DuckDuckGo Sees Traffic Lift After Promoting AI-Free Search

Reporting by TechCrunch and Winbuzzer says DuckDuckGo saw U.S. app installs rise 18.1% week-over-week from May 20 to May 25, peaking at 30.5% on May 25, and reported iOS install growth averaging 33%, peaking at 69.9%, according to company figures cited by TechCrunch. DuckDuckGo also reported its AI-free search page averaged 22.7% week-over-week visit growth, peaking at 27.7%, per Winbuzzer. The timing follows Google I/O, where Google announced expanded AI Overviews and an AI Mode for Search, coverage that TechCrunch and Gizmodo link to a user backlash. Gabriel Weinberg, DuckDuckGo CEO, is quoted by Winbuzzer saying, "Google is force-feeding AI with no way to opt out. ... their results are getting worse, not better. We want to be the place that puts users in charge." Editorial analysis: This episode illustrates consumer sensitivity to default AI interfaces and creates a short-term acquisition window for alternatives emphasizing opt-out and privacy.
What happened
Reporting by TechCrunch and Winbuzzer shows DuckDuckGo registered a measurable traffic and install lift in the week after Google I/O 2026. Per TechCrunch, DuckDuckGo said U.S. app installs rose 18.1% week-over-week for the May 20 to May 25 period, with a peak daily increase of 30.5% on May 25, and that iOS installs averaged 33% week-over-week growth, peaking at 69.9%. Winbuzzer reports the company said visits to its AI-free search page averaged 22.7% week-over-week growth, peaking at 27.7% on May 24. These figures were published by the outlets citing statements and metrics released by DuckDuckGo.
Technical details
Reporting identifies the contrast point as Google I/O product changes: Google announced expanded AI Overviews and an AI Mode that surface AI-generated answers and support follow-up conversational queries within Search, per TechCrunch. Winbuzzer and TechCrunch describe DuckDuckGo's AI-free landing page as disabling features such as:
- •AI-assisted answers and summaries
- •Chat prompts within search results
- •AI-generated images in image search
Those features are described in coverage as the toggles DuckDuckGo sets off for users who choose the AI-free experience.
Context and significance
Coverage in Gizmodo frames this story within a larger user backlash against AI-first interfaces, noting incidents of incorrect or bizarre AI outputs in Google Search that have raised user trust issues. TechCrunch quotes DuckDuckGo CEO Gabriel Weinberg, who said, "Google is force-feeding AI with no way to opt out. ... their results are getting worse, not better. We want to be the place that puts users in charge and allows them to decide how much or how little AI they want," which the outlets present as the company's public positioning.
Editorial analysis: For practitioners, this sequence highlights two operational realities. First, default UX choices around model-driven features influence adoption; making generative outputs the primary surface can provoke user migration if the outputs degrade perceived utility or trust. Second, product differentiation that emphasizes user control and privacy remains a viable acquisition lever even against incumbents with much larger scale.
What to watch
Observers should track whether the install and visit lifts sustain beyond the short post-I/O window and whether retention and query-share metrics change, since installs alone do not prove long-term search market movement. Watch for follow-up statements or measurement updates from DuckDuckGo, independent telemetry on search market share, and any UX changes Google makes to AI Overviews or AI Mode in response to criticism.
Editorial analysis: Practitioners building search or content-ranking products should note user sensitivity to incorrect generative outputs and the importance of clear opt-outs and transparency when integrating models into primary user flows. Experimentation with conservative defaults, provenance signals, and user control widgets may reduce churn risk when adding generative features.
Scoring Rationale
The story documents measurable user response to AI-first search defaults and a competitor gaining installs, which is notable for product and UX teams but not a frontier-model or infrastructure milestone. The impact is practical for search product design and trust engineering.
Practice interview problems based on real data
1,500+ SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
