Altman & Ive's secret AI hardware

Author

Jason - The AI Digest
September 24, 2024

Welcome, AI Enthousiasts!

In today’s AI newsletter:
→ AI Newsflash
→ Altman and Ive announce AI hardware device
→ Edit images in real-time with Krea and FLUX
→ Sam Altman on AI’s evolution: o1 to agents
→ Apple launches beta of AI-powered iOS 18.1
→ Snap’s new AR glasses are powered by AI
→ 5 Best AI PDF editing and management tools

Reading time: 6 minutes

AI Newsflash

Alibaba unveiled its Qwen 2.5 7B coder model, reportedly matching GPT-4's coding performance at a fraction of the cost, offering a competitive open-source alternative.

Amazon launched Amelia, an AI assistant for third-party sellers, designed to help resolve account issues and provide sales data insights. Currently, it's in beta for select U.S. sellers.

LinkedIn quietly introduced AI model training using user data, with users needing to opt-out via account settings and a separate form to fully exclude their data from future training.

Google committed $25 million to fund AI education programs, with a goal to train over 500,000 educators and students in AI skills.

Google also added 10 new voice options for Gemini Live, allowing users to personalize their AI assistant's voice.

Salesforce announced free access to its premium AI courses and certifications through 2025, available via its Trailhead platform.

Perplexity introduced a new "reasoning" feature for Pro users, offering up to 10 daily uses of OpenAI’s o1-mini model for solving puzzles, math, and coding challenges.

Former Apple design legend Jony Ive has officially announced a collaboration with OpenAI CEO Sam Altman on a new AI-powered hardware device, marking Ive’s first significant venture in tech since leaving Apple.

Here’s what you need to know:

  • Ive and Altman are working on an AI device designed to offer a computing experience that is “less socially disruptive than the iPhone.”

  • The project has secured private funding and aims to raise up to $1 billion in startup capital by the end of the year.

  • Ive's design firm, LoveFrom, is spearheading the device’s design, with support from former Apple executives Tang Tan and Evans Hankey.

  • While specific details are still under wraps, the device is expected to use generative AI to tackle complex user tasks that go beyond conventional software.

As OpenAI deepens its relationship with Apple through ChatGPT's integration on the iPhone 16, this collaboration between Altman and Ive stands out. Combining Ive’s renowned design expertise with OpenAI’s advanced models, like o1, could potentially bring about a truly useful AI hardware solution, something that most current options on the market have struggled to achieve.

AI Training
Edit images in real-time with Krea and FLUX

Krea AI is a cutting-edge image editing platform powered by AI, featuring real-time image generation through its Flux model. This allows users to quickly edit and enhance images on the go, with a substantial free tier available.

Step-by-Step Guide:

  1. Sign up or log into Krea AI and navigate to the "Generate" section.

  2. Upload the image you want to work with.

  3. Choose "Flux" as the model and input your desired prompt (e.g., "corgi with a red ball").

  4. Use the brush tool to specify areas where you want changes applied.

  5. Adjust the AI Strength slider to control how much the transformation affects the image.

Pro Tip: Play around with the AI Strength levels to find the perfect balance between your vision and the AI’s creative input.

At T-Mobile’s Capital Markets Day, OpenAI CEO Sam Altman delved into the new o1 model's advanced reasoning features, hinting at rapid progress toward developing AI agents, signaling that another breakthrough in AI may be close.

Here’s the breakdown:

  • The o1 model is touted as the first AI system with advanced reasoning skills, surpassing earlier models in handling complex problem-solving tasks.

  • Altman likened the current state of o1 to the ‘GPT-2 phase’ of reasoning models, implying that this is just the beginning, with a GPT-4 equivalent on the horizon.

  • He explained five stages of AI development, placing o1 at level 2 (reasoners) and suggesting that level 3 (agents) could arrive "relatively quickly."

While the gap between GPT-4 and o1 felt lengthy, Altman’s comments indicate that the leap from o1 to fully functional AI agents will be much quicker, thanks to the iterative nature of the models. If accurate, it may only take a few months before o1 vastly outperforms GPT-4 and other existing systems.

Apple has just launched the beta version of its much-anticipated AI software update, iOS 18.1, alongside the iPhone 16 release, bringing significant improvements to Siri and introducing a variety of AI-driven features.

Here’s what you need to know:

  • Available for iPhone 16 models and select premium iPhone 15 versions through the Apple Beta Software Program (users are advised to download with caution).

  • Siri now boasts enhanced conversational capabilities, along with a sleek new visual interface that features a rotating glowing light during interactions.

  • The update introduces several AI-powered tools, such as writing assistance, email summarization, and advanced photo editing, with more features expected in future releases.

  • Initially rolling out in the U.S., the AI features will expand to other English-speaking countries by December and reach additional languages and regions throughout the next year.

This beta offers a sneak peek at Apple's long-awaited AI enhancements, including major Siri upgrades and AI features aimed at delivering personalized experiences. As with all beta releases, users should proceed with caution, as this is not the final version and may contain bugs.

Snap has just introduced its fifth-generation Spectacles, standalone AR glasses powered by the new Snap OS, which integrate advanced AI features and elevate social interactions through augmented reality.

Here’s what you need to know:

  • These new Spectacles come equipped with multiple cameras and sensors, enabling multi-modal AI, contextual awareness, and full mobility—all while remaining lightweight and portable.

  • Built specifically for Spectacles, Snap OS lets users interact with My AI, launch immersive AR Lenses, and extend Snapchat experiences into 3D environments.

  • Snap also rolled out AI-driven video and portrait generation tools that create videos or images from a single photo, providing new ways to boost creative expression.

  • The Spectacles are now available to developers in the U.S., requiring a $99/month subscription with a 12-month commitment.

While this marks a significant leap forward for cordless, AI-enhanced AR devices, the limited 45-minute battery life and 46-degree field of view highlight areas for future improvement. Still, Snap is pushing the envelope in tech innovation, and it will be exciting to see how these advancements evolve over the next few years.

5 Best AI PDF editing and management tools

ChatPDF - An AI-powered PDF tool that enables users to engage in a chat-like interface with their documents, providing insights and answering queries directly from the content.

Foxit PDF Editor - A powerful PDF editor equipped with AI-driven tools for seamless editing, document management, and collaborative workflows.

AskYourPDF - An AI-driven tool that uses natural language processing to extract answers and insights directly from PDF documents.

AI PDF - An AI-powered PDF management solution designed to simplify document handling, enhance search capabilities, and improve user interactions with files.

Relayto PDF AI - A platform that leverages AI to transform how users interact with and manage digital PDFs, creating a more dynamic and intuitive experience.

That’s a wrap!

We had a lot to talk about, so let’s wrap it up. If you have any questions, feel free to shoot over an e-mail and we wil get you a response within 24 hours.

If you have specific feedback or anything interesting you’d like to share, please let us know by replying to this email.