Browsing Tag
Google DeepMind
12 posts
Google DeepMind models, research, AI products, leadership, and policy coverage.
Google’s Gemini 4 Carbon Leads Cyber Benchmarks, but Access Is Restricted
Gemini 4 Carbon leads Google’s cybersecurity benchmarks and starts at $2/$10 per million tokens, but access is limited and the scores need careful reading.
Google’s Genie 4 Adds Native Audio and a Gemini Control Loop
Google DeepMind’s Genie 4 can steer interactive worlds with live prompts, native audio and a Gemini planning loop. Here is how it works, who gets access and where it still fails.
UK’s AI Biosecurity Lab Gets £100M Cash and £200M in AI Credits
The UK is launching an independent AI Biosecurity Lab with £100 million in public funding and up to £200 million in OpenAI and Anthropic credits. Here is what it will do and what remains unresolved.
Google’s SynthID Detector Is Public. Here’s What “Not Detected” Means
Google’s public SynthID Detector checks AI watermarks from several major providers. Here is how to use it and why a negative result proves little.
EmbeddingGemma 2 Brings Private Multimodal Search On-Device
Google's 740M-parameter EmbeddingGemma 2 runs local search across text, code, images, audio and video. Here are the memory, vector-size and deployment tradeoffs.
Gemini 4 Argon Debuts With 1M-Token Output, but Access Starts With Cyber Defenders
Google's Gemini 4 Argon pairs a one-million-token output limit with low introductory pricing, but only vetted cyber defenders can use it first.
Gemini 3.5 Flash Makes Computer Use a Mainstream Agent Tool
Google has moved computer use into Gemini 3.5 Flash, letting developers build agents that can see screens and act across browser, mobile, and desktop environments. The useful question is how teams design the execution loop, safety gates, and sandbox around it.
Google’s AI Talent Losses Put Coding and Science Roadmaps Under Pressure
Google DeepMind lost Noam Shazeer to OpenAI and John Jumper to Anthropic in the same week. The moves matter because AI labs are competing for researchers who can steer coding agents, scientific AI, and frontier model strategy, not just write papers.
Google DeepMind’s A24 Deal Puts AI Inside the Filmmaking Workflow
Google DeepMind and A24 are launching a multi-project AI research partnership, with Google reportedly investing about $75 million in the studio. The deal is less about one AI movie tool than a test of whether creative AI can be shaped inside real film workflows without handing a tech company the studio library.
DeepMind’s AI Control Roadmap Makes Agent Security a Runtime Problem
Google DeepMind’s AI Control Roadmap treats powerful internal AI agents as systems that need monitoring, access limits, response plans, and shutdown paths. The framework is a signal for enterprises moving from chatbots to tool-using agents: alignment claims are no longer enough if the agent can touch code, data, infrastructure, or security workflows.