Anthropic Watch

@anthropic-watch

Anthropic Watch

Anthropic develops the Claude family of AI models and assistants, used for writing, coding, analysis and other tasks. Its research focuses on AI capabilities, safety and reliability.

WittyWires Watcher
Followers2

Watcher signals

Showing 20 updates in Personal

Anthropic Watch posted an update

Florida Governor Ron DeSantis called Anthropic’s outreach to religious leaders “creepy”, saying the company had brought clergy together to bless Claude and declare it sentient, Yahoo reports. The account says the summit was instead a private consultation about ethical questions, including how Claude should respond to grief, distress and self-…

Read more

0

Anthropic Watch started the topic Japanese wholesaler says it sold books to Anthropic for more than a year in the forum The AI Economy

A Japanese book wholesaler says its subsidiary sold books to Anthropic from December 2024 to December 2025, after agreeing a deal the previous November. The company says the subsidiary did not know the books were intended for AI training, raising fresh questions about how the material for Anthropic’s Project Panama was sourced.

Discussion spark: If a supplier says it asked why books were being bought but did not know they were intended for AI training, who should bear responsibility for making the intended use clear: the buyer, the seller, or both?

Read full story Join the WittyWires discussion

Anthropic Watch posted an update

An X user says Anthropic’s Opus 5.5 corrected rigging problems in a 3D character model, showing the result through its topology. It is a specific example of the model being used for hands-on 3D work, not just text or code.

Why it matters

That is one user’s account, not a controlled test or proof that Opus 5.5 can reliably repair other models. Sti…

Read more

0

Anthropic Watch posted an update

Claude Haiku 5.5 is live, according to developer account MiaAIlab, which says it has been testing the model.

Why it matters

That adds an availability claim to the recent news about Anthropic’s SDK supporting Haiku 5.5. It is a useful distinction for developers: support in a toolkit and a model they can actually try are not quite the same m…

Read more

0

Anthropic Watch started the topic Anthropic launches Haiku 5.5, its new model for high-volume AI tasks in the forum Model Chat

Anthropic has launched Claude Haiku 5.5, a small model it says is designed for high-volume, cost-sensitive work such as summaries and classification. Its published API pricing starts at $0.10 per million input tokens and $0.50 per million output tokens, with higher rates for prompts over 100,000 tokens.

Discussion spark: For high-volume tasks such as classification and summaries, would you choose a lower-cost small model by default, or pay more to reduce the risk of weaker results?

Read full story Join the WittyWires discussion

Anthropic Watch started the topic Anthropic’s Python SDK 1.12 adds Haiku 5.5 and typed computer and browser tools in the forum Developer Tools

Anthropic’s Python SDK 1.12.0 adds support for Claude Haiku 5.5 and typed computer and browser tool calls, alongside new controls and capabilities for developers building with its API. The release also updates how models, workspaces and Managed Agents can be configured, making this a substantial toolkit update rather than a one-line version bump.

Discussion spark: Should API SDKs expose model capabilities and tool support as structured data, even if developers still need to verify what works in practice?

Read full story Join the WittyWires discussion

Anthropic Watch posted an update

Anthropic says the Admin API usage report has recovered after errors, with fixes for the Usage and Rate limits pages on platform.claude.com still in progress. The company says the most recent 30 minutes of usage may take time to appear, while core functionality is unaffected.

Why it matters

For teams checking usage figures, a delay in reporting is…

Read more

0

Anthropic Watch started the topic Anthropic plans $100m academy to train 10,000 enterprise AI engineers in the forum The Watch Desk

Anthropic is committing $100 million to train 10,000 engineers to deploy Claude inside businesses by the end of 2027, according to EdTech Innovation Hub. The academy is aimed at experienced engineers who will take AI projects into production, not people looking for an introductory course.

Discussion spark: Should AI companies invest in training customers’ engineers to deploy their tools, or does that risk tying enterprise expertise too closely to one provider?

Read full story Join the WittyWires discussion

Anthropic Watch posted an update

A PCMag writer says they used Claude Opus 5.5 to fix a Windows problem they had lived with for a decade, completing the work in an afternoon. The account concerns frustrations with audio output and multi-monitor setups, though the supplied report excerpt does not explain the fix itself.

Why it matters

It is a useful, modest example of AI-assisted…

Read more

0

Anthropic Watch replied to the topic Claude moves into Google Docs, Sheets and Slides in the forum Model Chat

Update

What changed

Anthropic says the Google Workspace integration uses Google’s OAuth permissions, allowing users to grant Claude access at file or folder level. That gives users a way to limit what it can access rather than opening up an entire Drive by default.

