Japan's Three Major Banks to Obtain Access to Anthropic's Mythos by End of May, Public-Private Collaboration to Address Security Risks

According to reports, the banking divisions of Mitsubishi UFJ Financial Group, Mitsui & Co. Financial Group, and Mizuho Financial Group are expected to gain access to Anthropic's AI model, Mythos, as early as late May, marking the first time Japanese companies have been approved to connect to the system. The report states that the three major banks may have learned about this news during a meeting held in Japan on Tuesday, facilitated by U.S. Treasury Secretary Scott Bessenet.

Mythos is designed for defensive cybersecurity tasks, but its powerful capabilities have also raised concerns about potential threats to traditional software security. Anthropic discovered multiple critical vulnerabilities in the system during a preview test, affecting all major operating systems and web browsers.

The Japanese government responded quickly. After meeting with Bessenet on Tuesday, Japanese Finance Minister Osamu Tanaka stated that Japan will establish a public-private working group this week to address the cybersecurity risks posed by Mythos to Japan's financial system. The first meeting is expected to be held on Thursday.

Secret Messages to AI Overheard by Others? Meta Launches Invisible Chat

Meta launches 'stealth conversation' feature on WhatsApp, enabling fully private chats with Meta AI. Content operates in an isolated secure environment, inaccessible to anyone. Users simply tap the stealth icon in the chat interface to activate; standalone Meta AI app will also support this feature.....

Anthropic Announces Increased Claude Code Quotas: Weekly Limit Increased by 50% Until July 13

Anthropic announced a temporary 50% increase in the weekly usage limit for Claude Code, effective until July 13, 2026. This adjustment, which stacks with the previous 'doubled to 5 hours' limit, grants developers significantly enhanced code generation capabilities. The change is automatic and requires no user action, aiming to strengthen Anthropic's developer ecosystem.....

French AI Star Mistral Promotes Cybersecurity Model, Targets Banks Without Access to Mythos

French AI startup Mistral AI is collaborating with multiple European banks to launch a new cybersecurity AI model designed to quickly detect security vulnerabilities. The model is similarly positioned to Anthropic's Mythos, but Mythos has limited access and is not available to all banks. Mistral aims to provide a broader solution.....

Google Releases AI Notebook Platform Googlebook: Gemini Model Redesigns Pointer Interaction and System Infrastructure

Google launches Googlebook, an AI-driven notebook platform shifting personal computing from traditional OS to a model-centric logic. It deeply integrates Gemini at the system level, embedding AI into the mouse cursor to analyze screen content in real-time and proactively respond, revolutionizing human-app interaction.....

Kuaishou Technology Board Evaluates Restructuring of Ke Ling AI Business, Possible Introduction of External Financing

Kuaishou Technology announced that its board is evaluating a restructuring plan for Kuaishou Ling (Kling) AI assets, potentially involving external financing. Kling is Kuaishou's self-developed video generation large model, expected to launch in June 2024. On January 31 this year, the Kling 3.0 series was released, including image, video, and Omni versions, with technical upgrades offering richer content.....