Google Bard unveils Experiment Updates: What’s new in conversational AI?

Stay up-to-date with Google Bard’s latest features, improvements, and bug fixes through the new ‘Experiment Updates’ changelog, as the conversational AI platform integrates PaLM and offers more ‘Google It’ search topics.


Google Bard Gets ‘Experiment Updates’ Changelog

Google Bard now has an “Experiment updates” changelog, similar to GPT’s versioning system, which informs users of the latest features, improvements, and bug fixes. The team aims to provide a convenient place for users to stay informed and provide feedback. Bard now offers more “Google It” search topics and has integrated PaLM to improve math and logic responses, with coding capabilities on the way. The changelog can be accessed directly or through the side navigation drawer. It remains to be seen how Google will mark major updates for both Bard and LaMDA in the future.


AI Artist Imagines How The World’s Wealthiest Would Look If They Were Poor

Artists are using artificial intelligence to generate images with minimal effort. Gokul Pillai recently used Midjourney, an AI program, to create images of billionaires dressed in rags, standing against the backdrop of slum areas. The images feature Donald Trump, Bill Gates, Mukesh Ambani, Mark Zuckerberg, Warren Buffett, Jeff Bezos, and Elon Musk. Pillai captioned the post as “Slumdog Millionaires. (Did I miss to include anyone in the list?)”. The images have since gone viral on social media, with users praising the artist’s concept and the realism of the images.

Check out the Slumdog Millionaires here:


Google Teases Taking Pixel Call Screen ‘Even Further’ This Year With AI

Google plans to use conversational AI to solve the issue of unwanted calls, from spam to robocalling, according to the group product manager for the Phone by Google app, Jonathan Eccles. Speaking on the Made by Google Podcast, he said the company wants to create a future where users never feel unproductive or bothered by incoming calls, and teased that this year there will be new features added to solve for unwanted calls using AI. There is speculation that Google will apply its LaMDA model to Call Screen to enhance the feature’s ability to find out what a caller wants.


Microsoft Could Be Planning Bing Chat Third-Party Plugin Support

Microsoft may be planning to expand the functionality of its Bing Chat AI through third-party plugin support, according to Mikhail Parakhin, the CEO of Bing. Parakhin confirmed on Twitter that the expansion is currently under development, and while he did not provide details on when or how it will be available, it could allow third-party developers to integrate Bing Chat AI within their apps. Bing Chat, which is powered by OpenAI’s GPT-4 model, is currently in preview and has gone through many improvements to provide users with personalized and relevant search results.


Meta Will Debut Its Generative AI This Year

Meta, formerly Facebook, plans to launch its proprietary generative AI technology by the end of this year. The company has been investing in AI research since 2013 and recently created a new generative AI team to focus on commercializing the technology for improving ad effectiveness and applying it across its products, including Facebook and Instagram. The company also plans to incorporate the technology into the development of the metaverse to make content creation more accessible. Although generative AI holds great potential, concerns remain about its impact on human control over civilization. Meta’s CTO Andrew Bosworth believes that stopping progress is challenging, and it’s crucial to invest in responsible development. Meta’s focus on generative AI and its integration with their products and the metaverse demonstrate their commitment to being at the forefront of AI advancements.


If you’re tired of ChatGPT’s slow and unreliable free version, but don’t want to pay for the $20 per month ChatGPT Plus, OpenAI’s pay-as-you-go plan may be the solution for you. With the latest ChatGPT 3.5 Turbo API models, you can pay as little as $0.002 per 1,000 tokens. All you need is a unique set of API keys from OpenAI that you can use in any third-party ChatGPT app. Head over to OpenAI’s billing page, sign up for the PAY AS YOU GO plan, set a soft or hard limit of your choice, copy your API key, and you’re good to go. The world is your oyster!


Poe’s AI Chatbot App Now Lets You Make Your Own Bots Using Prompts

Quora‘s latest product, Poe, now allows users to make their chatbots with prompts that direct them to perform specific tasks. These bots use an existing bot, such as ChatGPT, as their base. Poe initially launched with a few general knowledge bots, which users can access by paying for subscriptions to more powerful language models, such as GPT-4. Users can access these bots via Poe’s iOS or Android app, desktop web interface, or mobile web. The platform plans to offer bot creators feedback on how their bots are used and eventually develop an API to allow anyone to host a bot.


Apple Allegedly Allowed Fake Baidu AI Apps In The App Store

Baidu, a Chinese tech company, is suing Apple and app developers, alleging they created counterfeit apps of Baidu’s Ernie AI chatbot and that Apple allowed them on the App Store. Baidu claims that any Ernie apps on the App Store are fake until an official announcement is made. Ernie is a chatbot similar to OpenAI’s ChatGPT and is only available to a select number of users. Baidu hopes to remove similar apps from the App Store and halt their creation in the future. The company still has not announced when Ernie will be available to the general public.