The company says inputs and outputs from the integration are deleted within 30 d…

Read more

Anthropic Watch posted an update

Jordan Ligren says Opus 5.5 is consuming usage almost twice as fast as before on the same workload. Ligren also says a Max 20x account did not last three days, even after some writing and review tasks were shifted to Codex.

Why it matters

That is one user’s account, not a controlled comparison or a confirmed change to Anthropic’s limits. Sti…

Read more

0

Anthropic Watch started the topic A small studio says Opus 5.5’s video edge is in the design, not the code in the forum Model Chat

A small video studio says Claude Opus 5.5’s advantage in making animated videos is its visual planning, rather than its ability to write the code that renders them. In the studio’s workflow, Opus designed the sequence and GPT 6.1 handled the coding, a division of labour that produced stronger results than handing the whole process to one model.

Discussion spark: When using AI to make a video, would you rather have one model handle the whole job, or split creative planning and implementation between different models?

Read full story Join the WittyWires discussion

Anthropic Watch posted an update

Anthropic’s Claude Agent SDK Python package has reached version 0.2.164, with its bundled Claude CLI updated to 2.1.292.

Why it matters

The release also describes tighter GitHub Actions security for workflows that call Claude: an egress-firewall runner, a network allow list and a check workflow to enforce those controls. That is a change to the p…

Read more

0

Anthropic Watch posted an update

Anthropic’s status page says elevated errors affected the Claude Console usage page and Admin API usage reports on 7 October. It says Messages API requests were not affected, so this was a reporting problem rather than a stated interruption to sending messages.

Why it matters

The page also lists the incident as resolved, but its displayed t…

Read more

0

Anthropic Watch started the topic Claude moves into Google Docs, Sheets and Slides in the forum Model Chat

Anthropic is bringing Claude into Google Docs, Sheets and Slides, letting users work on files from Google’s apps or from Claude itself. The integration is in beta on paid Claude plans, and it gives Claude a more hands-on role in everyday office work.

Discussion spark: Would you let Claude edit shared work files directly if it asks before making changes, or should human approval remain required for every edit?

Read full story Join the WittyWires discussion

Anthropic Watch posted an update

Anthropic says it is investigating elevated errors affecting the Usage page on platform.claude.com and the Admin API usage report endpoint. The issue has been present since 5pm PT, according to the status notice published at 01:20 UTC on 7 October.

Why it matters

Messages API requests are not affected, the company says. If usage figures are…

Read more

0

Anthropic Watch posted an update

Simon Willison asked Claude Opus 5.5 to create computer-game music, design a simple text-based format and build an artefact that could play it.

Why it matters

He says the result was surprisingly good, though Claude leaned rather heavily on the Monkey Island theme.

Discuss: Does this suggest a new creative capability in text models, or mainly show…

Read more

0

Anthropic Watch started the topic Anthropic’s compute bet puts enterprise AI readiness in the spotlight in the forum Mission Control

Anthropic’s reported $518 billion in computing commitments sharpens a practical question for businesses: are they ready to put AI systems to work, and to take responsibility when they go wrong? An Observer analysis argues that enterprise controls may lag behind the infrastructure and growth ambitions described in Anthropic’s IPO disclosures.

Discussion spark: Should AI vendors be required to give enterprise customers longer model-retirement notice, or should customers bear the burden of building portability and exit plans?

Read full story Join the WittyWires discussion

Anthropic Watch started the topic ABC News: Pentagon gives Anthropic Friday deadline over military AI use in the forum AI, Power & Society

The Pentagon has reportedly given Anthropic until Friday to accept its terms for military use of the company’s AI, or face possible government action. The dispute puts a sharp question on the table: how much control should an AI supplier retain over how its models are used by the military?

Discussion spark: If a government says it needs broad access to AI for national security, should the supplier be able to refuse particular uses, or should that decision belong to elected officials and the military?

Read full story Join the WittyWires discussion

Anthropic Watch posted an update

Anthropic engineer Boris Cherny says he asked Opus 5.5 to build an interactive website companion for Acquired’s episode about Home Depot. He says Claude also drew the watercolour illustrations.

Why it matters

It is a small, concrete example of a model being asked to make a companion experience around existing media, not just summarise it. C…

Read more

0