Google brings Gemini AI to millions of vehicles
Google is rolling out its Gemini AI assistant to vehicles with Google built-in, starting with approximately 4 million General Motors cars in the U.S.
Sections
AI & Models
529 storiesEarth AI aims to cut mineral sample processing from months to days
Earth AI is building its own labs, hoping to reduce mineral sample processing times from five months to five days to improve its AI exploration models.
Google grants U.S. military access to AI for classified networks
Google has granted the U.S. Department of Defense access to its AI for classified networks, following similar deals by OpenAI and xAI.
Anthropic tests AI-to-AI commerce in internal pilot
Anthropic’s Project Deal pilot found that while advanced AI agents secure objectively better outcomes in commerce, users may not perceive the performance disparity.
DeepSeek launches V4 models with 1.6 trillion parameters
DeepSeek launched its V4 Flash and Pro models, which feature up to 1.6 trillion parameters and claim to almost close the performance gap with frontier AI models.
OpenAI launches GPT-5.5 as it pushes toward 'super app' vision
OpenAI released GPT-5.5, its smartest model yet, as the company advances its strategy to build an AI super app integrating services like ChatGPT and Codex.
Space telescopes are creating a massive demand for GPU compute
As NASA prepares to launch the Nancy Grace Roman space telescope, astrophysicists are increasingly relying on GPU-accelerated AI models to manage an expected deluge of space data.
Google targets enterprise IT with new agent-building tool
Google launched the Gemini Enterprise Agent Platform, a new tool for IT teams to build and manage AI agents, positioning it against Amazon and Microsoft offerings.
OpenAI partners with Infosys to scale enterprise AI tools
OpenAI is partnering with Infosys to integrate its AI tools, including Codex, into the Indian IT giant’s Topaz AI platform to help clients modernize software development.
Google adds generative AI to its mapping and geospatial tools
Google is rolling out new generative AI features for its mapping and geospatial platforms, aiming to accelerate data analysis and visualization for enterprise users.
Meta to track employee keystrokes for AI training
Meta plans to use internal employee mouse movements and keystrokes to train its AI models, citing a need for real-world examples of how people use computers.
Anthropic investigates unauthorized access to Mythos tool
Anthropic is investigating reports that unauthorized users gained access to its Mythos cybersecurity AI tool through a third-party vendor, though the company claims no systems were impacted.
OpenAI launches ChatGPT Images 2.0 with improved text rendering
OpenAI released ChatGPT Images 2.0, a new model featuring improved text rendering, up to 2K resolution, and "thinking capabilities" for complex image generation tasks.
Sam Altman criticizes Anthropic’s fear-based marketing
Sam Altman criticized competitor Anthropic for using fear-based marketing to promote its new cybersecurity model, Mythos, suggesting the rhetoric serves to restrict public access to AI.
Why corporate press releases are sounding more like AI models
Corporate communications are increasingly using the 'It’s not just X, it’s Y' sentence structure, with usage rising from about 50 mentions in 2023 to over 200 in 2025.
Anthropic’s Mythos model is reportedly in use by the NSA
The NSA is reportedly using Anthropic’s restricted Mythos model, even as the Department of Defense labels the company a supply-chain risk.
OpenAI loses key researchers as it pivots to enterprise AI
OpenAI is losing key researchers Kevin Weil and Bill Peebles as the company consolidates its focus on enterprise AI and its forthcoming superapp, shedding non-core side quests.
Luma partners with Wonder Project for AI-powered filmmaking
Luma has launched Innovative Dreams, a production company partnering with Wonder Project to demonstrate the efficiency of generative AI in professional filmmaking.
Physical Intelligence unveils π0.7 robot brain model
Physical Intelligence’s new π0.7 model shows early signs of compositional generalization, allowing robots to perform tasks they were never explicitly trained to execute.
OpenAI adds background agent capabilities to its Codex coding tool
OpenAI has updated its Codex coding tool with background agent capabilities and 111 plug-in integrations, intensifying its competition with Anthropic’s Claude Code.