🤖 Copilot Gets Eyes and a Voice

Plus: OpenAI DevDay Releases and New AI Models Announced

October 03, 2024

Hello AI Enthusiast,

This week in AI, Microsoft’s Copilot gets new features, Google integrates AI into its latest Chromebooks, Alibaba and Nvidia release advanced AI models that compete with the best, and OpenAI presents new developer tools at its DevDay. Let’s dive in! 👇

The Big Picture 🔊

Microsoft Unveils New Copilot AI Features

Microsoft has launched new Copilot features, including Copilot Vision, which can analyze and respond to questions about what's on a user's screen. Additionally, Think Deeper enhances Copilot's reasoning abilities for complex problem-solving, and Copilot Voice introduces voice interactions with spoken responses, along with a personalization feature to tailor recommendations.

Helin Yontar
Helin YontarCPO and Polyglot

This time Microsoft is carefully rolling out new Copilot features, showing they've learned from past missteps. Interestingly, they're emphasizing privacy while also introducing personalization, a bit of a contradiction. Could we call this "privacy-washing"?

Despite not being first to market, Microsoft has advantages: a wide ecosystem, huge user base, and procurement strength.

The core issue is training. We have the tools, but users need to learn how to apply them to their needs. Our practical AI training helps you and your team automate the small, annoying tasks. Schedule a call with us to see how we can help.

Copilot's new features have just been released, but how are you using Microsoft's AI assistant so far?

Login or Subscribe to participate in polls.

New Chromebooks Come with Integrated AI

Google’s new Chromebooks introduce the Quick Insert button, which offers instant access to AI tools like “Help me write” and file retrieval from Google Drive. Other features rolling out include Live Translate, Welcome Recap, and the Recorder app for real-time transcription, enhancing productivity across devices.

Gioele Mottarlini
Gioele MottarliniCOO and Image Addict

Google's strategy: offer affordable Chromebooks (starting at $350) with built-in AI tools, then hook users into Gemini subscriptions. A digital Trojan horse!

This strategy reflects a bigger change in tech business models. It's no longer just about selling a device and leaving you with the electricity bill to run it. Now, companies foot the bill for running AI models in the cloud, which means they need to recoup costs through subscriptions.

Alibaba and Nvidia Release New Advanced AI Models

Alibaba and Nvidia have both unveiled new AI models. Alibaba's Qwen 2.5 series offers both open-source and proprietary options for various applications, including a text-to-video tool. Nvidia's NVLM 1.0 is a multimodal model excelling in visual and textual tasks.

Gianluca Mauro
Gianluca MauroCEO and AI Rockstar

While most companies are reaching similar capability levels, the real differentiators now are integration, user experience, and procurement strategies.

Interestingly, those benchmark tables companies use to compare models, whenever they launch a new one, are losing relevance. Even AI experts are shifting focus from raw performance metrics to practical applications and user-centric features.

OpenAI Unveils Realtime API and New Developer Tools at 2024 DevDay

At its 2024 DevDay, OpenAI introduced several new tools for developers, including a public beta of its "Realtime API" for building apps with low-latency, AI-generated voice responses (basically ChatGPT’s Advanced Voice Mode). OpenAI also launched vision fine-tuning for GPT-4o, a model distillation feature to improve smaller AI models, and a prompt caching feature to reduce costs and improve latency by 50%.

Bits and Bobs 🗞️

  • California Governor Gavin Newsom signed 18 AI-related bills into law, addressing key issues like AI risk analysis, privacy, deepfake pornography, and election deepfakes.

  • At ZapConnect 2024, Zapier unveiled new features, enabling users to build custom software and powerful AI-integrated workflows without coding.

  • Reddit has introduced an AI-powered translation feature allowing posts and comments to be auto-translated into a user's preferred language.

  • New research shows that image-recognition bots can now solve reCAPTCHA v2 challenges with 100% accuracy, matching human performance.

  • Microsoft plans to relaunch its AI tool, Recall, in November with new privacy measures, including making it opt-in rather than default.

From Our Channels 🤳

If you haven’t tried ChatGPT’s new voice mode yet, you’ve probably seen someone using it online. But do you actually understand the technology behind it and why it’s better than the previous version? Let Gianluca explain it to you in simple terms.

@gianluca.mauro

One of the craziest technologies out there, explained. #Ai #learnontiktok #artificialintelligence #business #machinelearning #product #ux ... See more

LOLgorithms 😂

The moment the two podcasters of NotebookLM realize they are not… human.

That's a wrap on our newsletter! Here’s a quick recap before you go:

Catch you next week! 👋