AI News from June 20, 2026

Security researchers warn that 7,000 exposed Langflow servers are being actively exploited via CVE-2026-5027, a critical path traversal flaw requiring no credentials.

Norway announced sweeping restrictions on generative AI use by students, effectively banning it for children aged 6–13 and limiting use for older age groups.

Times of India draws parallels between Anthropic's Fable 5 export ban and the 1999 US government classification of Apple's Power Mac G4 as a weapon.

Nobel laureate and Google DeepMind VP John Jumper announced his departure to join Anthropic, the latest in a wave of high-profile AI talent exits from Google.

The former CEO of Allbirds is launching Smartbird, an AI infrastructure provider targeting the seemingly bottomless demand for compute to train and run deep learning models.

ZoomInfo's GTM.AI is now natively integrated into Amazon Quick Suite, AWS's agentic AI workspace, grounding AI agents in verified data on 100M companies and 500M contacts.

VentureBeat explores how hypernetworks generate task-specific model adapters at inference time, sidestepping the costs of fine-tuning and context leaks of RAG.

Enterprise software company Elastic has agreed to acquire Deductive AI, a CRV-backed startup that uses AI to automatically catch and resolve software bugs, for up to $85 million.

The US government and ASML are at odds over whether the chipmaker's most advanced EUV lithography tool has ended up in China, raising export control concerns.

About 200 organizations in Anthropic's Glasswing program, including Cisco and Dragos, retained Mythos Preview access despite the US export control shutdown.