AI News Roundup: Key Highlights This Week

Key takeaways

Osmo is working on scent replication tech for health and wellness, with future plans for disease detection.

AMD is acquiring ZT Systems for $4.9 billion to enhance its AI capabilities.

Microsoft warns about AI deepfakes at the DNC, highlighting growing concerns about misinformation.

This week in the tech world has been eventful, with exciting developments in AI, significant acquisitions, and new tools being introduced. Here’s a breakdown of the key happenings.

Osmo's Scent Technology for Wellness and Health 

AI startup Osmo, founded by ex-Google researchers, is making strides in scent recognition technology. Currently, the focus is on wellness and consumer products, but the ultimate goal is to use this technology for detecting diseases in the future.

AMD's $4.9 Billion Acquisition of ZT Systems 

AMD has struck a $4.9 billion deal to acquire server solutions company ZT Systems. The acquisition, primarily paid in cash with a portion in stock, aims to bolster AMD's AI capabilities, putting them in a stronger position to compete with Nvidia. While ZT’s design and customer teams will stay, its manufacturing division will be sold off.

Microsoft Raises AI Deepfake Concerns at the DNC 

At the Democratic National Convention in Chicago, Microsoft hosted a luncheon on August 19 to discuss the growing threat of AI-generated deepfakes. With incidents like deepfakes in Chicago’s mayoral race and Slovakia’s parliamentary elections, there’s an increasing need to address AI-driven misinformation in global politics.

Trump Shares Fake AI Images of Taylor Swift 

Donald Trump shared AI-generated images on social media that falsely suggested Taylor Swift’s endorsement, sparking backlash from her fans. Swift has not endorsed any candidate for the upcoming election, although she did support Democrats in 2020.

FCC Fines Lingo Telecom $1M for AI Robocall Scam 

The FCC has fined Lingo Telecom $1 million for using AI to generate robocalls that mimicked President Biden’s voice, misleading voters in New Hampshire. The FCC has also introduced new verification rules to counter AI-related election manipulation.

Authors Sue Anthropic Over Copyright Violations 

A group of authors has filed a lawsuit against AI company Anthropic, accusing them of using pirated books to train their AI chatbot, Claude. This lawsuit, filed in San Francisco, challenges Anthropic’s claims of developing AI responsibly. 

Texas University’s AI Predicts Earthquakes with 70% Accuracy 

Researchers at the University of Texas have developed an AI algorithm that successfully predicted 70% of earthquakes during a seven-month trial in China. The team plans to continue testing in Texas, with hopes of improving predictions in earthquake-prone areas worldwide.

OpenAI Partners with Condé Nast for Content Integration 

OpenAI has teamed up with Condé Nast to incorporate the publisher’s content into its new AI-powered search engine, SearchGPT. This partnership aims to enhance the accuracy and quality of search results, continuing a trend of OpenAI collaborating with major media companies.

Windows 11’s AI-Based Recall Feature Delayed Until October 

Microsoft has postponed the release of the AI-based Recall feature in Windows 11 until October. Initially set to launch with Copilot+ PCs in June, the feature is undergoing additional security improvements. A new release date for the general public has not yet been announced.

Gmail Introduces 'Polish' Feature to Refine Email Drafts 

Gmail's new “Polish” feature in the “Help Me Write” tool helps refine email drafts, turning rough notes into polished messages. Available through shortcuts on Android and iOS, this feature is currently being rolled out to selected Google Workspace users.

McAfee Launches AI Tool to Detect Deepfakes 

McAfee has introduced the AI Deepfake Detector, designed to identify deepfakes, which is available on select Lenovo PCs. The tool runs locally to ensure privacy and is part of McAfee’s Smart AI Hub, aimed at educating users about AI-related scams.

Tech Giants Repurpose Old Power Stations for AI Expansion 

Microsoft and Amazon are converting old power stations into data centers to meet AI’s growing energy demands. This strategy leverages existing infrastructure to address land and energy constraints.

YouTube Unveils AI Tool to Help Creators Recover Hacked Accounts 

YouTube has introduced an AI tool to assist creators in recovering hacked accounts. Initially available in English and to a select group of creators, the tool guides users through troubleshooting steps, allowing account recovery without needing direct support from Google. A wider rollout is planned for the future.

Midjourney’s New Website Simplifies AI Image Creation 

Midjourney has launched a new website that lets users generate up to 25 AI images for free without needing to sign up. New users can choose to sign in with either Google or Discord, with subscriptions starting at $10 per month after the trial period. This update removes the previous requirement to use Discord for image generation.