There’s been a lot of buzz around Artificial Intelligence of late, but Google has been an AI first company for several years now.  At IO 2023, Google has outlined just how their AI projects will now begin powering experiences across their vast array of products and services. 

At the heart of the Google AI announcements at IO is the new PaLM2 AI language model which they are putting up against the likes of OpenAI’s GPT-4. 

The new PaLM 2 model has improved logic and reasoning over the previous generation, but the model itself is more behind the scenes. Google says that PaLM2 is powering over 25 new products in their catalogue including Docs, Slides and Sheets in Google Workspace and of course Google Bard.

Bard

The most obvious is the ChatGPT competitor, Google Bard. Launched earlier this year in limited access, Bard will be open to everyone in 180 countries in English, Japanese and Korean with 40 more languages to come.

Bard will see improvements in AI, including benefits from partnerships with Adobe which will see AI generations in Adobe Firefly, as well as with partnerships with InstKayak, OpenTable, ZipRecruiter, Instacart, Wolfram and Khan Academy. 

Google says Bard is optimised for delivering code, making it a great resource for developers. Bard can offer code, including debugging and code explanations in 20 languages. Something interesting to try out if you ever had an app idea. 

Workspace AI

AI is changing a lot about the way we work, including generating new content and Google is bringing AI to Workspace apps introducing new features in Docs, Slides and Sheets to help you get work done. 

Google is bringing a ‘Side Panel’ to their Workspace apps, an area for you to enter in data for the AI to build from. Google is also introducing the idea of Prompts as well as show you Prompts, ideas that you can use 

Help me Write

Available in Gmail, Help Me Write is an iteration of the Smart Reply feature first introduced in Gmail. The new Help Me Write feature will be available in both Gmail and Docs help generate documents including Job Descriptions, Blog posts and more, or in Gmail it can help generate professional replies. The feature will also trawl through your previous emails, adding in all the relevant information including names and more.

You can even fine-tune the results with options to make it more formal, or make it more elaborate if you want and the AI does the work. 

Slides

Presentations will improve thanks to AI generated images in Slides presentations. Using the Side Panel, you can enter simple text prompts to have images automatically added to your slides. 

Slides will also incorporate some of the new Prompts including an option to build speaker presentation notes for you.

Sheets

AI will be able to generate tables based on scenarios entered into the Side Panel. Google used the example of a dog walking business, with the query being to setup a table with rates, times and the AI simply added in a table 

Labs

Google has always had great side projects, or little things you can test but slowly killed off these projects known as ‘Labs’ over the years. Well, they’re back and this is where you’ll be able to try out a lot of these new Google AI powered Workspace features. 

If you want to try out any of the Workspace AI features, you can head to g.co/labs to check them out.

Photos

Google Photos is getting more features powered by AI, including a new Magic Editor which expands on the Magic Erase feature launched on the Pixel 7 last year. 

Magic Editor will allow you to ‘reimagine your photos with AI’, basically giving you pro level photo editing skills without the training. Magic Editor will help you do simple things like removing a strap, but also allow you to do more advanced features like reposition people in photos, edit the background including changing the weather and generating data like the balloons in the shot below.

Google hasn’t given a launch date for Magic Editor, but has said it will be coming to Google Photos later this year.

Maps

Building on their Immersive Mode for Maps, Google this morning introduced Immersive View for Routes. This new mode will let you get deeper views of your routes. This includes checking out AI generated previews of your planned routes, including predicted traffic and weather conditions and Air Quality.

Unfortunately Immersive View for Routes won’t be coming to Australia first, with only 15 cities announced including Amsterdam, Berlin, Dublin, Florence, Las Vegas, London, Los Angeles, New York, Miami, Paris, Seattle, San Francisco, San Jose, Tokyo, and Venice.

Google Search is their core product and AI is also coming to search results with Generative AI. 

Generative AI will create snapshot answers at the top of the results with options for more information, as well as suggested follow up questions. You can even try out a conversational mode about the topic you’re exploring. 

The new Generative AI search functionality appears to be starting out as Google Labs Experiment called SGE but only as a US-only release initially for Chrome desktop and the Google App for Android and iOS in the coming weeks.