AI Roundup — July 22, 2026
Glow Exits Stealth at $1.2B Valuation to Address AI-Era Endpoint Security
According to TechCrunch, a startup called Glow has emerged from stealth with a $1.2 billion valuation, positioning itself to address a new category of endpoint security risks tied to the enterprise adoption of AI agents and developer tools.
As reported, Glow is targeting the specific security challenges that arise when organizations deploy AI-powered tooling at scale across their internal environments. The company's focus is on the endpoint layer, which it argues has become a distinct and underserved attack surface as AI agents gain broader access to enterprise systems and data.
No specific funding round figures beyond the valuation were detailed in the report at the time of publication.
Synthesia Launches AI Roleplay Sessions for Enterprise Training
Synthesia, the AI video generation company, has announced the launch of AI Roleplay Sessions, an interactive training product aimed at enterprise customers, TechCrunch reports.
Rather than delivering pre-recorded video content, the new platform allows employees to practice workplace conversations in real time with AI avatars. According to the report, the system provides feedback, scoring, and analytics at the conclusion of each session, giving organizations a measurable way to assess training effectiveness.
The move marks an expansion of Synthesia's product line beyond its core video generation offering, extending into live, interactive coaching experiences. The company is targeting corporate learning and development use cases where conversational practice and performance tracking are key requirements.
Meta Tests AI-Powered Bedtime Story App
TechCrunch reports that Meta is in the early testing phase of an AI application designed to generate bedtime stories. The app represents one of Meta's exploratory consumer AI products being evaluated ahead of any potential broader release.
Few technical specifics were disclosed in the report, but the product appears to leverage generative AI to produce on-demand narrative content for users. Details regarding the underlying model, target platform, or timeline for a wider rollout have not been publicly confirmed by Meta at this stage.
These three developments reflect continued momentum across enterprise AI security, corporate training tooling, and consumer-facing generative AI applications. Further details on each product are expected as the companies provide additional disclosures.