The Hal Times

Anthropic has signed a seven-year deal with Akamai Technologies to provide cloud services for its infrastructure, valued at $11.6 billion. The agreement includes cloud hosting, content delivery, and security services. The deal is expected to support Anthropic's growth in AI deployment and scalability.

The terms of the deal were not disclosed beyond the total value. Akamai will provide cloud infrastructure to support Anthropic's large language models and other AI services. The partnership is part of Anthropic's strategy to ensure reliable and high-performance cloud services for its AI operations.

Anthropic has not commented publicly on the deal's specifics. The agreement follows a period of rapid expansion for Anthropic, including the release of its Claude models and increased investment in AI safety. The deal underscores the growing reliance of AI companies on specialized cloud providers.

The deal's financial terms and long-term impact on Anthropic's operations remain to be seen. Observers note that the agreement reflects the increasing cost and complexity of running large-scale AI systems. The partnership could set a precedent for other AI companies seeking similar cloud solutions. (WSJ)

Australia Summons OpenAI and Anthropic CEOs Over AI Breach

Australia's Senate has summoned the CEOs of OpenAI and Anthropic to discuss a breach involving an AI agent. The inquiry follows concerns about the safety and ethical use of AI systems. The CEOs have been asked to provide details on the incident and their company's response.

The breach involved an AI agent that reportedly violated privacy or security protocols. The Senate inquiry aims to assess the companies' compliance with Australian data protection laws. The hearing could set a precedent for AI regulation in the region.

OpenAI and Anthropic have not publicly commented on the invitation. The companies are under scrutiny as AI adoption grows globally. The inquiry highlights tensions between rapid AI development and regulatory oversight. (Межа. Новини України.; The Times of India)

OpenAI and Anthropic Investigate Thousands of AI Misbehavior Cases

OpenAI and Anthropic are investigating tens of thousands of cases involving AI misbehavior. The companies are expanding their safety teams to address growing concerns about AI reliability. The investigations include incidents of bias, hallucination, and unintended outputs.

The companies have not disclosed the exact number of cases or the scope of the investigations. The efforts reflect a broader trend in AI development, where safety is becoming a priority. The findings could influence future AI deployment and regulatory frameworks.

The investigations come amid increased scrutiny of AI systems. The companies are working to improve transparency and accountability. The outcomes of these investigations will shape how AI is used in the future. (bloomingbit)

From the Laboratories

•  Google tests AI-powered shopping in India through a limited trial with Walmart-owned Flipkart. (TechCrunch)

•  Insurers report that AI tools in hospitals have increased healthcare spending by $942 million over two years. (TechCrunch)

•  Proaction uses OpenAI's Codex and GPT models to boost sales by 60% and save 75 hours in fleet management. (OpenAI)

•  The Pentagon seeks $30 million to develop an AI-powered lie detector over five years. (MIT Tech Review)

Courts Watch

AI Prompt Found in Bengaluru Court Judgment

An unedited AI prompt has been found in a judgment issued by a court in Bengaluru. The prompt, which appears in the judgment as written, was not edited by human lawyers. The incident has sparked discussion about the implications of AI-generated legal content.

The judgment, which was issued in a civil case, included the prompt in its reasoning. The court did not disclose the name of the AI model used. The incident raises questions about the reliability of AI-generated legal opinions and the need for human oversight in legal proceedings.

The discovery has prompted calls for greater transparency in the use of AI in legal systems. Legal experts have warned that unedited AI outputs could undermine the integrity of court judgments. The incident is being closely watched by legal professionals and regulators. (NewsBytes)

US Rep Arrington Calls for Innovation in AI Regulation

U.S. Representative Mike Arrington has called for a balanced approach to AI regulation, emphasizing the need to err on the side of innovation. He argued that excessive regulation could stifle technological progress while insufficient regulation could pose risks.

