AIX Academy

OpenAI continues to attract significant investments, raising up to $6.5B despite Apple reportedly stepping back from the latest funding round.

Backed by giants like Softbank, Microsoft which is adding $1B to its previous $13B investment, and Thrive Capital, OpenAI’s valuation has soared to $150B.
While the AI leader shifts towards a for-profit model, big names are still betting on its success.

What’s next for the ChatGPT maker in this AI revolution?
Stay tuned for more updates on the world of AI! 📰

Google has just launched two new advanced versions of its Gemini 1.5 AI models! 🎉

The 1.5 pro-002 model is setting records in complex math benchmarks, while 1.5-flash-002 focuses on improving instruction following. These upgrades are all about giving developers a more powerful tool to build amazing things.

Try the new models for free here: https://ai.google.dev/aistudio

Google DeepMind has introduced AlphaChip, an AI system that designs computer chips in just hours instead of months!

Using reinforcement learning, AlphaChip has already been used to create Google’s Tensor Processing Units (TPUs), boosting performance and speeding up the design process. 📈

With MediaTek adopting this tech, it’s clear AI-assisted chip design is the future. Google is also sharing AlphaChip’s model weights to encourage further development.👩🏻‍💻

OpenAI has finally added its enhanced Advanced Voice Mode (AVM) this week for ChatGPT Plus and Teams subscribers.

This update brings a more natural, personalized AI experience. 🎙️ AVM now comes with improved accent recognition, memory integration, and 5 new voices!🗣️

Sadly, some regions, including the UK and EU, won’t have access yet. If you haven’t seen AVM on your app, try uninstalling and reinstalling. We’re excited to see how this brings us closer to more human-like AI interactions!

Exciting news! 🚗 Alibaba Cloud and Nvidia are joining forces to bring cutting-edge AI to autonomous driving.

By integrating Alibaba’s Qwen AI models with Nvidia’s Drive AGX Orin platform, this partnership will power smarter in-car assistants and enhance driving intelligence with visual and environmental data.
They’re also working to bring this AI to Nvidia’s next-gen Drive Thor platform, making future cars even more advanced. This collaboration is a step forward in advancing the automotive industry.

Apple has released the beta of iOS 18.1, a version of the new iOS packed with AI features.

Exclusive to iPhone 16 and select iPhone 15 users via the Apple Beta Software Program, it comes with an enhanced Siri, smarter writing tools, and advanced photo editing.📷

Siri now has better conversational skills and a glowing interface during interactions. The update is U.S.-only for now, with plans to roll out globally next year. Remember, it's still in beta, so expect some bugs!

Download if you're feeling adventurous: https://beta.apple.com/

YouTube recently launched a new range of AI tools to support their creators, offering features such as text-to-video generation, an upgraded Inspiration tab, AI dubbing capabilities and AI-enhanced reply suggestions to comments.
With Veo, YouTube Shorts can now be generated as six-second clips from text prompts. The AI-powered "Inspiration" tool will help creators brainstorm their future video ideas. The AI dubbing feature has been designed with the aim of more natural intonation.
All these updates are designed to boost human creativity, not to replace it.

You can read more about these updates here: https://blog.youtube/news-and-events/made-on-youtube-2024/

Microsoft recently revealed the latest updates to its AI Assistant, Copilot, bringing new features across integrated applications like Excel, PowerPoint, Teams, and more. One of the most exciting additions is Copilot Pages, a shared workspace where teams can collaborate with AI, allowing multiple users to work together on interactive content in real time.

Copilot Agents, another new feature, works quietly in the background to automate processes, helping streamline productivity. And for those who aren't technical experts, the new Agent Builder makes it easy to create custom AI tools without writing any code.

With Copilot’s new GPT-4 integration, responses are now twice as fast, and overall performance has dramatically improved, promising a better, smoother user experience.
This upgrade is set to transform the way millions of users interact with their everyday tools, from spreadsheets to presentations.

OpenAI recently released ‘o1’ (formerly Project Strawberry), an AI model with advanced reasoning now available in ChatGPT for Premium and Teams users.🧠
Using reinforcement learning and chain-of-thought processing, o1 mimics human problem-solving, solving 83% of the International Mathematics Olympiad problems and ranking in the 89th percentile for competitive programming. With more accurate responses and powerful capabilities, it is suitable for complex tasks in mathematics, science and more.⚗️

Google just unveiled the feature 'Audio Overviews' in NotebookLM. It turns your notes, slides, PDFs and more into engaging AI-generated audio discussions.
Using Gemini 1.5's multimodal capabilities, Audio Overviews creates conversations between two AI agents that summarise and connect your topics, from up to 50 sources with each allowing 500,000 words.
This new feature is perfect for auditory learners, as well as gives other users a new way to engage with content like academic papers or presentations.🎧

OpenAI is set to launch Strawberry, a new AI model focused on reasoning, in ChatGPT within the next two weeks!
Unlike other models, Strawberry will need time to "think" for 10-20 seconds before responding, aiming to handle complex queries without advanced prompt techniques. Initially, it will handle text only, not images. Some early testers say the slight performance boost over GPT-4o might not justify the wait.🤔
Would you give it a try?