="https://kinkycats.org/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a> <a href="https://kinkycats.org/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LLM</span></a> <a href="https://kinkycats.org/tags/journalism" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>journalism</span></a></p>
Pavel A. Samsonov<p>The biggest impact of <a href="https://mastodon.social/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LLM</span></a> has not been due to the technology's capabilities, but rather its flaws. Because the <a href="https://mastodon.social/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a> revolution must succeed at any cost, but the actual tech is flawed, orgs demanding more AI outputs MUST lower their bar for quality.</p><p>This creates space for <a href="https://mastodon.social/tags/UX" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>UX</span></a> malpractice to take root.</p><p>Designers who think they are fighting back against evil PMs who hate users and want to make bad products, are in reality committing an equally egregious sin.</p><p><a href="https://productpicnic.beehiiv.com/p/aura-ux" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">productpicnic.beehiiv.com/p/au</span><span class="invisible">ra-ux</span></a></p>
Denis 🌏<p>"However, the context does not specify individual rods for purchase separately."</p><p>Oops. Your slip is showing.</p><p><a href="https://hachyderm.io/tags/ikea" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ikea</span></a> <a href="https://hachyderm.io/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a> <a href="https://hachyderm.io/tags/llm" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>llm</span></a></p>
Denis 🌏<p>The IKEA support chat "My name is Chandan!" is definitely an AI. 🤮 </p><p><a href="https://hachyderm.io/tags/ikea" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ikea</span></a> <a href="https://hachyderm.io/tags/ai" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ai</span></a> <a href="https://hachyderm.io/tags/llm" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>llm</span></a></p>
C.<p>I posted this as a reply elsewhere, but think it stands on its own, fleshed out a bit. So here it is.</p><p>It's about "AI" - LLMs or chatbots, of course.</p><p>Ask an LLM "Is &lt;X&gt; a good idea?", and it will tell you why it's a good idea.<br>Ask the same LLM "Is &lt;X&gt; a bad idea?", and it will tell you why it's a bad idea.</p><p>It occurs to me that this behaviour - ask a chatbot a leading question, and it will give you an "answer" that backs up the assumptions buried in the question - is the realization of the computer system Douglas Adams called "Reason" (in Dirk Gently's Holistic Detective Agency).</p><p>Originally designed as a reasoning machine to help you find the correct solution to a problem you described, it instead turned out to be good at retroactively finding an incontrovertible argument to justify a decision you'd already made. It was so successful the Pentagon bought it lock, stock, and barrel, and military purchasing decisions started to have the hallmarks of its output.</p><p>So, they really have done a version of building the Torment Nexus from a book advising against it.</p><p><a href="https://mindly.social/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LLM</span></a> <a href="https://mindly.social/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a> <a href="https://mindly.social/tags/Reason" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Reason</span></a> <a href="https://mindly.social/tags/DouglasAdams" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>DouglasAdams</span></a> <a href="https://mindly.social/tags/DirkGently" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>DirkGently</span></a> <a href="https://mindly.social/tags/reasoning" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>reasoning</span></a> <a href="https://mindly.social/tags/TormentNexus" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>TormentNexus</span></a> <a href="https://mindly.social/tags/justification" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>justification</span></a> <a href="https://mindly.social/tags/chatbot" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>chatbot</span></a></p>
Metin Seven 🎨<p>🧵 ChatGPT 5 being candid… 😃</p><p>(Found in the ChatGPT Reddit)</p><p><a href="https://graphics.social/tags/ChatGPT" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ChatGPT</span></a> <a href="https://graphics.social/tags/OpenAI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>OpenAI</span></a> <a href="https://graphics.social/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LLM</span></a> <a href="https://graphics.social/tags/LLMs" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LLMs</span></a> <a href="https://graphics.social/tags/ChatGPT5" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ChatGPT5</span></a> <a href="https://graphics.social/tags/MachineLearning" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MachineLearning</span></a> <a href="https://graphics.social/tags/chatbots" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>chatbots</span></a> <a href="https://graphics.social/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a> <a href="https://graphics.social/tags/ArtificialIntelligence" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ArtificialIntelligence</span></a> <a href="https://graphics.social/tags/AISlop" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AISlop</span></a></p>
