Source: Google Bard transitions to PaLM 2 and expands to 180 countries
Summary
Google’s chatbot AI, Bard, is dropping the waitlist requirement and expanding to 180 countries. The newer version of Bard uses Google’s Large Language Model, PaLM 2, making it capable of generating better quality and higher fidelity responses as well as generating programming code. Google also plans to add new features such as visual responses, easier image input, and integration with Adobe’s Firefly generative AI.
Fact
💬 Bard chatbot now covers over 180 countries.
🧠 Bard uses Google’s PaLM 2, providing better responses and generating programming code.
👀 New visual responses to your queries and easier image input integration.
🤝 Partnering with Adobe to integrate BingChat-DallE2’s Firefly generative AI.
🛡️ New updates such as the ability to click on a line of generated code and export ability for generated programming code on other platforms.
Source: Google is incorporating Adobe’s Firefly AI image generator into Bard
Summary
Google and Adobe jointly announced that chatbot Bard will soon integrate Adobe Firefly and Express to allow users to generate, edit, and share AI images directly from the command line. The partnership also includes leveraging Adobe’s Content Authenticity Initiative to mitigate threats to creators from generative AI.
Facts
Source: Google’s Duet AI brings more generative features to Workspace apps
Summary
Google has unveiled Duet AI, its collection of AI tools integrated across Workspace applications such as Google Sheets, Google Docs and Google Slides. It includes a growing list of generative AI features, including “Help me write” for Gmail to draft email replies and contextual assistance that creates professional email responses. Duet AI also offers image generation for Google Slides presentations and the ability to generate custom background images for video calls on Google Meet. It also helps users create custom plans and enhance writing by proofreading suggestions.
Fact
Source: Google adds more context and AI-generated photos to image search
Summary
Google will soon allow users to access more information about an image through its image search function, including dates for when the image was first indexed and other places it has been posted online. This will help users determine if images have been generated using AI.
Fact
Source: Google’s Search Labs lets you test its AI-powered ‘products and ideas’
Summary
Google has launched Search Labs which offer users AI-powered search tools such as Search Generative Experience that helps users find and understand information, Code Tips providing programming snippets and “Add to Sheets”, enabling users to insert search results straight into a spreadsheet.
Fact
Source: Google Search’s new Perspectives tab will highlight forum and social media posts
Summary
Google announced a new feature called ‘Perspectives’ that will show users exclusively short- and long-form videos, images, and written posts shared on social media platforms and discussion boards. This is part of Google’s quest to “transform” search through AI. Perspectives will have a stand-alone module that will showcase the same results graphically resembling the Top Stories module.
Facts
Source: Google unveils its multilingual, code-generating PaLM 2 language model
Summary
Google has unveiled its new product PaLM 2, an advanced language model which can run on various devices. PaLM 2 comes in four sizes and its predecessor, PaLM, was the first large language model to perform an expert level on the US medical licensing exam. PaLM 2 has improved multilingual, reasoning, and coding capabilities, and is “smaller, faster, and more efficient.”
Fact