Opera Ads’ Post

It has nerver been easier to try out private Ai locally https://lnkd.in/dQ-7sHXi

Aria Browser AI upgrade Read more: https://lnkd.in/dQqh3p5i Aria's latest update: enhanced command line and Page Context Mode streamline browsing, offering tailored AI assistance without leaving your page.

Exciting News in the AI Space! 🚀 In a groundbreaking move, Opera has announced that its AI assistant Aria is getting integrated with Google Cloud's Gemini! 🌐 Launched last year, Aria is a powerful browser-based AI chatbot that stands out with its multi-LLM Composer AI engine. Unlike single-model chatbots like ChatGPT, Aria intelligently selects the best large language model (LLM) for each task, enhancing performance and versatility. Key Highlights: 🔹 Google Gemini Integration: With this new integration, Aria will provide the most current information with exceptional performance. 🔹 Conversational Responses: Aria now reads responses in a natural, conversational manner, similar to ChatGPT’s Read Aloud feature. 🔹 Enhanced Features: Last month, Opera introduced a free Imagen-2-powered text-to-image generator, further showcasing its commitment to AI innovation. 🔹 Diverse LLM Support: Aria leverages a range of models, including Meta’s Llama, Google Gemma, and Mixtral by Mistral AI, and supports over 150 local LLM variants. Resource: https://lnkd.in/gKBtn6FM #AI #Opera #Technology #TechNews

We’ve collaborated with Google Cloud to integrate Gemini AI models into Aria! Aria is powered by our multi-LLM Composer AI engine, which processes user intent and decides which model to use for which task. With Google’s Gemini - the company’s most capable model yet -  now in the mix, our users get the latest information, at high performance! This collaboration also sees Aria’s free image generation capabilities improved by utilizing the Imagen 2 model on Vertex AI. Aria will be able to read out responses in a conversational-like fashion, thanks to Google’s ground-breaking text-to-audio model. This update follows our AI Feature Drop program, which allows early adopters to test the newest, experimental, AI innovations in the Opera One Developer version of the browser.   Read more here: https://opr.as/3ba341 

Big news in the world of AI! I’m all for protections. Working with AI Consultants allows for expert advise in navigating both the use and regulation of AI. #AI #AIStrategist #AIConsultant https://lnkd.in/eUFV2Jpq

An interesting company, Muzaic is using AI to create soundtracks for your life! 🎶 Overarching Trends: ✅ Specialization of AI models (text, audio, video, lips, emotion...etc) ✅ Lower barrier to entry for creativity ✅ Polarization of content (really good vs. really bad, little in between) https://lnkd.in/gSGQYjHc #AI

UMG chairman-CEO Lucian Grainge spoke about it at length in a letter to staff obtained by Variety. In it, he gives very broad explanations of the terms, including (in the case of AI), “protecting human artists from being economically disadvantaged by AI; guarding against the use of AI-generated deepfakes; and requiring transparency in how AI companies train their models.” #AI #Musicbiz #musicindustry https://lnkd.in/grbBKrdU

Exciting news for AI enthusiasts! Opera is now letting users download and run large language models (LLMs) directly in their browser. This is a game-changer for anyone who wants to experiment with AI locally and explore the capabilities of different models. Here are some key points: Over 150 models available, including big names like Llama from Meta and Gemma from Google. Requires the Opera One developer stream and utilizes the Ollama open-source framework. Models can be large (over 2GB!), so make sure you have enough storage space. What are your thoughts on this development? Is this a step towards a more AI-powered browsing experience? #AI #MachineLearning #OperaBrowser

As the intersection between AI and HI grows smaller protections need to be established. Although not perfect it is the long awaited first step.

Last year everyone got very excited about #GenerativeAI. This year we keep seeing headlines about #GenAI used poorly. In the last weeks... 1. A really poor, and probably unnecessary use of #AI generated imagery for a simple FB ad. The image is just clearly flawed & obviously fake - its lazy. https://lnkd.in/eWvsXYis 2. The probably now infamous use of AI for the #advertising of the "Wonka experience" https://lnkd.in/emP4RSTv 3. Trump supporters target voters with AI fakes: https://lnkd.in/eZaawZdV 4. An example that impacted a friend of mine thats a little different: having submitted an application to an organisation: he was asked to rewrite it without the use of AI. The problem? He didn't use AI. The organisation appeared to have used AI to flag applications generated using AI... and clearly got it wrong. 5. My own experience... I used #ChatGPT to generate Anagrams for a pub quiz. Seemed like a simple enough ask. I didn't check the anagrams... and when the quiz came along we had some very confused quizzers as the answers were being read out... as the anagrams weren't even close to being correct. Significant egg on my own face. The threads that tie these together? 1. AI is not a replacement for human diligence. It can speed things up - and save some money, let you do things you could never have done before - but you as a human are still responsible for the output. Don't assume AI will get it right. Keep a human in the loop. 2. If you use AI to try and pull the wool over peoples eyes you will be found out. Just don't do it. Use AI to express reality in new and creative ways - don't use it to invent new realities. 3. We are clearly already developing a crisis of trust caused by Gen-AI. Advertising that is misleading. People being paranoid about applications being AI generated. Politicians being targeted with fakes both to their benefit and detriment. (and this coming ahead of a major election year: We are all going to have to be very very careful what we share or repeat). AI is an amazingly powerful tool - but humans need to stay in the loop. If you're running a company - consider how you deploy it to enhance or possibly enable you to redeploy your staff - not replace them. And finally... any politician who suggest that AI doesn't need to be regulated (Looking at you Rishi Sunak) is living in a dream world. As a #LibDem I'll be advocating for the accelerated use of AI in our country - but with pragmatic safeguards and regulation in place.