Google launched its AI chatbot in Canada after a delay tied to the company’s standoff with Ottawa over online news.
The web search giant expanded the rollout of its chatbot, now called Gemini, to Canada along with an iOS app and a fleet of generative AI tools set to launch soon. Gemini was previously called Bard.
The announcement comes months after the company excluded Canadian users from its chatbot’s “biggest expansion” – to more than 230 countries and in more than 40 languages – in July. Canadians can now access Gemini in both English and Québécois French.
“Our team in Canada was still working to find a constructive resolution on Bill C-18,” Jack Krawczyk, product lead, Gemini experiences at Google, told The Globe and Mail.
In November, the federal government and Google reached an agreement over the Online News Act – previously known as Bill C-18 – after months of tense negotiations, including Google’s threat to block Canadians’ ability to search for news on its platform. The tech giant agreed to pay $100-million a year to Canadian news organizations.
“Fortunately, the Government of Canada has committed to addressing our core issues with the bill…that really helped us clear a path to launch Gemini in Canada,” Mr. Krawczyk said.
Google parent company Alphabet Inc. GOOG-Q announced the launch of its chatbot service to test users in February last year, with a wider rollout to users in the U.S. and Britain in March. The product rivals ChatGPT, a similar generative-AI chatbot from Microsoft-backed MSFT-Q AI company OpenAI, which launched in November, 2022, and expanded to include a mobile iOS application in May last year.
Google’s Gemini app is available for iOS and Android as of Thursday, with plans to launch the mobile application in Canada on Monday, Mr. Krawczyk said.
Despite users in Britain getting first dibs on the chatbot in March last year, Mr. Krawczyk confirmed that Britain, along with a handful of countries, would be excluded from the approaching global launch of the Gemini app.
“When we expand globally, we work with policymakers in these regions and we want to make sure that we continue that commitment,” Mr. Krawczyk said, adding that “we certainly hope to be able to expand to the UK and more countries soon.”
In a December blog post, the company said Gemini Ultra (Google’s largest and most capable AI model) beat GPT-4, the AI model that powers ChatGPT’s paid service, on 30 out of 32 large language model benchmarks.
Free versions of the newly launched Gemini are powered by Gemini Pro, but Google launched Gemini Advanced, a paid model which gives users access to Ultra 1.0, its largest and most capable AI model for $26.99 a month through a new Google One AI Premium plan.
In blind evaluations, Mr. Krawczyk said third-party tests showed Gemini Advanced to be most preferred paid chatbot among its competitors.
Despite that, visits to OpenAI’s ChatGPT far outweigh visits to Google’s chatbot, with an average of 1.6-billion monthly visits to ChatGPT versus 313.2-million average monthly visits to Bard between November, 2023 to January, according to Similarweb, an analytics site that tracks website traffic.
Alphabet and Microsoft both companies reported healthy increases in earnings for the December quarter, but costs surged as the companies both placed heavy investments in severs, data centres and research.
As for how the company plans to make up for its investments in AI, Mr. Krawczyk pointed to the subscriptions Google offers for premium versions of its AI like Gemini Advanced. Alphabet CEO Sundar Pichai has said that Google’s subscription services generated combined revenue of $15 billion last year.
