Week 22

*|MC_PREVIEW_TEXT|*

View this email in your browser Hi Humans, Welcome to the first AI.HUB newsletter! The purpose of this weekly newsletter is to keep you updated on what is happening in the world of AI, provide you with links to exciting articles, and hopefully take the edge of your AI urge. With no further ado, let us get started. AI news from the past week: Google presented its newest AI search algorithms, which didn’t go well. Employees leaving OpenAI because of safety concerns. Elon Musk wants to build a 10,000-semiconductor supercomputer to train xAI’s upcoming Grok model. Apple signs a deal with OpenAI for iOS. Google’s AI recommends smoking doing pregnancy Due to the speed of the AI world, a new feature from 5 days ago is old news. However, we want to use a few minutes to talk about Google’s ‘AI Overview’ — AI-powered answers above the search result — released in the US on May 14. Since its release, AI Overview has been a laughingstock. Filling the internet with  examples of the AI recommending 2-3 cigarettes per day during pregnancy or adding glue to pizza sauce to ensure your cheese is sticking to your pizza. Disclaimer: Doctors don’t recommend smoking, and glue in the pizza sauce is not a healthy idea either! Why It Matters This isn't the first time Google has struggled with an AI tool launch. When Gemini (the old Bard) was introduced, it was full of errors. Later, controversy arose when Google's image generator produced racially diverse depictions of Nazis and Founding Fathers. However, there's a critical difference between the misinformation on Gemini, which includes a disclaimer that states, “Gemini may display inaccurate info, including about people, so double-check its responses,” and the misinformation on Google Search, a trusted service used for billions of search queries daily (and lacking any such a disclaimer). I expect Google to fix this issue, and I believe we have to be patient before the ‘AI Overview’ is released to the Danish market. In the meantime, we can always use Perplexity or patiently wait for the arrival of the OpenAI Search feature. ‘OpenAI is shouldering an enormous responsibility on behalf of all humanity’ This statement is made by Jan Leike, who left his position as head of super alignment at OpenAI on Friday, May 17th. Jan Leike announced his resignation on the social media platform X, where he explains how OpenAI is no longer concerned about safety or security, but is more focused on shipping shiny products. Why It Matters This is a no-brainer. With more than 180.5 million people using and sharing data with ChatGPT, data security and safety should always be at the top of the agenda to ensure that the risk of privacy violation and the potential for misuses remain as low as possible. I think we, as users and humans, need to demand data transparency and easy-to-understand data policies (I know it is easier said than done) and start thinking critically about how we are sharing our data. Otherwise, we will all end up in a 'Joan is Awful' episode — it’s a reference to a Black Mirror episode, and it is absolutely worth watching! Thank you for reading! Everything starts from 0; if you have ideas, inputs, or improvements - please let me know :) Until next time // Annemette Møhl This email was made in collaboration with Lukas Guldstøv. Follow Us Copyright (C) *|CURRENT_YEAR|* *|LIST:COMPANY|*. All rights reserved. *|IFNOT:ARCHIVE_PAGE|**|LIST:DESCRIPTION|**|END:IF|* Our mailing address is: *|IFNOT:ARCHIVE_PAGE|**|HTML:LIST_ADDRESS|**|END:IF|* Want to change how you receive these emails? You can update your preferences or unsubscribe