Courts rule in favor of LLM training in copyright cases

🚧 Attention, language enthusiasts and AI aficionados! 🚧 Ever wonder what barriers large language models face in their quest for linguistic domination? Our latest blog post dives deep into the perplexing challenges these behemoths encounter. From limitations in understanding context to the nuances of human communication, we’re peeling back the layers on what stands in the way of AI becoming the all-knowing oracle we dreamt it would be. Curious to know how these hurdles impact the evolution of AI and what it means for the future of communication? Don't miss out on this insightful read that's sure to make you think (and maybe chuckle). Check it out here: https://ift.tt/E3Pob9Z

BGAD News: Tilde AI Releases TildeOpen LLM: An Open-Source Large Language Model with Over 30 Billion Parameters and Support Most European Languages: Latvian language-tech firm Tilde has released TildeOpen LLM, an open-source foundational large language model (LLM) purpose-built for European … http://dlvr.it/TMwt1f bgadconsulting.com

Hugging Face has "TildeOpen LLM is an open-source foundational (base) language model built to serve underrepresented Nordic and Eastern European languages. Developed with European Commission funding and trained on the LUMI supercomputer, this 30B+ parameter model addresses the performance gaps that speakers of 19 focus languages—representing over 165 million people—face with existing AI systems." https://lnkd.in/e5Qye97Z. #tildeopen #largelanguagemodel #nordiceuropean #easterneuropean #lumi #artificialintelligence #europeanunion #huggingface

I sat down with Rebecca Avery to talk about her new guest essay for Engines of Change: “We Talk Too Much About AI.” She makes the case that it is time to move past the usual AI debates and start asking bigger questions about localization, accessibility, and the influence media leaders already have in shaping what comes next. Watch our conversation here, then head over to Engines to read the full piece: https://lnkd.in/gPQXQffQ

Is the em dash a tell-tale sign of AI? 🤖 The short answer... probably not. We looked into this debate to give English teachers and internet sleuths the real scoop. https://ow.ly/elCI50WYjCC

AI agents – capabilities, challenges, and what the future might look like. Engage in complex goals w/o instructions across voice, video, and/or text with a natural language prompt. https://lnkd.in/gB5TcvR4 Melissa Heikkilä

Human-first language services will never be able to compete with AI when it comes to prices, but that doesn't mean they're going to lose their edge. The 11th edition of CSA Research's CEO Insights reveals how LSPs are adapting, and although CEOs are all taking their own approach, their strategies have one thing in common. They're continuing to put people at the centre. Reframing professional language services as 'responsibly sourced' is not only catching the eyes of ethically minded clients. It's also shining a light on the processes behind the work clients receive, which is helping them to understand why professional services are priced the way they are. Source: https://lnkd.in/eq66T7B5 #FutureOfLanguageServices #LanguageServices #MachineTranslation #AddingValue

Agentic AI presents new scientific frontiers, like deciding what language agents should speak and modeling agentic negotiations. In this analysis, Penn professor and Amazon Scholar Michael Kearns explains how researchers will need to grapple with questions around sharing context without compromising privacy and understanding users' commonsense policies: https://amzn.to/4piTFzp

In the race to shape the future of AI, language matters, both literally and politically. That’s why European initiatives like #Apertus (Switzerland’s open-source LLM) and AI Sweden’s collaboration with OpenAI’s open models are worth paying close attention to. Today’s session by Magnus Sahlgren and hosted by Henric W. (AI Sweden) offered a timely deep dive into what strategic value language models can create when aligned with local context, multilingual capability, and digital sovereignty. We’re seeing a shift: → From importing models to building and fine-tuning them closer to our needs → From passive adoption to active ownership of AI infrastructure → From English-first to Europe-aware LLMs The question is no longer if we can build large-scale models in Europe, it’s how we do it well, and who we bring with us in the process. Curious to hear: → What would you expect from a European LLM strategy? → Where do you see the biggest gaps today? #AI #LLM #EuropeanAI #TrustworthyAI #Apertus #OpenAI #AISweden #DigitalSovereignty #LanguageModels #OpenModels #AIInfrastructure

"In a paper published by the Royal Society last year, three researchers from the Oxford Internet Institute made the provocative argument that generative AI models should have a legal duty to tell the truth, just like, in different ways, lawyers, advertisers and doctors are obliged to do through their professional codes." https://lnkd.in/dk94zbrv The paper: https://lnkd.in/dSZzamDy