Sunflower Björnskalle 🌻<p><a href="https://mastodon.online/tags/ChatGPT" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ChatGPT</span></a> and <a href="https://mastodon.online/tags/DeepSeek" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>DeepSeek</span></a> are so incredibly wrong when you ask it even basic questions about <a href="https://mastodon.online/tags/typetheory" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>typetheory</span></a>. I asked for the top 10 simplest recursive types. Only 4 of them even were for DS and ChatGPT wasn't much better but a bunch of its formulations were even wrong. One of the things it gave wasn't even semantically let along syntactically valid. Also the ways they were wrong were braindead: the unit type and empty types aren't even close.</p><p><a href="https://mastodon.online/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LLM</span></a> <a href="https://mastodon.online/tags/ai" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ai</span></a></p>
Will Berard 🫳🎤🫶<p>Is <a href="https://mastodon.acm.org/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a> influencing your reality? <br><a href="https://youtube.com/shorts/GjVDqQU5PNk" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="">youtube.com/shorts/GjVDqQU5PNk</span><span class="invisible"></span></a><br>(That's not pandering, that's diffusing research.)<br><a href="https://mastodon.acm.org/tags/chatgpt" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>chatgpt</span></a> <a href="https://mastodon.acm.org/tags/llm" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>llm</span></a> <a href="https://mastodon.acm.org/tags/AISafety" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AISafety</span></a> <a href="https://mastodon.acm.org/tags/MentalHealth" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MentalHealth</span></a> <a href="https://mastodon.acm.org/tags/ChatbotPsychosis" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ChatbotPsychosis</span></a></p>
Marquis de Geek<p>More <a href="https://ohai.social/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LLM</span></a> messings:</p><p>VIRGILIA:<br>Therefore!<br>Why, for Lord, my frience good will thy son him busining sir; And much shall good a consentrance as and hours, on unto remingenal desent;</p>
Dr. Stefan T. Siegel<p>Warum ich niemals KI nutzen würde, Sie es dennoch im Studium lernen sollten</p><p>Absolut sehenswerter Videoessay von Kevin Schumacher (KIT-Bibliothek, HoC-Schreiblabor): <a href="https://lnkd.in/dNy2C7gw" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="">lnkd.in/dNy2C7gw</span><span class="invisible"></span></a></p><p><a href="https://bildung.social/tags/ai" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ai</span></a> <a href="https://bildung.social/tags/ki" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ki</span></a> <a href="https://bildung.social/tags/llm" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>llm</span></a> <a href="https://bildung.social/tags/learning" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>learning</span></a> <a href="https://bildung.social/tags/teaching" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>teaching</span></a> <a href="https://bildung.social/tags/sustainability" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>sustainability</span></a> <a href="https://bildung.social/tags/competence" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>competence</span></a> <a href="https://bildung.social/tags/technology" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>technology</span></a> <a href="https://bildung.social/tags/automatization" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>automatization</span></a> <a href="https://bildung.social/tags/growth" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>growth</span></a> <a href="https://bildung.social/tags/bias" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>bias</span></a> <a href="https://bildung.social/tags/innovation" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>innovation</span></a> <a href="https://bildung.social/tags/writing" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>writing</span></a> <a href="https://bildung.social/tags/cognitiveoffloading" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>cognitiveoffloading</span></a> <a href="https://bildung.social/tags/illusionofcompetence" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>illusionofcompetence</span></a> <a href="https://bildung.social/tags/society" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>society</span></a> <a href="https://bildung.social/tags/ethics" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ethics</span></a> <a href="https://bildung.social/tags/digitaledrecksarbeit" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>digitaledrecksarbeit</span></a> <a href="https://bildung.social/tags/mentalhealth" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>mentalhealth</span></a> <a href="https://bildung.social/tags/copyright" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>copyright</span></a> <a href="https://bildung.social/tags/unemployment" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>unemployment</span></a> <a href="https://bildung.social/tags/criticalthinking" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>criticalthinking</span></a></p>
Assn for Computing Machinery<p>“If we continue on the current path … we are basically playing Russian roulette with humanity,” warns <a href="https://mastodon.acm.org/tags/ACMTuringAward" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ACMTuringAward</span></a> recipient Yoshua Bengio.</p><p>Agents have been around for decades. But in recent months, a new class of agents has arrived on the scene: ones built using large language models. How to make them work safely and securely? <a href="https://www.technologyreview.com/2025/06/12/1118189/ai-agents-manus-control-autonomy-operator-openai/" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="ellipsis">technologyreview.com/2025/06/1</span><span class="invisible">2/1118189/ai-agents-manus-control-autonomy-operator-openai/</span></a> </p><p><a href="https://mastodon.acm.org/tags/TechNews" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>TechNews</span></a> <a href="https://mastodon.acm.org/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LLM</span></a> <a href="https://mastodon.acm.org/tags/AIAgent" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AIAgent</span></a></p>
