July 2026 Summaries

Cohere has become one of the first companies to sign the EU Code of Practice on Transparency of AI-Generated Content, underscoring its dedication to responsible AI governance and compliance with the EU AI Act. This voluntary code helps AI system providers demonstrate adherence to Article 50 of the EU AI Act, which emphasizes the transparency and integrity of AI systems and outputs in Europe. By signing this Code, Cohere seeks to support Europe's strategic objectives for transparency, accountability, and innovation in AI, particularly in serving enterprise and public sector clients. This move aligns with Cohere's broader commitment to trusted AI development, as reflected in their previous signing of the EU AI Act Code of Practice on General Purpose AI Models. Cohere intends to continue its proactive collaboration with EU partners and the AI Office to facilitate the successful implementation of the AI Act, contributing to Europe's leadership in responsible AI governance.

In a high-pressure wealth management environment, integrating AI tools like North's agentic AI platform has transformed Dave's workflow by automating tasks that were once manual and time-consuming. By utilizing a series of AI agents—Client Intelligence, Research Synthesis, and Client Communication—Dave is able to gain comprehensive insights into clients' portfolios, synthesize market data quickly, and produce draft communications efficiently, allowing him to focus more on relationship-building and strategic decision-making. These tools also enhance compliance by providing structured documentation and scalable workflows that adapt to regulatory complexities, while operational agents improve post-trade processes by reducing settlement friction and enabling faster, more accurate exception handling. The implementation of AI not only streamlines Dave's day-to-day tasks but also improves client experiences and operational resilience by offering faster, more consistent communications and reducing the need for manual oversight.

Cohere has launched North Automations, an AI-driven workflow orchestration platform designed to bridge the gap between AI potential and practical implementation for enterprises, addressing challenges such as narrow workstreams, lack of governance, and high costs. North Automations aims to shift from isolated task automation to coordinated, outcome-driven workflows, providing a centralized layer for coordinating diverse agents while ensuring enterprise-grade governance and security. This new platform allows users to convert complex workflows into simple outputs, maintain control over each step, and govern AI usage safely at scale. Implementing North Automations across various business units, Cohere has demonstrated its capability to streamline processes, such as marketing operations, customer success management, and sales preparation, by integrating various data sources and providing actionable insights in real time. The platform is available to all North customers, offering integration flexibility and robust security measures, marking a significant advancement in making agentic AI both practical and scalable for daily business operations.

Cohere and the University of Toronto have announced a multi-year partnership aimed at advancing the responsible adoption of AI across the university. This collaboration will integrate Cohere's sovereign, enterprise-grade AI technology into U of T's upcoming university-wide AI platform, aiding in teaching, research, and administrative tasks while ensuring data security. Cohere's North platform will serve as an orchestration layer, allowing faculty, staff, and students to efficiently manage tasks and access information. Additionally, the partnership includes support for U of T's AI Kitchen, a secure environment for exploring AI tools. This initiative not only strengthens Cohere's ties to its alma mater but also positions U of T to become a leader in responsible AI adoption on a global scale.

Cohere offers a range of AI-driven products and solutions, including North, Compass, Command, Transcribe, Embed, Rerank, and customization options, catering to industries such as technology, energy, financial services, healthcare, manufacturing, public sector, and telecommunications. The company provides various deployment options and maintains a Model Vault for users. Cohere supports its community through resources like blogs, customer stories, developer events, and LLM University, alongside comprehensive documentation and release notes. The company emphasizes security and trust through its Security Trust Center and offers legal resources through its Legal Center. Cohere invites businesses to stay updated on AI advancements by subscribing to their updates, with the assurance of privacy and the option to unsubscribe at any time.

Expedition Tiny Aya is paving the way for technology that adapts to human needs by utilizing open-source datasets, benchmarks, and methodologies to foster AI that bridges languages, cultures, and communities, thereby democratizing knowledge and opportunities globally. The initiative, which is part of the Cohere Labs Open Science Community, has gained traction with its core research being accepted to COLM 2026 and participation in projects like the Hugging Face Build Small Hackathon. Cohere Labs is a global network of researchers, students, and practitioners working collaboratively to advance AI research, offering programs, mentorship, and community events to encourage diverse contributions to AI's future. Madeline Smith, Operations and Community Manager at Cohere Labs, emphasizes the community-driven nature of this research and invites interested individuals to engage with their initiatives.

Speculative decoding (SD) is an innovative method designed to expedite large language model (LLM) inference by proposing multiple tokens with a smaller draft model, which are then verified by a larger target model, optimizing the use of GPU resources. Traditional SD faces challenges in production environments due to dynamic batch size (BS) changes, which can render it less effective when inference becomes compute-bound. Hardware-aware dynamic speculative decoding (DSD) enhances this process by adjusting the number of draft tokens (K) based on the interaction between the model and the hardware, increasing efficiency during memory bandwidth-bound scenarios and reducing K when compute-bound. This adaptability proves advantageous across various batch sizes and model architectures, as demonstrated in benchmarks comparing vanilla, fixed-K SD, and DSD configurations. Recent contributions to vLLM incorporate DSD, ensuring compatibility with async scheduling and full CUDA Graph, thus optimizing inference efficiency and maintaining the framework's performance in dynamic environments.

Cohere has launched Cohere Transcribe Arabic, an open-source automatic speech recognition (ASR) model designed to accurately convert spoken Arabic into text, excelling in capturing the nuances and dialectal richness of the language, which is spoken by over 300 million people across numerous dialects. The model demonstrates superior performance compared to leading alternatives like Whisper and OmniASR, achieving the lowest word error rate (WER) among open-source models and excelling in bilingual speech (Arabic-English) and varied acoustic conditions. Cohere Transcribe Arabic is optimized for enterprise use, providing high throughput and accuracy, and is available under the Apache 2.0 license, allowing developers to access and deploy it freely. This development addresses the gap in AI services for Arabic speakers, offering a solution that respects linguistic diversity and is adaptable to different professional contexts.

Cohere has developed a groundbreaking solution in the field of Automated Speech Recognition (ASR) with the launch of Cohere Transcribe Arabic, which is designed to bridge the gap in AI accessibility for Arabic-speaking users, who have historically been underserved compared to English-speaking markets. This model offers state-of-the-art accuracy by achieving the lowest word error rate on the Hugging Face Arabic ASR Leaderboard and excels in preserving dialectal nuances, code-switching, and maintaining enterprise terminology. Cohere Transcribe Arabic is optimized for high-throughput, making it suitable for production environments, and allows users sovereignty over their data and models by running efficiently on consumer hardware without external cloud dependencies. This open-source model is accessible through the Cohere API and Hugging Face, promoting democratized access to speech AI technology and encouraging developers to innovate and provide feedback through various platforms.