A growing backlash against automatically enabled generative AI features has found its sharpest articulation yet in a Wired opinion piece arguing that technology companies should stop forcing users to opt out of AI and instead default to opt-in for sensitive features.
“Opt-out toggles for automatically enabled generative AI features” have become standard practice across the technology industry, wrote Wired’s Reece Rogers. “It’s past time to make ‘opt in’ the default setting for sensitive features.”
The piece arrives amid a wave of high-profile AI opt-out controversies. In early July, Meta rolled out a feature allowing anyone using its AI app to tag public Instagram accounts and generate images using their likeness, automatically opting users in unless they navigated to a settings menu and manually disabled the feature. Google has similarly faced criticism for enabling AI Overviews in search by default, and for training its AI models on user data unless users find the buried opt-out toggle.
Wired’s argument centers on what it calls a fundamental asymmetry of effort. Companies design AI features to be turned on by default, knowing that most users, particularly less technically sophisticated ones, will never find or understand the opt-out mechanism. The burden of protecting privacy falls entirely on the individual, while the company reaps the benefit of broader data access and engagement.
“Opt-out assumes consent. That’s the wrong way around,” Rogers wrote.
The piece cites research showing that default settings are one of the most powerful determinants of user behavior in digital systems. When a feature is opt-in, adoption rates may be lower, but users who do adopt it have made an active, informed choice. When the same feature is opt-out, adoption is higher but includes users who would have declined if asked.
Regulators are beginning to take notice. The EU’s Digital Markets Act and Digital Services Act both include provisions that push platform operators toward more transparent consent mechanisms. California’s Delete Act, which took effect in 2026, allows residents to make a single opt-out request that applies to all data brokers, a recognition that the current system places an unreasonable burden on individuals.
Sources: Wired