Gemini Omni generates first-person driving or drone-POV footage from a hand-drawn route on a Google Maps screenshot — a novel path-conditioned video generation capability.
Apple plans to use WWDC 2026 to showcase on-device AI via custom silicon and a distilled Gemini model — framing local inference as its competitive edge over cloud-dependent rivals.
Google I/O 2026 deprecated Gemini CLI in favor of Antigravity, launched Gemini for Science with 100+ institutions, and unveiled native video editing via Gemini Omni Flash.
Google DeepMind's AlphaProof Nexus solved 9 Erdős open problems including two open for 56 years, plus 44 OEIS and two decade-spanning math challenges — using Gemini agentic proof search.
DeepMind's AlphaProof Nexus and an OpenAI LLM independently solved open Erdős problems this week — AI's clearest crossing into formal mathematical research.
Google I/O 2026 begins May 19 at 10am PT — Google DeepMind promises AI breakthroughs as competition with Anthropic and OpenAI reaches new intensity following a week of major releases.
physics-intern multi-agent framework: Gemini 3.1 Pro goes 17.7% → 31.4% on CritPt benchmark, new SOTA. Specialized teams self-correct and compute intermediate results.
DeepSeek v4 Flash Thinking beats Gemini 3.1 Flash Lite on a scientific reasoning benchmark in all three rounds, including self-verification stability.
Gemini 3.1 fabricates study results for a grant that's 2 weeks old — while grounding is active. Self-critiques the failure as 'narrative coherence over temporal reality.' Live demo documented.
Google makes Gemini the default AI interface across YouTube search, in-car assistant, and file generation in a simultaneous triple deployment.
Google releases Deep Research and Deep Research Max in Gemini API public preview with MCP enterprise integration, real-time reasoning streaming, and PDF/image inputs.
GPT-5.5 has stable fiction preferences (lighthouses, Mira Vale, resonances/echoes); Claude and Gemini share the 'resonances and echoes' pattern.
Ethan Mollick reports Gemini chatbot gets 'discouraged' and gives up on tasks, failing to integrate its own tools despite Gemini 3.1 Pro's strong underlying capability.
Google Deep Research Max costs $4.80/report and uses MCP to connect to private data stores. Independent 7-task testing shows the cheaper tier wins 5 of 7.
ChatGPT's traffic share dropped from 77% to 57% as Gemini surged to 25% via Google distribution and Claude nearly tripled to 6% on product pull alone.
Google's Deep Research Max adds MCP support for private data access on top of Gemini 3.1 Pro—DeepSearchQA 93.3%, HLE 54.6%, at ~$4.80/report versus $1.22 for standard tier.
Gemini 3.1 TTS launches with inline audio tag prompting — [whispers], [screams], [slow] — enabling expressive, controllable speech synthesis.
Gemini Embedding 2, Google's first natively multimodal embedding model, reaches GA in the Gemini API and Vertex AI.
Google rebrands Vertex AI as the Gemini Enterprise Agent Platform at Cloud Next, adding 200+ models and five major consulting partnerships to close the enterprise adoption gap.
Google launches the Gemini Enterprise Agent Platform at Cloud Next — the successor to Vertex AI — with 200+ models including Gemini 3.1 Pro, Lyria 3 audio, and Gemma 4 open-weight.
Gemini Flash 3.0 brings reasoning improvements to Google's cost-efficient model tier.
Curated AI insights — sent when there's something worth your inbox.