Google and the AI War
We see how Google is the company that has invested the most in AI and how it is more than ready for the AI war among Big Tech.
It has been a week full of news for Big Tech, with the announcement of the new experimental conversational artificial intelligence service, Bard, today in Paris, Big G demonstrated how artificial intelligence can be the focal point of many Google product development at the next level. Earlier in the day, Microsoft had announced the integration of ChatGPT into Bing (link to waitlist here).
Google’s Artificial Intelligence strategy: create an ecosystem with Responsible, Sustainable and Ethical AI at its core
Google has spent over $100 billion on Artificial Intelligence in the past few years, more than anyone else in the industry, and has laid the groundwork for innovations by other companies such as OpenAI, in fact ChatGPT was born out of the evolutions of Transformer the first open source text-based Generative AI model (from which Bert, MUM and others emerged).
Google’s strategy is to use AI at the center of its ecosystem, and the numerous projects such as LaMDA, PaLM for language, Imagen for images and the latest addition MusicLM for music are there to prove it as well as Tensor Flow regarding Machine Learning management.
Google’s news on Artificial Intelligence.
Let’s see what Google announced at yesterday’s event in Paris:
Google Search
Google is working on merging image, text, audio and video search and generation streams.
For Google, search is far from dead, it is just being transformed for the umpteenth time, and Google is declining it far beyond its homepage, both in the format it will soon be possible to do searches that include text and images together, and locally with an increasingly augmented reality Google Maps. All these efforts make Bard’s integration into Google search look like a mere complement.
Google Lens
“The camera will be your next keyboard.”
Google has spent a lot of money in this area and has done deep work on reading images by AIs, and Google Lens is already being used to do research 10 Billions times at month, with this upgrade the usage will increase a lot.
Google Maps
Google has used AI to create an immersive experience in Google Maps with an eye toward augmented reality and a future driving and navigation experience. See some examples in these images:
Google Bard — LaMDA
Google Bard is based on LaMDA of which you can see the 2021 Google I/O presentation above, (here is an article of Google about it).
Bard, will be released in a lightened version and made agile for external use in the coming weeks to “Trusted Testers” to understand its limitations and potential, but Google has already announced that APIs will also be released soon to make the language model available to companies and developers.
Initially, Bard will use a simplified version of LaMDA. This much smaller model requires significantly less computing power, allowing us to open it up to more people and thus receive more comments and suggestions. We will combine external suggestions with our own internal testing to ensure that Bard’s answers meet a high level of quality, security, and substantiation of real-world information. This testing phase, which we are excited about, will allow us to continue to learn and improve Bard’s quality and speed.
Here is an interview with Blake Lemoine the Google Data Scientist whose statements about a sentient AI (he was referring to experimentation with LaMDA) had caused a stir in the fall of 2022.
Here is what Matt Brittin president of Google EMEA said:
One of my favorite Google tools is Translate. Over 150 billion words are translated on the platform every day: and the three most commonly used phrases are “thank you,” “how are you,” and “I love you.” It’s a beautiful example of how people can connect — build bridges where there used to be language barriers. Since February 2022, we have seen translations of Ukrainian into other languages soar as displaced Ukrainians seek refuge across Europe and try to find support and connection in new communities. Translate now has over 100 languages online and, using the power of artificial intelligence, we are working to reach 1,000.
Today we shared some more ways we’re using artificial intelligence to make our services even more useful for billions of people and take many of your favorite Google products to the next level:
➡️ We’re making Search even more intuitive and natural. You can already use Google Lens, which lets you search what you see, directly from your camera or images, right from the search bar, with over 10 billion uses each month. Now we’re making it even easier to “search your screen” on Android through Assistant globally-so you can search for what you see in photos or videos through the websites and apps you know and love, without having to leave the app or experience.
➡️ And we are introducing multiple search globally, so you can search with an image and text at the same time. Soon, you’ll also be able to add “near me,” making it easier than ever to find what you want, how and where you need it.
➡️ Search is not the only tool receiving an update. On Maps, we are implementing immersive visualization, which uses artificial intelligence and computer vision to give you an idea of a place, before you visit it: as well as sharing information about weather, traffic and activity.
➡️ We are also launching Live View on maps, which uses artificial intelligence and augmented reality to help you find things around you, such as ATMs, restaurants, parks, and transit stations: simply by lifting your phone while on the road.
➡️ These are not the only updates Maps is getting. We’re also introducing great new features like Live View for interiors, help you navigate difficult places like airports, train stations, and shopping malls, and other features for drivers of electric vehicles that have Google built in.
I’m really excited about this next step in our AI journey, combining both our forward-looking ambitions and how we’re using AI to improve the tools you already have in your pocket.
🔗 A few links for further exploration
- Google Search: From images to videos, how AI is helping you search visually
- Google Maps: New ways Maps is getting more immersive and sustainable
- Google Bard: Google’s Blog post by CEO Sundar Pichai
