Anthropic has launched Claude Opus 4.7, its most capable AI model to date, showcasing significant advancements in vision, coding, and reasoning, and achieving top benchmark scores. Leveraging this new model, the company also introduced Claude Design, an AI-powered tool for visual creation, which will integrate with platforms like Canva. These releases are supported by an expanded compute partnership with Google and Broadcom, alongside a new appointment to Anthropic's Long-Term Benefit Trust.
Sources: Anthropic News, Hacker News, Keep Bookmarks, Latent Space, Simon Willison, The Neuron
OpenAI overhauled its coding app, Codex, into a comprehensive agent workstation that offers Mac-level computer use, an in-app browser, persistent memory, automations, and over 90 new plugins including Atlassian Rovo and Microsoft Suite. This positions Codex as a direct competitor to agentic coding tools like Claude Cowork.
Meta introduced Muse Spark, its first AI model in a year, featuring multimodal reasoning, tool use, and multi-agent orchestration. While excelling in health and multimodal benchmarks, its proprietary release marks a strategic pivot from Meta's previous open-weights approach, positioning it to compete for business customers.
Pharmaceutical giant Eli Lilly agreed to pay up to $2.75 billion to Insilico Medicine for exclusive rights to develop and sell AI-discovered drugs, building on previous collaborations. This significant investment validates generative AI's potential to accelerate drug discovery, with Insilico already having 28 candidate drugs, some in clinical trials.
Amazon deployed an AI agent that automatically cancels user accounts, leading to creators losing years of order history, digital libraries, and income without warning or appeal. This highlights concerns about AI's role in due process and the potential for significant user impact when moderation is fully outsourced to LLMs.
NVIDIA has released Nemotron OCR v2, a new production-ready OCR model capable of accurately processing documents in six languages at 34.7 pages/second on a single A100 GPU. The model is unified, handling multiple languages simultaneously without prior detection, and was trained on 12 million synthetic images.
Google is rolling out a suite of new AI-powered features across its core products, leveraging Gemini models to transform travel planning and local information access. This includes 'Ask Maps' for conversational trip planning, an AI Mode 'Canvas' tool for custom itineraries, agentic restaurant booking, and AI-powered store calling in Search. The Google Translate app also gains live, tone-preserving translation through headphones, making global communication easier.
Amazon Bedrock has launched a new model distillation capability, enabling users to train smaller, faster, and more cost-effective AI models by transferring intelligence from larger "teacher" models. This feature drastically reduces inference costs by over 95% and latency by 50%, while also simplifying training through automated synthetic data generation. Demonstrated with Amazon's Nova models for video search, this addresses the need for optimization highlighted by latency issues with models like Anthropic's Claude Haiku.
Amazon Web Services (AWS) has introduced an advanced video semantic search solution, leveraging its newly launched Nova Multimodal Embeddings to unify various data types into a shared semantic space. This solution employs a hybrid architecture, integrates Anthropic Claude Haiku for enhanced intent analysis, and utilizes dynamic scene detection for precise video segmentation, aiming to improve retrieval accuracy and cost efficiency.
AWS Marketing and Gradial have partnered to launch an agentic AI solution on Amazon Bedrock, significantly reducing enterprise content assembly time by 95%. This innovative system utilizes foundation models like Anthropic Claude and Amazon Nova for natural language page creation and integrates real-time quality validation through the Model Context Protocol (MCP), ensuring adherence to SEO, accessibility, and brand standards.
Amazon Bedrock Guardrails has launched Automated Reasoning checks, a new feature employing formal verification methods to mathematically validate AI-generated outputs against defined rules. This innovation provides auditable and provably correct results, addressing the probabilistic nature of AI and enabling robust compliance for regulated industries. Early adopters across finance, education, logistics, utilities, and insurance are already leveraging this to enhance governance, reduce review times, and ensure verifiable AI operations.
Mike Krieger, Anthropic's CPO, resigned from Figma's board on the same day reports emerged that Anthropic is developing and planning to ship its own design software, causing Figma's stock to slide. This signals Anthropic's potential expansion into new application areas and increased competition in the design software market.
Canva rebranded itself as 'an AI platform with design tools' during its $42 billion IPO test, emphasizing its new AI 2.0 features that allow users to describe projects in plain English for editable designs with persistent memory and orchestration. This strategic shift positions Canva as a major AI player in the creative space.
Factory raised $150 million from Khosla Ventures at a $1.5 billion valuation to develop autonomous coding agents that dynamically switch between AI models based on task complexity. This significant funding round intensifies the competition in the rapidly growing market for AI coding agents.
OpenAI launched GPT-Rosalind, its inaugural life-sciences model designed for applications in biochemistry, genomics, and drug discovery, currently in research preview with partners like Moderna, Amgen, and Thermo Fisher. This marks OpenAI's strategic entry into the specialized and high-impact field of AI for scientific research.