artificial intelligence will improve searches and Maps, Bard will be available in 180 countries (but not in Italy)

FROM OUR REPORTER
MOUNTAIN VIEW – «It has been a busy year for artificial intelligence (AI), today we will show a useful use of this technology for users and for work». With these words Sundar PichaiCEO of Alphabet, opened the proceedings of Google I/O 2023, the annual conference for developers in Mountain View (Here the summary with all the highlights). And he mentioned in the very first moments of his speech the need for a “responsible approach”: a way of responding to those who have raised concerns about artificial intelligencemonopolizing the debate on the topic.

Beware of artificial intelligence

Only a few days ago (May 4) US Vice President Kamala Harris gathered in Washington the leaders of the main companies dealing with the issue, reminding them that companies have «the ethical, moral and legal duty to guarantee the safety» of their products. But – as it was easy to predict – safety in the research and use of artificial intelligence was the common thread that linked numerous speeches, such as that of the senior vice president James Manyika, who leads the area of ​​Google called Technology and Society. “We believe it is essential to have a responsible approach to artificial intelligence – he said -. Developing AI responsibly must be a collective effort, involving researchers, social scientists, experts, governments, creators, publishers, as well as the people who use it every day».

Manyika added that we need to understand that AI is an emerging technology still in the making, and that it detects “natural tension” between “bold” and “responsible” and for that they must be accountable from the start. «It is important to be aware that AI also has the potential to make some societal challenges worse existing ones, such as the influence of unjust prejudices, and to create new ones as it becomes more advanced and new uses emerge».

Against fake news

The problem is not (only) the potential bias, but also the disinformationwhich the senior vice president himself mentions in a passage, explaining how yes, generative AI allows you to create new content very easily, but at the same time it also leads to the birth of new tools for evaluating information online: one of these, About this image, will be introduced in the search engine in the coming months to provide context to images shared online.

Here is Bard

Among the main innovations presented by Google during the keynote, one of the most eagerly awaited is certainly the implementation of Bard, the Mountain View chatbot launched less than two months ago and a direct challenger to ChatGpt (our test here). It will be available in 180 countries in English (until now it was only for the US and the UK), it will soon add Japanese and Korean, and soon, they say on stage, it will work with 40 different languages. It will work together with other tools, such as Adobe Firefly: in this way it will be possible to ask Bard for create new images starting from a caption. And it will use the photos to make the answers more complete: for example, you can ask Bard what to see in one day in Washington and the chatbot will answer in a more complex way. Big G also presented Palm2the latest version of its large language model (Llm) which is also able to recognize errors in the code by translating them between multiple languages ​​(a clarification: from today Bard runs on Palm2).

The AI ​​enhances the photos by itself

Google Photo will be enriched with Magic Editor, a feature that will allow you to edit photos without necessarily having to have professional editing knowledge. This too is made possible by artificial intelligence (including generative AI) which will allow you to modify parts of a shot, such as the subject or background, to achieve the desired result. For example, the subject of a photo can be repositioned at the best point, to obtain a more harmonious composition. Or even remove objects that spoil a shot.

How Maps changes

But artificial intelligence will also enrich Maps, thanks to the Immersive View applied to the routes: this will allow us to view the road we will have to take before setting off. How is it possible? Merging billions of Street View and aerial images to create a digital model of the world. It will then be possible to preview cycle paths, sidewalks, intersections and parking lots along our journey. Immersive View for Routes will roll out in the coming months for Amsterdam, Berlin, Dublin, Florence, Las Vegas, London, Los Angeles, New York, Miami, Paris, Seattle, San Francisco, San Jose, Tokyo, and Venice.

The new research

It will also improve the researchthanks to’Generative AI: This will allow you to ask more complex questions and receive an overview of key information as an answer from the search engine. Duet AI will arrive for Google Workspace, which will allow thanks to the news on Gmail to send complex response emails using only a few words as suggestions, but also to create original images from text, directly in Google Presentations.

And here are the devices

At the end of the event, after showing the new “cheap” Pixel 7a smartphone (here our preview review) and the Pixel Tablet, Google dropped the ace with the most awaited product: Pixel Foldits first foldable smartphone (here is our insight). It will cost $1,799 and arrive next month. But only in the USA, Germany, the United Kingdom and Japan.