Google Ads language targeting removal shifts Search and Performance Max to AI

The removal of manual Google Ads language targeting marks a definitive pivot from auditable user controls to black-box model inputs. As privacy-focused browser changes like Chrome’s header reduction limit deterministic data, platforms are forced to rely on behavioral inference to reach multilingual audiences. This shift aligns with the broader industry trend of converting granular settings into AI-driven signals, similar to recent changes in Meta’s location targeting and Google’s Demand Gen lookalikes. For strategists, this complicates performance attribution during a month already crowded with AI Max upgrades and bidding changes. Watch for whether Nielsen’s revised Spanish-speaking household estimates in late September trigger significant volatility in automated valuation for multilingual inventory.

Google's removal of manual language targeting fits within a broader pattern of the company consolidating advertiser controls into automated systems. In May 2026, Google announced that AI Max for Search campaigns would expand to all advertisers with new creative optimization and keyword expansion features, effectively bundling bidding, targeting, and creative decisions under a single algorithmic layer. Performance Max campaigns, which already operate with minimal manual input, saw average conversion lifts of 18% when advertisers migrated from standard Shopping and Display campaigns, according to Google's own published benchmarks. These moves signal that language targeting's removal is not an isolated change but part of a systematic reduction of manual levers across Google Ads. The IAB Tech Lab has been working to standardize how language and audience signals are passed in programmatic transactions, which affects how platforms like Google infer user attributes. In March 2026, the IAB Tech Lab released updated OpenRTB 3.0 specifications that include a revised language signal taxonomy designed for privacy-compliant inference, replacing deterministic browser-language headers with probabilistic content-based signals. Nielsen, meanwhile, has been recalibrating its audience measurement methodology to account for multilingual viewing. Nielsen announced in July 2026 that its revised Spanish-language household estimates would incorporate streaming-first panel data, a change that could affect how automated systems value multilingual inventory in programmatic auctions. Display and Video 360, Google's programmatic buying platform, is expected to align its language inference logic with these updated IAB standards. From a technical standpoint, the shift from explicit language targeting to behavioral inference mirrors changes already underway in other ad platforms. Microsoft Advertising introduced automated audience expansion for its Search campaigns in early 2026, using similar content-signal models to determine ad relevance across language boundaries. The privacy dimension is equally significant: Chrome's ongoing reduction of Accept-Language header granularity, part of the Privacy Sandbox initiative, means that deterministic language signals available to ad platforms have declined by roughly 40% since 2024, forcing platforms to build inference layers that reconstruct what was once freely available. For advertisers managing multilingual campaigns, the practical implication is that performance attribution will increasingly depend on conversion-level signals rather than targeting-level controls, making A/B testing across language segments more complex without third-party verification tools.