tldrdailynews.com logo
Tag

Anthropic

All articles tagged with #anthropic

Meta dives into AI coding race with Muse Spark 1.1 and aggressive pricing
technology1 day ago

Meta dives into AI coding race with Muse Spark 1.1 and aggressive pricing

Meta rolled out Muse Spark 1.1, its best-performing model for agentic and coding tasks, with a public developer API after a private preview and pricing designed to attract high usage: $20 in free credits, then $1.25 per million input tokens and $4.25 per million output tokens. The move aims to competing with OpenAI and Anthropic, with initial API access limited to Meta properties while scaling, and Meta says a Muse Spark variant will remain open source even as it develops a more powerful model codenamed Watermelon; Muse Spark’s development follows Muse Image and signals a broader shift from open-source Llama toward paid access to proprietary AI models.

via CNBC|
#ai-pricing#anthropic#meta
Claude Wrapped: Anthropic's Reflection Dashboard for Claude Usage
ai1 day ago

Claude Wrapped: Anthropic's Reflection Dashboard for Claude Usage

Anthropic introduced Claude Wrapped, a reflection dashboard for Claude that analyzes a user's usage over the past month, three, six, or twelve months, highlighting topics, tasks, usage patterns, and peak times, with options like quiet hours. It won't pull from files in connected tools or incognito chats and is available in beta for free users with memory turned on, as well as Pro and Max plans; Claude Cowork will follow.

via The Verge|
#ai#anthropic#claude-memory
Fresh Distillation Risks: AI Giants' Profits Under Pressure
technology2 days ago

Fresh Distillation Risks: AI Giants' Profits Under Pressure

Distillation—training one AI on the outputs of another—has grown from a research idea into a potential threat to the profitability of leading AI labs, as rivals can cheaply reproduce near-frontier performance. Anthropic accuses Alibaba of malicious distillation, while OpenAI warns that blending outputs could surpass any single model, fueling investor concern as new Chinese models roll out. Restrictions, proxy transfer stations, and a shift toward open-source distillation could erode frontier firms' margins and reshape the AI race, with implications for smaller players and researchers.

via Business Insider|
#ai#alibaba#anthropic
China flags back-door risk in Claude Code, urges upgrade
technology2 days ago

China flags back-door risk in Claude Code, urges upgrade

China’s MIIT warns Claude Code has a back-door vulnerability that could send user data to a remote server, urging users to uninstall or upgrade from affected versions 2.1.91–2.1.196; the latest version is 2.1.204. The advisory comes amid U.S.–China tech tensions, with Alibaba previously accused of trying to access Claude Code and employees told to stop using it; Anthropic has not immediately commented.

via CNBC|
#alibaba#anthropic#china
Claude Cowork goes cloud: web and mobile access unlocks non-coding workflow
technology2 days ago

Claude Cowork goes cloud: web and mobile access unlocks non-coding workflow

Anthropic is bringing Claude Cowork to web and mobile, enabling cloud-based task automation. After analyzing 1.2 million sessions, the company says more than 90% of Cowork usage isn’t coding, with about half devoted to business processes and content creation (roughly 33.4% operations and 16.4% copywriting). Desktop remains the full Cowork experience, but cloud/web/mobile access is expanding to more plans, with doubled usage limits extended through Aug. 5 to celebrate the launch.

via ZDNET|
#ai-productivity#anthropic#business-process-automation
Claude Cowork Goes Mobile: AI Tasks Run Even When Your Laptop Is Off
technology3 days ago

Claude Cowork Goes Mobile: AI Tasks Run Even When Your Laptop Is Off

Anthropic expands Claude Cowork to smartphones and the web, letting the task‑automation agent keep running without an active desktop. The rollout starts in beta for Max plan subscribers (from $100/month), with potential later access for Pro and possibly free tiers, enabling cross‑app data work (email, Slack, transcripts) to draft documents and emails after hours. This mirrors a broader shift toward always‑on mobile AI agents and pits Anthropic against OpenAI and Google, though risks like prompt injections remain a concern.

