Skip to content

Anthropic

Anthropic launches Claude Corps program

Written by an AI model from the linked sources and checked against them by a second model. How we work

Image: anthropic.com

The company is investing $150 million in a fellowship program for early-career professionals. Over the next year, 1000 participants will work at US non-profits, deploying Claude-based tools for process automation and data analysis.

The program targets people with less than two years of work experience. Each fellow will receive a $85,000 annual stipend, benefits, and intensive AI training. Partners include CodePath, which handles hiring and training, and Social Finance, which will assess program effectiveness.

The inaugural cohort will work with 400 organizations, including food banks and education centers. Applications for the first 100 spots are open until July 17, 2026. Anthropic plans to make program infrastructure open-source to scale the initiative globally.

Anthropic

Anthropic makes Fable 5 safety mechanisms transparent

Anthropic admits that Fable 5’s hidden safety filters were a mistake. Now, when safety mechanisms trigger, the model switches to Opus 4.8 with an explanation. This aims to reduce unexpected blocks, though it may increase false positives temporarily as Anthropic tunes its classifiers.

No comments yet. Be the first.

Join the conversation

Sign in to leave a comment. It appears after moderation.