Keldura Daily Open Keldura

Keldura Daily · Technology and AI

AI acquisitions, agent tooling, and safety tensions dominate the latest batch of tech updates

The evidence points to a fast-moving week in AI: rumors of a major robotics acquisition, new products aimed at AI-native workflows, a test of AI-generated children’s stories, a cybersecurity incident involving pre-release models, and fresh model releases focused on agent performance and safety. The common thread is how quickly AI is moving from model capability into products, infrastructure, and real-world operational risk.

The field note

1 source · 2 items
  1. Physical Intelligence had reportedly raised more than $1 billion and was said to be in talks this spring for an…
  2. OpenAI and Anthropic have both been on acquisition sprees, and both are also preparing to go public after confi…
  3. Anthropic has not built a hardware robotics lab of its own, while OpenAI has publicly moved back into robotics…
Story 011 source

Anthropic-Physical Intelligence rumors highlight the race for robotics expertise

A weekend rumor that Anthropic was acquiring Physical Intelligence spread quickly, even after a denial from Physical Intelligence’s CEO, because the company is a well-funded robotics startup with a widely used robot-brain model and prior acquisition talks with Anthropic this spring were later reported. The reporting also notes that OpenAI is already an investor in Physical Intelligence, which complicates any potential deal dynamics between the two rivals. [1]

Why it matters

Robotics is emerging as a strategic frontier in the AI race, and ownership of a strong robotics team could shorten the path to systems that understand the physical world. The story matters because it shows how acquisition rumors, investor relationships, and IPO timing can shape who controls the next layer of AI infrastructure. [1]

Key insights

  • Physical Intelligence had reportedly raised more than $1 billion and was said to be in talks this spring for another $1 billion round at an $11 billion valuation. [1]
  • OpenAI and Anthropic have both been on acquisition sprees, and both are also preparing to go public after confidential IPO filings in June. [1]
  • Anthropic has not built a hardware robotics lab of its own, while OpenAI has publicly moved back into robotics through OpenAI Robotics. [1]
  • OpenAI’s existing investment in Physical Intelligence could give it leverage if a sale to a competitor were ever on the table. [1]
Story 021 source

Meta is piloting StoryKit, an AI bedtime-story app for children

Meta is testing StoryKit, an AI storytelling app that generates personalized children’s stories with custom characters, settings, lessons, and music. Meta says the pilot is running in select countries, is limited to users over 18, and includes AI safety filters and no social features. [2]

Why it matters

This is another example of Meta pushing AI into everyday consumer behavior, including family and parenting use cases. The product raises questions about whether generative AI is becoming normalized as a substitute for human creativity in intimate, emotionally meaningful activities. [2]

Key insights

  • The app lets parents create a character by snapping a photo of a toy or person and then building a story world and lesson around it. [2]
  • Meta says StoryKit is meant to help craft personalized, imaginative storybooks for children. [2]
  • The product is available only to adults, suggesting Meta is trying to control direct child access while targeting parents as the users. [2]
  • The app fits into a broader pattern of Meta shipping experimental AI features that attract criticism. [2]

Create your own daily briefing — start free. Keldura monitors the sources you choose and gives you a private, grounded daily digest with cited answers.

Create your own daily briefing — start free