Hraban (fiëé visuëlle)<p>Ich bleibe dabei:<br>„KI“ (im Sinne von LLMs und Generativer KI) ist eine durch und durch zerstörerische Technologie.</p><p>Sie zerstört<br>– Umwelt (Energie &amp; Wasser)<br>– Jobs (IT, Kreation…)<br>– Wissen, Bildung &amp; Wissenschaft<br>– Arbeits- &amp; Lebenszeit (Schadensmanagement, Analyse…)<br>… um Lügen zu verbreiten:<br>– fake news, fake science, deep fakes<br>– Vorurteile (Sexismus, Rassismus…)</p><p>Ihr privater Einsatz ist mit Littering zu vergleichen, ihr gewerblicher Einsatz ist ein Verbrechen.</p><p><a href="https://literatur.social/tags/KI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>KI</span></a> <a href="https://literatur.social/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a> <a href="https://literatur.social/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LLM</span></a></p>
Deborah Preuss, pcc 🇨🇦<p><span class="h-card" translate="no"><a href="https://mastodon.social/@bruces" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>bruces</span></a></span> … consider, also:</p><p>"GPT-5’s average energy consumption for a medium-length response is just over 18 watt-hours, a figure that is higher than all other models they benchmark except for OpenAI’s o3 reasoning model, released in April, and R1, made by the Chinese AI firm Deepseek."</p><p>This is “significantly more energy than GPT-4o.”<br><a href="https://www.theguardian.com/technology/2025/aug/09/open-ai-chat-gpt5-energy-use" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="ellipsis">theguardian.com/technology/202</span><span class="invisible">5/aug/09/open-ai-chat-gpt5-energy-use</span></a></p><p>Another report I read said <br>Energy usage: <a href="https://cosocial.ca/tags/chatGPT" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>chatGPT</span></a> 5 = (5 x chatGPT 4).<br><a href="https://cosocial.ca/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LLM</span></a> <a href="https://cosocial.ca/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a> <a href="https://cosocial.ca/tags/ClimateCrisis" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ClimateCrisis</span></a></p>
Shufei 🧮<p><a href="https://merveilles.town/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LLM</span></a> <a href="https://merveilles.town/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a> is not an aberration, not a discontinuity. </p><p>The ills of AI are the direct outgrowth of a rot long infesting modern human society’s digital networks - spurious language, superficial text, fetishization of the vulgar, the stupidity of crowds and their writing by crowdsourced committee, data mining, misappropriation of organic communities by colonizing actors… AI is merely the synthesis of this utter abdication of responsibility by human society toward our own education. </p><p>It’s no wonder that the AI produced by this society would eviscerate deep perspective when the mass society’s ontological values were already utilitarian slop. As humanity continues to slide into a dark age of “post-fact” postmodernism, we should at least have the spine to recognize the fault is endemically our own.</p><p><a href="https://merveilles.town/tags/ML" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ML</span></a> <a href="https://merveilles.town/tags/ButlerianJihad" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ButlerianJihad</span></a></p>
KB<p>“We are in dire need of a better metaphor. Here’s my suggestion: instead of seeing AI as a sort of silicon homunculus, we should see it as a bag of words.”</p><p>⬆ This.</p><p><a href="https://www.experimental-history.com/p/bag-of-words-have-mercy-on-us" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="ellipsis">experimental-history.com/p/bag</span><span class="invisible">-of-words-have-mercy-on-us</span></a></p><p><a href="https://mastodon.scot/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a> <a href="https://mastodon.scot/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LLM</span></a> <a href="https://mastodon.scot/tags/Claude" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Claude</span></a> <a href="https://mastodon.scot/tags/ChatGPT" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ChatGPT</span></a></p>
Wulfy<p>How does "Parallel reasoning" work in the new <a href="https://infosec.exchange/tags/llm" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>llm</span></a> models?</p><p>It mirrors how humans collectively make decisions.<br>Multiple reasoning sessions report their findings, then they talk it out amongst themselves.<br>The results are impressive.</p><p><a href="https://infosec.exchange/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a></p>
Thomas<p>What always stopped me from experimenting with LLMs was that sending my prompts to OpenAI, Microsoft, Meta or Google felt wrong (they'd learn more about me than I about their LLM, and that's a non starter).</p><p>After the release of Gemma-3 270M last week, it's possible for anyone to experiment with LLMs locally. </p><p>I installed LM Studio on an old laptop computer, a Lenovo T460 (6th gen Intel i5, no GPU) running Debian 13. </p><p>The performance is OK, 17 tokens/s, with the fan idling.<br><a href="https://mas.to/tags/llm" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>llm</span></a> <a href="https://mas.to/tags/debian" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>debian</span></a> <br>1/2</p>
TrendingLive feeds
About