ai

Copilot Voice and Vision are now rolling out.

Eine Studie zeigt, dass ChatGPT & Co. auch strukturelle Einschätzungsmuster übernehmen und Vorurteile reproduzieren. Das reicht bis zur Körpertemperatur.

The 1 petaflop DGX Spark system runs AI models with 200 billion parameters locally for $4K.

Microsoft Viva Insights “helps people and businesses thrive with data-driven, privacy-protected insights.” It’s workplace spyware for your boss. [Microsoft; Microsoft] Viva Insights now has an exci…

Aleph Alpha sollte die deutsche Antwort auf OpenAI werden. Stattdessen verlor das Start-up an Relevanz. Jetzt muss Gründer Jonas Andrulis den Chefsessel räumen.

Anthropic research on data-poisoning attacks in large language models

Der Hype um Künstliche Intelligenz ist zur einzigen Wette an den Märkten geworden und gilt zu oft als Allheilmittel. Ökonomen und Finanzexperten schlagen Alarm. Ein Kommentar.

In reference to #LLMs, I want to really underscore the difference in speaking and writing. What is so astonishing about the speaking human is that by the time we are speaking fluently, we have adopted and incorporated a large amount of the cultural rules, norms and prejudices that piggyback on any given language. So when we open our mouths and speak, it comes out fast, often fairly unprocessed, and full of that baggage. 1/ https://mathstodon.xyz/@jonmsterling/115300004399760594

This is a comic about AI art.

Die Beratungsfirma hatte in der ersten Version des Berichts nicht angegeben, KI bei der Erstellung genutzt zu haben

Attached: 1 image And here we go. I never wrote this article, and yet it is cited here. https://www.liberalbriefs.com/geopolitics/power-without-mandate-the-perils-of-germanys-military-ambition And of course, it sounds so plausible, I seriously checked whether I had forgotten it, or the footnote was slightly wrong. The other sources are hallucinated too. #AIisnotresearch

Attached: 1 image Please teach your students not to do this. I understand there is a continuum between spellcheck/grammar check and LLMs, and there could be ways to use LLMs to improve wording or grammar, etc. But please teach your students not to use LLMs to “draft content”. I don't care if they disclose it (tbh, I prefer they don't!). I don't care if they "take responsibility for the content". The thing about writing is that it has a social function. The social function is *not* to cause text to exist. The social function is to communicate from your mind to my mind. Text is the means, not the end. If you must use an LLM, skip it and just send me the prompt. (No, I'm not going to link to the paper. The point is not to dunk on people who end up on the wrong side of a social question. The point is to do better.)

Attached: 1 image @derpostillon@mastodon.social hats geschafft. #Gemini benutzt ihn jetzt auch als vertrauenswürdige Quelle. https://www.der-postillon.com/2017/05/laserpointer.html #AI #LLM

: Artificial intelligence works when humans use it wisely

Das Internet wird überschwemmt von KI-generiertem Müll. Automatisierte Bots produzieren eine Flut aus KI-generierten Inhalten, der das Internet zu ersticken droht. Der Wissensraum wird mit Desinformation, Propaganda und synthetischen Bildern und Tönen geflutet. Werden wir bald nur noch Informationen bekommen, mit der die KI uns füttert?

Attached: 1 image An ad for Smalltalk/V from the 1986 Byte magazine. Marketing of "AI" hasn't changed much in the last 40 years. "Watching someone use an AI workstation is a vision of what the computer was meant to be. Fingers dance across the keys as windows dilate, shift, overlap, and disappear on the bit-mapped display. ... The interface vanishes, man and machine are one. Smalltalk/V brings that experience to your IBM-PC." #smalltalk #ai

OpenAI-Chef Sam Altman kündigt an, dass Sora zukünftig Copyrights besser achten soll. Zum Launch war die App ein Freifahrtschein für Urheberrechtsverletzungen.

Late in the afternoon of May 11, 1997, in front of the cameras of a small television studio 35 floors up a Manhattan skyscraper, Garry Kasparov sat down at a chessboard. The boisterous and temperamental World Chess Champion had never lost an official match, but entering the sixth and final game he was tied with his opponent at two games each (the third game having ended in a draw).

A Microsoft blog post about "vibe working" broke me.

: Use your home subscription with your work Microsoft 365 account

Attached: 1 video You don’t need to understand the math behind large language models in order to get that they are not minds.

Current AI models are unlikely to be able to make novel scientific breakthroughs, Thomas Wolf, co-founder of Hugging Face said.

OpenAI's planned AI data center projects would consume as much as the entire city of New York City and San Diego combined.

Meta plans to use data from your interactions with AI products to sell targeted ads on Facebook and Instagram.

A blog about computers, functional languages, Lisp, and Scheme.

Soundtrack: Queens of the Stone Age - First It Giveth Before we go any further: This is, for the third time this year, the longest newsletter I've ever written, weighing in somewhere around 18,500 words. I've written it specifically to be read at your leisure — dip in and out

Attached: 1 image “Microsoft says its (vibe working mode) in Excel has an accuracy rate of 57.2 percent in SpreadsheetBench, a benchmark for evaluating an AI model’s ability to edit real world spreadsheets” Imagine being a hiring manager and hiring somebody to use Excel who says they do their job wrong about half the time. Welcome to AI.

Tech companies are pouring hundreds of billions into data centers, taking on heavy debt, but current revenue is relatively tiny. Critics warn of a new dot-com bubble.

GenAI does some neat, helpful things, but it’s not yet the engine of a new economy — and it might not ever be.