Claude's AI research mode now runs for up to 45 minutes before delivering reports
Briefly

The article discusses the recent performance of Claude Research, highlighting the potential for confabulated quotes, where AI-generated content combines multiple sources inaccurately. Although the report was well-crafted overall, deeper verification of sources is recommended for serious research applications. Additionally, Anthropic has launched a new 'Integrations' feature that enhances Claude's capabilities to search and interact with various applications. Currently, Claude is compatible with ten services including Atlassian and Zapier, illustrating significant advancements in AI data access and productivity integration.
One must always be careful with confabulated quotes in AI because even outside of this Research mode, Claude 3.7 Sonnet tends to invent plausible ones to fit a narrative.
Claude Research did a relatively good job crafting a report on this particular topic. Still, you'd want to dig more deeply into each source and confirm everything if you used it as the basis for serious research.
The new 'Integrations' feature allows Claude to work with remote Model Context Protocol (MCP) servers across web and desktop applications.
At launch, Claude supports Integrations with ten services, including Atlassian's Jira and Confluence, Zapier, Cloudflare, Intercom, Asana, Square, Sentry, PayPal, Linear, and Plaid.
Read at Ars Technica
[
|
]