This week, we look at OpenAI challenging Google with the launch of its ChatGPT Atlas browser, as well as recapping Meta’s recent moves in the advertising space. On today’s MadTech Daily, we cover Amazon introducing a service built for real-time bidding, Reddit targeting Perplexity with a data-related lawsuit, and Alibaba’s Quark app launching an AI chatbot. 

It has been a momentous week for OpenAI, who launched ChatGPT Atlas, a web browser to directly compete with Google Chrome. The AI-powered browser builds on ChatGPT’s capabilities, offering users a more seamless way to access information and complete tasks. The browser also offers users the option to browse in Agent mode. 

As well as facing a significant public challenge from OpenAI, Google recently decided to fully retire its Privacy Sandbox initiative. The tech titan cited limited industry uptake of the technologies as the reason for the discontinuation. 

Elsewhere in the kingdom of Big Tech, Meta announced the tightening of AI chatbot controls for children, aiming to address recent scandals regarding reports of its chatbots engaging in inappropriate conversations with underage users. New controls for parents will be introduced starting in 2026. Parents will be able to block their children from engaging with AI characters on Instagram, Facebook, and the Meta AI app, among other features.  

The move closely follows Meta losing its brand safety accreditation from the Media Rating Council, mere months after it had been granted. The accreditation had signalled that Meta’s platforms met rigorous standards for brand safe ad placements. 

Meta also confirmed it is testing a new ad format: unskippable ads on Instagram Reels. The test aims to assess whether unskippable ads will enhance business discovery.