🎵 Everyone's talking about this new music tool - find out why!

AI-HOI and good morning, AInauts!

Today we’ve put together a colorful AI goodie bag for you, with the most important news, tools & hacks.

That's what we have in store for you today:

• 🎵 Everyone's talking about this new music tool - find out why!

• ⛅️ Google's cloud conference - top or flop?

• 📰 AI-News-Quickie: The HAI-lights

• 🤭 AI-Fun - camouflage sweater or anti-stress

Let's go!

🎵 Everyone's talking about this new music tool - find out why!

⚠️ Read on at your own risk: you'll probably spend the next half hour playing around with this great new tool.

Curious? Alright, the world belongs to the brave ones: Welcome to the new era of music creation!

To give you a taste, here's a song by us Udio (video via Canva). The song is in German, because Udio can do so.

We recently presented Suno.ai and Stable Audio, but honestly: they can pack their bags ... Udio plays simply in a different league!

You can use it to create music in seconds. Simply define a few settings, add lyrics if you wish, and the platform will create two different short tracks! You can then remix them, make them longer and share them with the community.

You can create up to 1,200 songs per month for free!

It's definitely fun to experiment with it - and even serious musicians will get a kick out of it, because it offers possibilities for fine-tuning and instrument selection.

Behind the platform are four former employees of Google DeepMind. And the company just got a cash injection of $10 million dollars from angel investors such as rappers will.i.am and Common, and Mike Krieger (Co-Founder & CTO of Instagram).

And there are no limits to your creativity, as these examples show:

Our take: A Udio song that storms the charts? Sure, but ...

How will Udio shake up the music industry? Only time will tell, but it's pretty certain that an AI song will storm the charts sooner rather than later. Is this the end for all musicians, singers, songwriters, producers, ...?

Wait, wait - let’s take a step back!

AI-generated music will certainly be used as background music in some areas (after all, it doesn't cost any royalties), but it won't replace our favorite songs any time soon. The (AI-generated) Spotify playlist remains the first choice.

But AI tunes also open up completely new real-time use cases, as this video impressively shows (reminds us of this track)!

Above all, the platform is living proof of the unstoppable influence of AI on all creative processes.

It forces us to ask tricky questions: Is AI here an artist or just a tool? Will AI music replace real artists?

The crux of the matter is that the general demand for music will not increase - but the supply will. The stars have less to worry about, but the oversupply makes it difficult for lesser-known artists in particular to be heard ...

⛅️ Google's cloud conference - top or flop?

From hip start-up to powerful industry giant. Let's start with the facts and highlights from the Google Cloud Next conference - and then we'll take a look at the AI features you can already use.

First things first: Google presented an impressive range of new generative AI solutions at its Cloud Next conference.

• Vertex AI Agent Builder: The new best friend for all coders - With the new Agent Builder, you can develop your own AI agents in a low-code/no-code environment - from customer service to purchase advice, etc., everything is possible (101 customer cases here and here) - at least that's their promise.

  We were excited by this announcement and immediately tried it out - with sobering results. The usability needs to be improved, there are only a few helpful templates (three, to be precise), you have to provide a lot of input and examples, and the provided default models are not really the best. But there is a positive side aspect too: it can only get better!

• Gemini 1.5 Pro: Maximum context window - Geminis gigantic context window of 1 million tokens is impressive, and you can already use it!

  This massive memory ensures that the model has a memory like an elephant and thus offers real added value for certain use cases.

  In addition, you can give multimodal content as input (text, image, audio, video, code). The proof: almost all Harry Potter books can be uploaded and analyzed in one go!

• Imagen 2: Images that live - With the new text-to-live function, you will soon be able to create not only images, but also create animated video clips from text - similar to Runway, PikaLabs, etc. But does it come close to those? Take a look and judge for yourself ... The images are also provided with a digital SynthID watermark.

• Gemini Code Assist: The helper for developers - This tool understands code, and thanks to the 1 million token context window, it has a real advantage over other tools. But the important market test is yet to come: will developers switch to it or stick with the market leader Github Copilot (Microsoft)? It will be usable from July 11, then we'll know more ...

• Gemini in the Google Cloud: A feast for developers - Gemini is now also available for databases, Android apps and other areas, and Google also supports open source efforts.

• AI magic for your photos: professional tools in your pocket - Users of the newest Google Pixel phones already know (and appreciate) the various AI features such as the "Magic Editor", "Magic Eraser" or "Photo Unblur". These will be rolled out in Google Photos to (almost) all users from mid-May. We love these features, thumbs up - very helpful!

• AI Meetings Add-on: More productive with Google Workspace - With "Take notes for me", you can focus on the meeting, and let the AI take notes. For $10 per user and month, you can have Google Meet automatically create notes and translate them into many languages. And more ...

• Google Vids: the simple video editor for all of us - Google Docs, Sheets, Slides and Co. are getting a new addition with Google Vids - a video editor which allows you to instantly create appealing content. Voice over, stock videos, animations etc. PowerPoint Google Slides on steroids, so to speak. Exciting - and available for testers starting mid-June.

