Huge change to Google Australia announcedThe way everyone uses Google is going to change – and the benefits Google claims it will have are remarkable.Alice Clarke3 min readFebruary 9, 2024 - 8:21AMInternetDon't miss out on the headlines from Internet. Followed categories will be added to My News.Google is about to enter its “Gemini era”.GoogleCEO Sundar Pichai, announced on Friday that Google is now going all-in on its generative AI platform, which will be renamed from ‘Bard’ to ‘Gemini’.The biggest change for Australians, is that Google Assistant is now being phased out in favour of Gemini.As of next week, AustralianGoogle Assistantusers on Android phones will be able to opt-in to use Gemini instead of the standard Google Assistant, marking the biggest change to Google Assistant since it was first introduced in 2016. Apple iPhone users will get access to Gemini in the next few weeks.Google CEO Sundar Pichai announced that Google Assistant will be phased out and replaced by AI platform, Gemini. Picture: Josh Edelson / AFPGoogle Assistant is the Google equivalent of Apple’s Siri or Amazon Alexa, and currently answers simple questions and does basic tasks, like turning smart lights on or off. Gemini will build on that byincorporating generative AI to fulfil tasks.Users can use their voice (or type) to ask Gemini to generate a custom image for a birthday invitation, create an Instagram caption for a photo, or answer questions about an article you’re reading.Google claims you can even take a photo of your flat tyre on the side of the road and ask what to do next. Unlike Google Assistant, Gemini will be able to access more context for questions, which is how it knows which photo you want a caption for, or which article you’re reading.Gemini is coming to Australia. Picture: SuppliedIf users are uncomfortable with generative AI, or with having an AI assistant scanning their screen for context, then they can stick with the basic Google Assistant for now. But it raises the bar for what Apple and Amazon will need to do with their smart assistants to compete during this current generative AI trend.For users ofBard, not much will change except for the web address switching to say Gemini. Bard is a generative AI platform like ChatGPT or Midjourney, and uses short prompts to create pieces of writing or images.Google users in Australia will be able to opt into using Gemini from next week. Picture: SuppliedGoogle is the biggest search engine in the world, and Google Assistant is one of the most popular smart assistants in Australia, so these announcements signal further integration of generative AI in our daily lives, and further mainstream the technology.But generative AI is not without controversy.One common complaint about generative AI is that the Language Learning Models (LLMs) on which they’re built are very good at approximating what a human sentence might look like, but aren’t very capable of fact checking.For example, if you Google “which countries in Africa start with a k?”, the top AI selected (and allegedly written) answer says: “As of my last knowledge update in September 2021, there is no sovereign country in Africa whose name begins with the letter ‘K’.“However, please note that situations can change over time, so it’s always a good idea to verify this information with more recent sources if needed.”This almost an improvement on the previous top selected answer of “There are no countries in Africa that start with the letter ‘K’. The closest is Kenya, which starts with a ‘K’ sound.”Both answers are almost as funny as they are wrong.Fact checking isn’t AI’s strong point. Picture: SuppliedDo not say this to a rape survivor. Picture: SuppliedA significantly less funny, example of AI going wrong is if you google: “What to say to a rape victim”. One of the top AI suggested answers is, “it was your fault”.Another complaint is that the models these AIs are trained on include thousands, if not millions of pieces of copyrighted content used without permission, with artists and media organisations concerned that these models trained on their work will put them out of business. As reported by Euronews, OpenAI (the company behind ChatGPT) recently said: “Because copyright today covers virtually every sort of human expression – including blog posts, photographs, forum posts, scraps of software code, and government documents – it would be impossible to train today’s leading AI models without using copyrighted materials.”It will be interesting to see how that defence holds up in court.The new Gemini experience available to Australians on Android from next week.More CoverageGoogle sued after dad’s horror Maps deathIsabel Keane – New York PostMeta wants industry-wide labels for AI-made imagesOriginally published asHuge change to Google Australia announcedMore related storiesSocial‘Manfluencer’s’ far-right threat revealedControversial “manfluencer” Andrew Tate is facing rape and sex trafficking charges and Aussie experts warn his rhetoric also overlaps with far-right radicalism.Read moreInternet‘Time they paid’: Call to tax tech titansTech giants such as Meta, Amazon, Uber and Microsoft could be hit with a hefty tax under a new election policy demand put forward by the Greens.Read more