Anthropic
Anthropic business and news from across the web.- AI Yi-Yi!NVIDIA and Google Cloud are partnering to advance agentic and physical AI technologies. Separately, Mozilla utilized Anthropic's Mythos tool to identify and resolve 271 bugs within the Firefox browser.
'Defenders finally have a chance to win, decisively': Firefox CTO raves about Claude Mythos' bug hunting capabilities after it finds 271 vulnerabilitiesMozilla partnered with Anthropic to use an early version of Claude Mythos, an AI tool, in its bug hunting efforts for Firefox. The AI successfully identified 271 vulnerabilities, with Firefox CTO Bobby Holley stating that this technology gives defenders a significant advantage in finding zero-day exploits before they can be exploited.- Anthropic is investigating 'unauthorized access' of its Mythos cybersecurity toolAnthropic is investigating reports of unauthorized access to its Claude Mythos cybersecurity tool, which was accessed through a third-party contractor portal. The company stated that the access occurred within a vendor environment and that they are looking into the claims. Claude Mythos, part of Project Glasswing, was in a limited preview with companies like Amazon, Microsoft, and Apple, and has been noted for its ability to find security flaws.
Mozilla using Claude Mythos AI Preview to help fix major security issues in FirefoxMozilla is utilizing Anthropic's Claude Mythos AI Preview, part of Project Glasswing, to identify and fix security vulnerabilities in Firefox. The AI has helped address 271 issues in the recent Firefox 150 release, though Mozilla notes it has not discovered vulnerabilities beyond the capabilities of elite human researchers.- SpaceX and Cursor strike partnership that might end in a $60 billion acquisitionSpaceX and AI company Cursor have announced a partnership that could lead to a $60 billion acquisition of Cursor by SpaceX later this year. The collaboration aims to leverage Cursor's AI coding tools and SpaceX's supercomputing infrastructure to advance AI model training. This potential deal follows SpaceX's earlier acquisition of Elon Musk's AI company, xAI.
- Mozilla says it patched 271 Firefox vulnerabilities thanks to Anthropic's Claude MythosMozilla has reported that its team successfully identified and patched 271 vulnerabilities in the Firefox browser using Anthropic's Claude Mythos Preview AI model. The foundation stated that the AI has proven capable of finding all categories and complexities of vulnerabilities that humans can discover. This initiative highlights a positive application of AI in enhancing cybersecurity.
- ChatGPT Images 2.0 is better at rendering non-Latin textOpenAI has released ChatGPT Images 2.0, an updated image generation model with improved instruction following, text rendering, and object placement capabilities. The new system features reasoning abilities, allowing it to perform tasks like web searches and output verification. It shows significant gains in rendering non-Latin text and is more flexible with aspect ratios and resolution, making it potentially useful for game prototyping and storyboarding.
- AI Yi-Yi!Amazon is investing up to $25 billion in AI company Anthropic as part of a broad deal. Meanwhile, Google is offering higher AI Studio limits to its AI Pro and Ultra subscribers, and Claude Desktop is undergoing software changes.
- Amazon will invest up to $25 billion in Anthropic in a broad dealAmazon is significantly deepening its partnership with AI company Anthropic by investing up to $25 billion. This includes an initial $5 billion investment and up to $20 billion in additional payments tied to milestones. Anthropic has committed to using Amazon's custom silicon and spending over $100 billion on AWS technologies over the next decade, while also making its Claude platform available to AWS customers.
- LinkedIn's new Crosscheck feature lets premium subscribers test competing AI models for freeLinkedIn has launched a new feature called Crosscheck, allowing Premium subscribers in the United States to test various AI models from companies like OpenAI, Anthropic, and Google without token limits. Users are presented with two AI-generated answers to a prompt and choose their preferred one before seeing which model produced it. While currently text-based, the feature aims to help users compare AI models and will eventually expand to more countries and free users.
Anthropic had a 'productive and constructive' meeting with White House officials after the preview release of its new cybersecurity-challenging AI modelAnthropic CEO Dario Amodei met with White House officials, including Secretary Scott Bessent and Chief of Staff Susie Wiles, following the preview release of its new AI model, Claude Mythos. The meeting was described as 'productive and constructive,' focusing on collaboration and safety protocols for AI development. This comes after a period of tension between Anthropic and the US government regarding AI safeguards.- Morning Safety DanceAnthropic's Mythos AI model has raised concerns about its potential to enhance hacking capabilities. Separately, Square Enix has taken action against a Japanese content creator for allegedly harassing staff working on Final Fantasy 14, marking the second such incident in two months.
- The NSA is reportedly using Anthropic's new model MythosThe National Security Agency is reportedly using Anthropic's new AI model, Mythos Preview, which is designed for computer security tasks. This comes despite ongoing legal battles and previous government restrictions on Anthropic's services. The NSA is among approximately 40 organizations granted access to the model.
- AI Yi-Yi!The White House held a meeting with AI company Anthropic to discuss concerns surrounding its Mythos model. Separately, an article from Android Authority details the author's decision to cancel subscriptions to ChatGPT Plus and two other AI services, indicating a shift in user sentiment or value perception.
- AI Yi-Yi!A dispute between Elon Musk and Sam Altman is highlighted as a battle for the soul of OpenAI, with implications for the future of artificial intelligence development. Separately, Anthropic's AI model Claude Opus was reportedly used to write a Chrome exploit for a significant sum.
- Anthropic now has a design assistant tooAnthropic has launched Claude Design, a research preview that allows subscribers to use its AI model, Claude, to generate designs, prototypes, and slides. The tool, powered by Opus 4.7, functions as a design assistant rather than a simple image generator, allowing users to refine outputs through conversation, comments, and direct edits, including adjustable sliders for visual elements. Claude Design integrates with an organization's existing visual language and can export projects to other platforms like Canva.
- Perplexity brings its Personal Computer AI assistant to MacPerplexity has launched its AI assistant software, Personal Computer, for Mac users. This software builds upon previous orchestration capabilities and acts as a suite of computer use agents that can interact with files, apps, and the web to complete complex workflows. Users can prompt the assistant via voice or from their phone, with actions performed in a secure, auditable, and reversible sandbox.
- Meta isn't setting its Oversight Board free just yetMeta's Oversight Board is seeking to expand its role beyond the company, exploring partnerships with AI firms. However, Meta has not yet approved changes to its governing documents that would allow for this. The board, funded by Meta, is also facing internal questions about its future and investment as Meta shifts its content moderation priorities.
- Anthropic will ask Claude users to verify their identities 'for a few use cases'AI company Anthropic is implementing identity verification for certain Claude chatbot features, requiring users to submit government-issued photo IDs and selfies. This move has drawn criticism regarding data privacy, especially given the use of Persona, a service also used by OpenAI and Roblox, which has ties to Palantir. Anthropic assures users that data will be encrypted and not used for model training or shared.
- AI Yi-Yi!This article discusses the increasing integration of Artificial Intelligence in various sectors, including healthcare and web browsing. It highlights new features like saving AI prompts for one-click reuse in Chrome and the growing trend of hospitals adopting chatbots to manage patient inquiries.