Google I/O 2024: The AI-Packed Keynote Recap

Google mentioned 'AI' one hundred and twenty-one times in Google I/O Keynote 2024 yesterday. Evidently, many of their products will be heavily powered by the AI systems that they continue to develop and design. And here are some of the things announced in this year's Keynote.

Google Lens and Photos get even smarter

With Lens, you can search based on images. As though that wasn't useful enough, they are adding a new feature that allows you to search based on video as well. That is, you could make a video of an object and inquire about it, and Lens will provide relevant information from the web.

image.png

You can now "Ask Photos" questions about your photos with Google Photos. This addition will make sifting through photos even easier and more efficient. Ask Photos allows Gemini to go over your Photos library to provide answers to your questions. As demonstrated by CEO Sundar Pichai, Gemini was asked about a plate number, and it responded with the plate number and pulled up a photo of it.

Project Astra

A new multimodal model by Google, Project Astra, is an AI that the company intends to be a more nuanced virtual assistant that will see, hear, and understand its environment and do things more akin to a human assistant, like remembering where your things are.

In a demo, Project Astra was asked to indicate when it sees something that makes sound through the phone's camera, and it did so when it spotted a speaker and described a part of the speaker, the tweeter that produces high-frequency sounds.

More impressively, the model was asked if it knew where the user's glasses could be, and the model described their location as being near a red apple on a table.

Veo, a video-generation AI model

As expected, Google released their own version of Sora, a video generation AI model by OpenAI, the one that took everyone by surprise when it was announced and demonstrated. Google presents Veo, their new generative AI model that runs on text, image, and video prompts and outputs 1080p video quality.


image.png

Videos can be produced in a variety of styles, like aerial shots or timelapses, and can be tweaked with more prompts. The company is already offering Veo to some creators for use in YouTube videos, but it is also pitching it to Hollywood for use in films. The Verge

Customisable Chatbots in Gemini

You can now customise Gemini with Gems, a custom chatbot creator. Gems will let users instruct Gemini on how it should respond and determine its specializations.

So you can make it whatever you want it to be by instructing it, and Gemini will be just that.

Circle to Search becomes more helpful

You can now circle something like a math problem on your screen or some other problem to solve, and you will get help solving it. The AI won't just solve the problem for you, however. It'll walk you through the steps, making it easier to find and understand the solution.

image.png

This particular feature comes in handy for students with their homework, helping them learn rather than cheat.

Upgrades in SynthID AI Watermarking

In efforts to keep the development and use of AI ethical, Google is improving the capabilities of SynthID for watermarking AI-generated videos and content made with the new Veo AI video generator. SynthID is now able to detect AI-generated video content as well.

https://www.youtube.com/live/XEzRZ35urlk?si=HEYHxmHVJHkbse_o

These new developments by Google and their products are going to change how we use our devices to interact with the world and augment us into being more efficient and effective with their capabilities. We can only expect things to improve and advance from here. What do you say?

By the way, make earnings with your content on Hive via InLeo while you truly own your account. If you're new, sign up in a few minutes by clicking here! And here's a guide on navigating.

All images belong to Google

H2
H3
H4
3 columns
2 columns
1 column
Join the conversation now