Google expands AI training data collection by default
Google has quietly expanded its data collection practices to fuel AI training, shifting default privacy settings to permit broader use of user information for model development. This move reflects the intensifying competition for training data among frontier labs, where scale directly correlates with model capability. The change underscores a critical tension in the AI industry: companies need vast datasets to remain competitive, yet face mounting pressure from regulators and users over data governance. The availability of opt-out mechanisms signals regulatory awareness, but the burden now falls on individual users to actively protect their data rather than companies seeking explicit consent upfront.
Modelwire context
Analyst takeThe more pointed issue isn't the opt-out itself but the direction of the default. Shifting from opt-in to opt-out is a structural policy choice that quietly expands the training corpus without requiring any active user decision, and that asymmetry is where the real leverage sits.
This move fits directly into the data sourcing pressure we covered when Cloudflare began enforcing separation between search indexing and AI training crawlers (July 1), effectively tightening the external supply of web content for training. When third-party data becomes harder to acquire through crawling, the logical response is to extract more value from first-party user data, which is exactly what Google's default change accomplishes. The Platformer piece from July 2 on the widening gap between deployment pace and harm mitigation is also relevant here: regulators and users are already skeptical, and moves that shift consent burdens onto individuals rather than companies will accelerate that friction.
Watch whether the EU's data protection authorities issue a formal inquiry into this default change within the next 90 days. A challenge there would force Google to either reinstate opt-in defaults in European markets or defend the policy under GDPR, which would set a concrete precedent for how other frontier labs structure their own first-party data collection.
Coverage we drew on
This analysis is generated by Modelwire’s editorial layer from our archive and the summary above. It is not a substitute for the original reporting. How we write it.
MentionsGoogle
Modelwire Editorial
This synthesis and analysis was prepared by the Modelwire editorial team. We use advanced language models to read, ground, and connect the day’s most significant AI developments, providing original strategic context that helps practitioners and leaders stay ahead of the frontier.
Modelwire summarizes, we don’t republish. TechCrunch - AI originally reported this story as “If you use Google, you’re training its AI. Here’s how to opt out.”. The full content lives on techcrunch.com. If you’re a publisher and want a different summarization policy for your work, see our takedown page.