Our take: Start by using what is already available to you

Google wants to make one thing clear above all with these innovations: that it wants to be a leader in the development of AI and is pulling out all the stops to achieve this. Whether it succeeds remains to be seen. At least the announcement wasn't overshadowed by OpenAI this time ...

Of course, such events are always intended to stake out the playing field - and regular users like us usually have to wait months before we can get our hands on it.

So here are a few tips on what you can already use today (if necessary with a VPN and US IP address).

Try out Google's AI Lab experiments at https://labs.google!

There are some highlights in Google Labs, such as AI-powered search, practical Chrome AI features and NotebookLM(it's really good!).

Use Google Gemini and the integration with Google Docs, Sheets, Slides, GMail, etc.

You can generate images for presentations at the touch of a button or remove backgrounds, create texts, analyze tables, answer and summarize emails (currently only in English), create charts, etc.

The tools are there, you just have to try to integrate them into your processes. Google has even created a Gemini manual for us (PDF here).

The crux of the matter is that humans are creatures of habit and it takes time, repetition and a sense of achievement for this to become second nature.

And yes, we know that this is sometimes frustrating because it doesn't always work the way you want it to at first go. Nevertheless, this is the only way to get to know the possibilities and limitations.

📰 AI-News-Quickie: The HAI-lights

That was just the most important news from Google. But there has also been a lot going on in the rest of the AI universe, so here are a few more tidbits for dessert!

OpenAI Updates

• Remember the interview of OpenAI CTO Mira Murati that pretty much was a sh*tshow? The Wall Street Journal has now published the whole clip for paying subscribers. But if the link to the video is in the source code of the page, you can also view view it directly 😎.

• OpenAI has released an update that wakes ChatGPT Plus from hibernation and gets it ready for spring. Did you notice a difference yet? Apparently, more internet search results are also being used to provide more up-to-date answers.

• In the Arena Leaderboard (you can use all the important models for free and pit them against each other!), GPT-4 Turbo regained the throne after briefly relinquishing the crown to Claude Opus.

• You can now use Siri and Shortcuts with an iPhone or iPad in the ChatGPT iOS app.

• Developers can also rejoice: GPT-4 Turbo with Vision can be used via API! And GPT-5 is apparently coming soon.

• But things are starting to get dicey with a whole host of lawsuits OpenAI has on its hands ... over two dozen high-profile lawyers have been hired, and lobbyists in Washington and Brussels are feverishly working the politicians.

• OpenAI also wants market share and a bigger slice of the enterprise pie - and is trying to steal Microsoft's thunder. Satya won't be happy about that ...

Microsoft, Meta and Google

• Microsoft AI (recently under the new leadership of Mustafa Suleyman) has announced a new AI hub in London - which is also home to Google Deepmind. The "war for talent" has started ...

• Google Deepmind CEO Demis Hassabis has confirmed behind closed doors that OpenAI has a real head start with Sora, and that it will not be easy to catch up with.

• Meta plans to release with Llama 3 in the next few days. As die-hard Ray Ban Meta fans, we are particularly looking forward to this upgrade!

• In Chile, the first surgery was successfully carried out with the support of the Meta Quest headset. AR, baby!

• Meta has also confirmed that new proprietary processors are going to be used as early as this year.

• Google has announced new CPUs for its own cloud under the name Axion.

• As we can see, chips are the new gold - and are being diligently subsidized from all sides!

Market & Stocks

• We don't have any stock recommendations - so you'll have to guess our portfolio - but here and here and here and here are a few interesting articles on the subject.

• And maybe you'll also find something suitable companies in the Forbes AI Top 50 list?

• Forbes, together with Sequoia and Meritech, have analyzed how the top 50 AI companies are driving a productivity revolution in various industries.

Industry & Thought Leaders

• Mistral has announced a new model once again "dropped a torrent" - and taking pole position on the openLLM leaderboard!

• xAI's Grok has also introduced version 1.5 with the Vision feature - and the benchmarks are on a par with GPT-4 Vision!

• Elon confirms: "AI will be smarter than the smartest human next year!"

• Until then, you can still sign up for the monitored self-drive Tesla, for a mere $99/month (only in the USA) - or soon be chauffeured around in a Tesla Robotaxi.

• TikTok wants to launch AI avatars that can create ads.

• The video platform Viggle.ai goes viral. You can give it a try here.

• AI luminary Andrew Ng is now part of the Amazon board.

• Apparently Adobe Firefly used AI-generated images from Midjourney. There will be repercussions! 

• To train a video model, Adobe is now buying videos of everyday scenes for $3/minute. Definitely a better way than burning your fingers on YouTube...

• At the world's first AI music tournament, various streamers competed against each other with the help of Suno AI.

🤭AI-Fun - Camouflage sweater or anti-stress

Oh ... the camouflage sweater brings up a new idea for a print-on-demand business. And if you need a bit of "anti-stress therapy" after all this news, we've also found the right animations for you.

We made it! But no need to be sad. The AInauts will be back soon, with new food and a surprise.

See you soon, Fabian & Reto

Follow us on Twitter & LinkedIn!

Your feedback is essential for us. We read EVERY comment and feedback, just respond to this email. Tell us what was (not) good and what is interesting for YOU.