Google rolls out new features for generative AI tool Bard

Alphabet-owned Google has introduced new features for its generative artificial intelligence platform Bard to make it more accessible in Arabic and other languages.

Arabic-speaking users can now upload images on to the platform with Google Lens, get Search images in responses and modify Bard’s responses, the company said on Tuesday.

The platform has also been integrated with other Google services such as Gmail, Docs, Drive, Maps, YouTube, Google Flights and Google Hotels. This feature is only available in English now, the company said.

To quell concerns about data privacy, Google said it will not use this information to train Bard’s public model, nor will it be seen by human reviewers.

“We are delighted by the positive feedback we have received from Arabic speakers using Bard. We are committed to continuously improving Bard and bringing more capabilities to Arabic speakers in the future,” according to a quote generated by Bard.

The AI service is powered by Google’s language model, PaLM2, which has been further developed based on users’ feedback “to become more intuitive and respond with greater quality and accuracy in all languages”, Google said.

Bard is Google’s generative AI experiment, launched in English, in March. Bard has since been expanded to more than 40 languages, including Arabic, Chinese, German, Hindi and Spanish, in July.

The Alphabet-owned company, which is wrestling Microsoft-backed Bing and ChatGPT for a greater share of the generative AI market, also introduced Bard in 59 places, including Europe and Brazil.

The conversational AI service focuses on creating ways to engage with information, from language and images to videos and audio.

Bard comes with generative AI capabilities. For example, users can ask Bard to give them tips to reach their goal of reading more books this year, explain quantum physics in simple terms, write a customised job description, draft an invitation for a Halloween-themed birthday party or quickly write an outline for a blog post.

The tool is powered by Language Model for Dialogue Applications (LaMDA) technology that was launched by the company two years ago. AI-led LaMDA comes with next-generation language and conversation ability.

In March, the company opened limited public access to select consumers in the US and the UK in English.

In May, Google removed the waiting list for Bard at its annual conference and introduced the technology in nearly 180 countries and territories, and added Japanese and Korean languages.

However, Google avoided launching the service in Europe in the initial phases.

In February, Google's parent company Alphabet lost $100 billion in market value after Bard made a factual error in a promotional video.

With the updated features, Bard is making its existing English language features available to more than 40 new languages, as well as launching new capabilities in English and soon to other languages, Google said on Tuesday.

As part of the new features, Bard allows people to use images in prompts using the capabilities of Google Lens, a product that recognises and analyses objects, texts and images. Users will also be able to receive images from Google Search in Bard’s responses.

Users can also modify Bard’s responses by changing the tone and style to five options: simple, long, short, professional or casual, according to the statement.

They can also continue their conversations with Bard, deciding whether to ask additional questions about the topic or use it as a starting point for new ideas.

This feature is available if the user has a public link to another Bard chat, Google said.

The global generative AI market is expected to be worth $188.62 billion by 2032, growing at an annual rate of more than 36 per cent, from $8.65 billion last year, data from The Brainy Insights market research company showed.

The North American region dominated the market last year.

Generative AI could also drive a 7 per cent – or almost $7 trillion – increase in the global economy and lift productivity growth by 1.5 percentage points over a 10-year period, Goldman Sachs estimated.

However, the US investment bank expects a more delayed impact on emerging market economies.

Globally, generative AI could cost the world the equivalent of 300 million full-time jobs to automation across major economies, the report indicated.