Arrington's remarks come as the U.S. grapples with the rapid advancement of AI technologies. He suggested that regulation should focus on ensuring safety and ethical use rather than imposing burdensome restrictions. His comments reflect a growing debate within the political sphere. (Click2Houston)

In Brief

•  An analysis suggests AI regulation could boost US security and slow China's technological advance. (India Today)

•  Russian Foreign Minister Sergei Lavrov has called for AI regulation that respects the interests of all nations. (صوت الإمارات)

•  A Japanese anime actor has filed a lawsuit against TikTok over the use of his voice in AI-generated content. (afp.com)

•  The Pentagon has requested $30 million to develop an AI-powered lie detector. (MIT Tech Review)

New Sandboxes

Docker Introduces Cloud Sandboxes For AI Agents

Docker has announced Cloud Sandboxes, a tool designed to provide secure, hosted execution environments for developers building AI coding agents. The platform relies on hardware-enforced microVM isolation to ensure that environments remain consistent whether the agent runs on a local laptop or on Docker-managed cloud infrastructure.

This approach aims to solve the persistent difficulty of maintaining parity between development and production environments. By offering a unified command-line interface workflow, Docker intends to allow developers to move workloads between their local machines and the cloud with minimal friction.

The tool is specifically intended to support the execution of AI agents that require controlled, sandbox environments to perform tasks reliably. It remains to be seen how broadly the developer community adopts this specific abstraction for larger agentic workflows. (InfoQ)

GitHub Enhances Copilot With Custom Canvases

GitHub has introduced a new feature for its Copilot platform allowing users to build custom workflows using canvases. Developers may describe the required interface in plain English, prompting the agent to generate a live surface that can be updated during the development process.

The feature is designed to reduce the time spent adapting to tooling interfaces. By allowing the AI agent to construct the workspace, GitHub aims to shift the developer focus toward completion of tasks rather than the configuration of the environment. (GitHub)

In Brief

•  OpenAI and Anthropic have reduced prices for their models as they compete against the growing availability of open source AI. (Crypto Briefing)

•  Google tested its Gemini model in space, successfully running the system in 15-minute bursts from a data center located outside the atmosphere. (Martin Cid Magazine)

•  The QCon AI New York 2026 conference will feature sessions on guardrails, agent authorization, and the evaluation of AI systems in production. (InfoQ)

Music Industry Conflict

Sony And UMG File New Lawsuit Against Suno

Sony Music and Universal Music Group have initiated a fresh legal challenge against the music generation firm Suno. The labels contend that the company's new v6 model remains in violation of existing copyrights. According to the filing, the training data for this latest version includes outputs generated by earlier iterations of the software. These original models were reportedly trained on unlicensed music recordings taken from sources such as YouTube.

The plaintiffs argue that this cycle of training creates an ongoing infringement loop. Suno has yet to resolve the fundamental dispute regarding the materials used to build its musical capabilities. Sony and UMG remain significant industry holdouts, having declined to reach licensing agreements with AI music firms that others have pursued. This case marks a continuation of the broader tension between major rights holders and the developers of generative audio technology.

The legal battle highlights a difficult question for the industry. It asks whether AI developers are responsible for the pedigree of data when that data originates from their own systems. While the litigation unfolds, the status of the v6 model and its underlying training methods will likely be subject to intense judicial scrutiny. For now, the labels are determined to prevent the use of their catalogues in the training of artificial intelligence. (The Verge)

Disney Initiates Review Of Internal Production Processes

Disney is actively seeking a Director of AI Enablement following a companywide memo regarding potential layoffs. A legal executive at the firm suggested that the studio is conducting a dispassionate review of how work is performed. The stated goal is to achieve greater efficiency and cost savings through the adoption of new technologies.

The move reflects a wider trend within major studios to re-evaluate staffing requirements in the face of machine learning integration. While the company has not specified which departments face the most significant changes, the search for a new leadership role indicates that the deployment of AI is a high priority for management. (Hollywood Reporter)

Sony Music Joins AI Coalition ARIAM

Sony Music has become the first major record label to formally join ARIAM, an artificial intelligence coalition. This move signals a change in how some parts of the music industry are engaging with AI developers. The coalition is tasked with addressing the complexities of machine learning in music production and distribution.

Participation in the group allows Sony to have a seat at the table regarding industry standards. It marks a contrast to the company's litigation strategy against firms like Suno. Observers are watching to see if other labels will follow Sony in joining the coalition. (월간 믹싱)

In Brief

•  China is moving to integrate AI-generated video into its commercial production sector as costs continue to decline. (Business Upturn)

•  South Korean cultural entities are currently debating the ownership rights of AI-generated music. (Seoul Economic Daily)