#Safety — DailyPrompt

▮#Safety

Business11 Jul 2026 · 1 min

NYT-led group asks judge to sanction OpenAI over hidden evidence

News outlets allege OpenAI falsely claimed it could not search its models for copyrighted material, while having already done so before the suit was filed.

Business10 Jul 2026 · 1 min

US closes loophole letting Chinese firms buy Nvidia chips overseas

Updated Commerce Department guidance requires export licenses for Blackwell-series chips reaching any China- or Macau-headquartered entity, anywhere in the world.

Business10 Jul 2026 · 1 min

Siri AI rollout skips the EU and China at launch

Apple's new assistant reaches US customers in English first, with international availability delayed by regulatory requirements abroad.

Business10 Jul 2026 · 1 min

Government of Alberta deploys Claude to hunt cybersecurity flaws

The Canadian provincial government is using Anthropic's model to find and fix vulnerabilities across its own systems.

Business9 Jul 2026 · 1 min

Appeals court hears first fair-use argument in AI training case

The Third Circuit's review of Thomson Reuters v. Ross Intelligence is expected to shape every generative AI copyright case still in the pipeline.

Tools9 Jul 2026 · 1 min

Midjourney seeks to expose studios' own AI use in copyright fight

The image generator wants discovery into how major studios themselves use generative AI, arguing it undercuts their infringement claims.

Business9 Jul 2026 · 1 min

Academic integrity worries schools even as AI adoption accelerates

Roughly four in ten students and educators cite integrity concerns, while under half of schools have offered any formal AI guidance.

AI8 Jul 2026 · 1 min

White House set to finalize voluntary AI jailbreak standards

Anthropic, OpenAI, Google, Microsoft and Amazon are expected to adopt a shared severity scale and 30-day pre-release review by early August.

AI8 Jul 2026 · 1 min

Anthropic proposes a CVSS-style scale for AI jailbreak severity

The Cyber Jailbreak Severity scale rates exploits across four axes, summing them into five bands from Informational to Critical.

Business8 Jul 2026 · 1 min

Report: 88% of organizations hit an AI agent security incident this year

Nearly half of surveyed organizations now expect agentic AI to be the top attack vector by the end of 2026.

Models7 Jul 2026 · 2 min

Anthropic releases Claude Fable 5, its first Mythos-class model

Fable 5 tops agentic coding leaderboards and introduces a new safety tier system that separates general availability from approved-organization access. Demo content.

Business6 Jul 2026 · 1 min

EU AI Act reaches full applicability on August 2

The remaining provisions of Europe's landmark AI law take effect this summer, two years after it first entered into force.