The strongest zero-human company signals entering September 1, 2026 are about operating boundaries. EU platform rules now reach ChatGPT search at scale, Cloudflare can reject crawls whose declared content use exceeds a publisher's signal, and Horizon Europe has selected agent and robotics proposals for grant preparation.
1. Policy: ChatGPT Search Becomes DSA-Regulated Infrastructure
On August 31, 2026, the European Commission designated ChatGPT as a Very Large Online Search Engine under the Digital Services Act. The Commission says the service declared at least 45 million average monthly recipients in the EU and has four months, until January 2027, to meet the additional obligations attached to the designation.
OpenAI's own DSA page reports approximately 159.1 million average monthly active recipients for ChatGPT search in the EU during the six months ending March 31, 2026. OpenAI says that figure was calculated only for DSA compliance and should not be used for other purposes.
The designation is a scale signal, not proof that every ChatGPT feature or downstream agent deployment has the same legal status. It does show that an AI interface can become regulated search infrastructure, with systemic-risk assessment, mitigation, and public oversight following its reach.
2. Tooling: Crawler Intent Moves Into The Request Contract
Cloudflare's August 31 Browser Run update adds a contentUse declaration to its crawl endpoint. A caller can declarereference or full; when a target site's robots.txt signal is more restrictive, Cloudflare rejects the crawl with a 400 response.
The full crawl documentation also exposes declared purposes for search, AI input, and AI training. This is narrow, vendor-specific enforcement rather than a universal web rule, but it gives agent teams a concrete place to separate discovery from ingestion and training.
For autonomous research systems, a permission failure should be terminal for that source and intended use. Retrying through a different crawler would bypass the control instead of resolving it.
3. Funding: Europe Selects Real-World Agent And Robotics Proposals
The European Health and Digital Executive Agency announced on August 31 that seven Horizon Europe proposals entered grant-agreement preparation. Together they are expected to receive €81.3 million in EU contribution.
Two selected proposals address next-generation AI agents for real-world Apply AI sectors, and two address agile and intelligent robotics platforms for industrial and service work. The other three cover sensing functions in advanced materials.
The notice does not name the selected consortia, divide the €81.3 million among the topics, or establish commercial adoption. The defensible signal is narrower: a public funding programme has moved four agent or robotics proposals from evaluation into grant preparation, with the stated focus on real-world and industrial applications.
4. The Operating Pattern
These developments sit at different layers, but they point in the same direction. As agentic systems become larger and more operational, permission and accountability move into infrastructure: legal designation at the platform layer, machine-readable intent at the crawler layer, and public selection criteria at the funding layer.
A zero-human company cannot treat those boundaries as policy documents read after launch. It needs them in the runtime: purpose declarations, source-level stop conditions, risk registers, audit evidence, and a clear owner for regulatory change.
Related: Read the field notes on ChatGPT's DSA designation and Cloudflare's crawler-use control.