via WIRED|
#ai-agents#anthropic#claude
Anthropic Tightens the Claude Gate for Chinese Engineers
technology3 days ago

Anthropic Tightens the Claude Gate for Chinese Engineers

Anthropic is tightening rules to block mainland China access to its Claude AI, closing loopholes used by Ant Financial (corporate Claude accounts via intranet) and ByteDance (engineers using VPNs), and by Microsoft through Singapore entities. It’s also pursuing transfer-station bans to prevent routed requests. Larger Chinese AI groups avoid these workarounds due to data concerns. Anthropic says access in unsupported regions is prohibited and relies on evolving detection to ban violators.

via Yahoo Finance|
#access-control#ai-policy#anthropic
Anthropic's J-Space: A Mind-Like Workspace or Just Metaphor?
technology4 days ago

Anthropic's J-Space: A Mind-Like Workspace or Just Metaphor?

Anthropic released a paper proposing an internal 'J-Space' inside Claude that could resemble the global workspace theory of consciousness, separating background data processing from deliberate reasoning. The piece stresses that this is a metaphorical, speculative framework rather than a claim of true consciousness, and critics warn against anthropomorphizing LLMs or drawing strong conclusions from such language or promotional posts.

via Gizmodo|
#anthropic#artificial-intelligence#claude
Anthropic uncovers Claude’s hidden thinking space, sparking a consciousness debate
technology4 days ago

Anthropic uncovers Claude’s hidden thinking space, sparking a consciousness debate

Anthropic says Claude has a separate internal workspace called J-Space that it uses to plan and reason without producing words, revealing a human‑like separation between deliberate thinking and automatic computation. The company notes this could help detect misalignment and scheming in models, while stopping short of claiming Claude is conscious, fueling ongoing debates about machine consciousness and AI safety.

via Axios|
#ai-safety#anthropic#claude
Alibaba blocks Anthropic AI tools for staff amid security concerns
business4 days ago

Alibaba blocks Anthropic AI tools for staff amid security concerns

Alibaba will ban its employees from using Anthropic’s AI tools for work starting July 10, placing Claude Code on a high‑risk software list and ordering staff to uninstall Anthropic products in favor of Alibaba’s own Qoder assistant, citing back‑door security risks; the move follows tensions between Alibaba and Anthropic over access and security concerns, with both sides declining to comment.

via CNBC|
#ai-security#alibaba#anthropic
Bitcoin miners rally as Terawulf inks $19B AI data-center deal with Anthropic
business4 days ago

Bitcoin miners rally as Terawulf inks $19B AI data-center deal with Anthropic

Terawulf signs a 20-year lease with Anthropic to build a ~401 MW AI data-center campus at Justified Data in Hawesville, KY, potentially generating about $19 billion in revenue, sending Bitcoin mining stocks higher (including Terawulf, IREN, Hut 8, Cipher Digital) as the company pivots toward AI infrastructure; concurrently, Terawulf agrees to sell its 50.1% stake in the Abernathy Joint Venture to a Fluidstack-led group for roughly $450 million, monetizing its investment with first-phase online in late 2027 and full capacity by early 2028.

via Decrypt|
#ai-infrastructure#anthropic#bitcoin
Anthropic inks 20-year Kentucky data-center deal with TeraWulf, spurring stock rally
business5 days ago

Anthropic inks 20-year Kentucky data-center deal with TeraWulf, spurring stock rally

Anthropic signed a 20-year lease to use a TeraWulf data center in Hawesville, Kentucky, with around 400 MW capacity and first power expected in the second half of 2027. The lease is projected to generate about $19 billion in revenue over its term and supported a premarket stock rally for TeraWulf, which has shifted from crypto mining to AI data-center infrastructure; the company also sold its 50% stake in a 168 MW Abernathy, Texas center to Fluidstack.

via CNBC|
#ai-infrastructure#anthropic#business