Google’s stake in AI startup Anthropic revealed as 14 percent

Google owns 14 percent of AI startup Anthropic, according to legal filings obtained by The New York Times. As reported by Cade Metz, Nico Grant, and David McCabe, Google’s investment comes with significant limitations: the tech giant cannot own more than 15 percent of the company and holds no voting rights or board seats. The …

Read more

AI-generated content overloads social media through brute force

Jason Koebler, co-founder of 404 Media, reports that generative AI is being used as a “brute force attack” on social media algorithms, flooding platforms with low-quality content at unprecedented scale. In his article, Koebler explains that AI creators can produce dozens of posts in minutes, allowing them to quickly identify and exploit what performs well …

Read more

Google’s Gemini model used to remove image watermarks

A recent discovery shows that Google’s Gemini 2.0 Flash AI model can remove watermarks from images, including those from Getty Images and other stock photo providers. According to reporting by Kyle Wiggers for TechCrunch, users on social media platforms have been sharing examples of this controversial capability. Unlike some competing AI models such as Anthropic’s …

Read more

Amazon to end local processing of Alexa requests

Amazon will eliminate a privacy feature that allowed Echo users to process Alexa requests locally without sending voice recordings to the cloud. According to Scharon Harding at Ars Technica, this change takes effect March 28 and coincides with the rollout of Alexa+, Amazon’s new subscription-based assistant. All voice commands to Echo devices will now be …

Read more

Mistral AI launches small but powerful open-source multimodal model

Mistral AI has released Mistral Small 3.1, a new open-source AI model with only 24 billion parameters that reportedly outperforms larger models from competitors like OpenAI. According to Michael Nuñez’s article, the French startup claims their model matches or exceeds the capabilities of GPT-4o Mini while offering text and image processing, multilingual support, and a …

Read more

Moonvalley launches AI video model trained exclusively on licensed data

Moonvalley has introduced Marey, a new AI video model trained solely on legally acquired and licensed content. According to Carl Franzen of VentureBeat, this approach addresses copyright concerns plaguing many competing AI video models that use scraped web content. Moonvalley CEO Naeem Talukdar stated, “People said it wasn’t technically feasible to build a cutting-edge AI …

Read more

Google to replace Assistant with Gemini on mobile devices by late 2025

Google has announced plans to replace its Google Assistant with Gemini on most mobile devices by the end of 2025. The transition, which has already begun for millions of users, represents what Google calls a “platform shift” driven by generative AI technology. What is changing Google Assistant, launched in 2016 with natural language processing and …

Read more

Snapchat introduces AI-powered video lenses for premium users

Snapchat has launched its first AI Video Lenses powered by its in-house generative video model. According to Aisha Malik at TechCrunch, these new features are exclusively available to Snapchat+ subscribers who pay $15.99 monthly. The initial release includes three lenses: “Raccoon,” “Fox,” and “Spring Flowers,” with more planned weekly. Users can access these lenses through …

Read more

Tech giants push for relaxed AI regulation under Trump administration

OpenAI and Google have both submitted policy proposals to the Trump administration’s upcoming “AI Action Plan,” advocating for minimal government intervention and looser copyright restrictions for AI development. These submissions follow President Trump’s revocation of Biden’s AI executive order, which had focused on safety and trustworthiness. OpenAI’s proposal emphasizes “the freedom to innovate in the …

Read more

Google introduces personalized Gemini responses based on search history

Google has launched a new experimental feature for its AI assistant Gemini that personalizes responses based on users’ search history and other Google services. According to Dave Citron, Senior Director of Product Management for the Gemini app, this capability aims to make Gemini “a personal AI assistant” that understands individual users rather than just answering …

Read more