Every LLM release, update, and milestone.
Microsoft is developing a new authentication and verification system designed to help users distinguish authentic content from AI-generated material circulating online. The initiative addresses the growing problem of AI-enabled deception now prevalent across social media and digital platforms.
Google DeepMind is pushing for moral behavior in large language models to be evaluated with the same technical rigor applied to coding and math benchmarks. As LLMs take on roles like companions, therapists, and medical advisors, the research group argues current evaluation standards are insufficient.
Google Deepmind has integrated its Lyria 3 model into Gemini, enabling users to generate 30-second music tracks with vocals, lyrics, and cover art from text prompts or uploaded media. The model represents an expansion of Google's multimodal AI capabilities into creative audio generation.
World Labs, the AI startup founded by Fei-Fei Li, has raised $1 billion in new funding to develop spatial intelligence—AI systems capable of understanding and operating in three-dimensional physical environments. The capital will fund the development of world models, a class of AI architecture designed to reason about spatial relationships and physical interactions.
Google announced Gemini 3.1 Pro, positioning the model for complex problem-solving tasks requiring deeper reasoning than previous versions. The release follows Gemini 3 Pro (November 2025) and Gemini 3 Flash (December 2025).
Google is rolling out Lyria 3, its music generation model, to all Gemini app users. The expansion follows recent releases of audio overviews, image generation, and video capabilities in the Gemini ecosystem.
American International Group (AIG) has deployed an agentic AI system with an orchestration layer, reporting faster-than-expected productivity gains in underwriting and portfolio management. The deployment demonstrates measurable improvements in throughput and workflow efficiency, according to recent investor disclosures.
Alibaba has released the Qwen 3.5 series, an open-source model that claims performance comparable to frontier proprietary models while running on commodity hardware. The release signals a shift in AI model economics, offering enterprises lower inference costs and greater deployment flexibility than closed alternatives.
Goldman Sachs is deploying Anthropic's Claude model in trade accounting and client onboarding operations. The deployment represents a broader adoption of generative AI among large financial institutions to improve operational efficiency in back-office processes.
Google has integrated Lyria 3, its music generation model, directly into the Gemini app. Users can now create custom 30-second music tracks from text descriptions and images without additional tools or subscriptions.
A new pruning method challenges the conventional wisdom inherited from autoregressive LLMs about preserving attention-sink tokens. Researchers demonstrate that attention sinks in diffusion language models are substantially less stable than in AR models, enabling more aggressive pruning without retraining.
Researchers introduce UniLID, a language identification method that leverages tokenizer-based unigram distributions to identify languages and dialects with high sample efficiency. The approach achieves over 70% accuracy on low-resource languages with only five labeled examples per language, substantially outperforming existing systems like fastText, GlotLID, and CLD3 in low-resource settings.
A new research paper introduces methods for predicting when humans will intervene in autonomous web agents by analyzing distinct interaction patterns. The work, which includes a dataset of 400 real-user web navigation trajectories with over 4,200 interleaved human-agent actions, shows that intervention-aware models improved agent usefulness by 26.5% in user studies.
Researchers have identified a fundamental difference in how attention mechanisms work in diffusion language models versus traditional autoregressive LLMs, enabling a new pruning strategy that removes unstable attention sinks without retraining. The finding challenges existing pruning assumptions inherited from autoregressive models and promises better quality-efficiency trade-offs during inference.
A new technique called Variance Controlled Policy Optimization (VCPO) addresses a fundamental problem in asynchronous reinforcement learning for LLMs: high variance in policy-gradient estimates from stale rollouts. The method scales learning rates based on effective sample size and applies a minimum-variance baseline, reducing long-context training time by 2.5x while maintaining synchronous performance.
Segmind has released SegMoE, a mixture-of-experts (MoE) diffusion model designed to accelerate image generation while reducing computational overhead. The model applies MoE techniques traditionally used in large language models to the diffusion model architecture, enabling selective expert activation during inference.
OpenAI and Italian media company GEDI announced a strategic partnership to integrate Italian-language news content into ChatGPT. The deal expands OpenAI's content partnerships beyond English-speaking markets, following similar agreements with news organizations in other regions.