The Archive

Anthropic releases Claude 3.5 Sonnet and Haiku with computer use capability for autonomous task execution across applications.

Anthropic updates Responsible Scaling Policy with new safety commitments and evaluation standards for frontier model development.

Anthropic outlines measures to prevent Claude misuse for U.S. election manipulation and disinformation.

Anthropic introduces Contextual Retrieval technique to improve RAG performance by processing retrieved documents with extended context.

Salesforce integrates Claude into Einstein platform to enhance CRM AI capabilities and customer data applications.

Anthropic expands model safety bug bounty program to solicit vulnerability reports and strengthen adversarial defenses.

Anthropic makes Claude accessible in Brazil, expanding geographic availability for regional users.

Anthropic partners with Menlo Ventures to launch Anthology Fund for AI startup investing and ecosystem development.

Anthropic enables fine-tuning of Claude 3 Haiku via Amazon Bedrock for custom enterprise deployments.

Anthropic launches third-party model evaluation initiative to establish independent benchmarking standards.

Anthropic expands Claude access for U.S. government agencies with compliance and security features.

Anthropic releases Projects feature enabling persistent, multi-turn collaboration workflows in Claude interface.

Anthropic releases Claude 3.5 Sonnet with improved performance across reasoning, code, and multimodal tasks.

Anthropic outlines red teaming methodology challenges and scalability limits in adversarial AI evaluation.

Anthropic details testing protocols and mitigations for election-related AI risks and misuse scenarios.

Anthropic announces Claude availability in Canada through regional partnerships.

Jay Kreps, Confluent co-founder, joins Anthropic's Board of Directors.

Anthropic references Golden Gate Claude in San Francisco regional deployment or marketing campaign.

Mike Krieger joins Anthropic as Chief Product Officer, overseeing Claude's product direction.

Anthropic publishes mechanistic interpretability research on LLM internals, advancing understanding of model reasoning.

Anthropic outlines responsible scaling policy balancing capability gains with safety and governance measures.

Krishna Rao appointed Chief Financial Officer at Anthropic.

Anthropic updates usage policy governing acceptable Claude use cases and content restrictions.

Anthropic publishes child safety principles framework for AI model deployment and governance.

Anthropic researchers document many-shot jailbreaking attack enabling adversarial model manipulation at scale.

Anthropic advocates third-party testing as essential mechanism for AI safety assurance and policy compliance.

Anthropic partners with AWS and Accenture to deliver enterprise AI solutions built on Claude.

Anthropic makes Claude 3 models available on Google Cloud Vertex AI, expanding enterprise deployment options.

Anthropic releases Claude 3 Haiku, a lightweight model optimized for speed and cost in enterprise applications.

Anthropic announces next-generation Claude models with improved capabilities across reasoning, coding, and multimodal tasks.