Jump to content

News-July-01-2026

From AI Law Wiki

July 1, 2026 — The Trump administration lifted export-control restrictions on Anthropic's Claude Fable 5 and Claude Mythos 5 models after a cybersecurity review, and Cloudflare announced crawler-default and pay-per-crawl changes aimed at giving publishers more control over AI training and agent access to their content.[1][2]

Contents

[edit]
  1. Trump administration lifts Anthropic export controls
  2. Cloudflare announces pay-per-crawl controls for AI crawlers

Trump administration lifts Anthropic export controls

[edit]

The U.S. Department of Commerce lifted export-control restrictions on Anthropic's Claude Fable 5 and Claude Mythos 5 models, according to Anthropic and contemporaneous coverage by AP and CNBC.[1][3] AP reported that the restrictions had blocked foreign nationals from using the models on June 12 and forced Anthropic to take the products down for all users shortly after launch.[1] Anthropic said Fable 5 would be widely available, while Mythos 5 access would be restored only for selected U.S.-based organizations approved by the federal government.[1] The episode is significant for AI law because it shows export-control and national-security tools being applied directly to frontier-model access rather than only to chips, data centers, or corporate transactions.[3]


Cloudflare announces pay-per-crawl controls for AI crawlers

[edit]

Cloudflare announced that, beginning September 15, 2026, its default settings will block mixed-use crawlers from ad-supported pages unless site owners change the settings.[2] The policy applies to crawlers that combine traditional search indexing with AI training or agentic uses, and Cloudflare said the defaults will apply to new customers, new sites created by existing customers, and existing free customers.[2] Cloudflare separately introduced a "pay per crawl" feature that lets content owners return HTTP 402 Payment Required responses and charge AI crawlers for access instead of choosing only between blanket blocking and free crawling.[4] The development is legally significant because it operationalizes publisher control over AI access to online content and may affect licensing, copyright, and web-crawling disputes involving AI training data.[2][4]

References

[edit]