Google’s AI chatbot, Bard, was launched for Australian users on Thursday. It is a significant milestone in the company’s advancements in artificial intelligence. Previously available only in the US and the UK, Bard will now be accessible to users in over 180 countries, including Australia.
Bard is a chat program built on Google’s large language model, PaLM2, similar to OpenAI’s GPT. It offers various functionalities such as providing information, coding, language translation, and image analysis.
During Google’s annual I/O conference, the company announced its plans to enhance Bard by incorporating visual responses alongside text-based ones. Users will soon be able to upload images to Bard through Google’s Lens application. For instance, users can upload a photo of two dogs and ask Bard to generate a funny caption. Bard will not only identify the breed of dogs but also draft appropriate responses.
To address the issue of AI hallucinations, where AI-generated content is presented as factual information, Google will annotate any externally sourced information provided by Bard and include a link to the source.
Google is committed to expanding Bard’s availability in more than 40 languages. Though the process is gradual due to concerns about potential harmful language output from PaLM2-based systems.
Furthermore, Google has plans to integrate Bard into various products such as Gmail, Docs, Drive, Maps, and others. This integration, known as Duet AI, will allow users to receive assistance with writing and other tasks within these applications. Google emphasizes that users will retain control over their privacy and the integration of these tools.
In line with Google’s commitment to ethical AI development, the company’s CEO, Sundar Pichai, stated that all of Google’s AI models will incorporate watermarking and metadata, ensuring that AI-generated content is clearly identified.
Looking ahead, Google intends to enhance its search capabilities by incorporating AI-generated text results alongside links. Additionally, Google plans to introduce an “about this image” tool in search results, providing context about the origin and presence of similar images online.
Google has also introduced a new experimental AI video subbing service called the universal-translator, which accurately translates a speaker’s voice while matching their lip movements. However, Google acknowledges the potential misuse of this technology by malicious actors despite its significant potential.
With Bard’s expansion and Google’s ongoing efforts to develop AI ethically, the company aims to bring the benefits of artificial intelligence to a wider audience while addressing potential risks and privacy concerns.
