I Tested 5 “Private” Browsers — Only One Didn’t Spy

Product reviewed

The video is a comparison/test of five “private” desktop browsers using Wireshark packet captures on a clean install, measuring what each browser transmits during ~15 minutes of idle startup (before the user visits sites). The goal is to see whether real network behavior matches the browsers’ privacy marketing.

Key points mentioned about the browsers

1) Brave (generally strong, but has unexpected advertising-network connections)

Scale/benchmarks mentioned

• ~80M monthly active users; positioned as privacy-focused and Chrome-compatible.
• privacytests.org rating: 143/156 for blocking tracking/fingerprinting (near the top).

What the network capture showed (startup/idle)

• “Normal/expected” items: update checks, component verification, and safe browsing database updates through Brave’s own infrastructure rather than Google (framed as an improvement vs Chrome defaults).
• Unexpected: connections related to Brave Ads advertising infrastructure appeared during idle startup even on a fresh install, where the user had not enabled Brave Rewards or opted into advertising.
• Those connections were described as limited and not transmitting identifiable behavioral data like classic ads, but the presence of “ad-adjacent servers” contradicted the expectation for users trying to escape advertising infrastructure.

Overall take on Brave

• Performed better than most in the comparison, but marketing and real idle behavior diverge due to Brave Ads infrastructure showing up by default/early.

Pros

• Strong tracker/fingerprinting blocking reputation (per benchmark).
• Safe browsing database handled via Brave infrastructure (vs Google default).

Cons

• Ad-network infrastructure connections present even without opting in.

---

2) Firefox (privacy strengths require configuration; default idle behavior includes outside services/telemetry)

Reputation/context mentioned

• Longstanding “trustworthy alternative” narrative (non-Chromium, open source, nonprofit Mozilla, auditable code).

What the network capture showed (idle startup)

1. Google safe browsing lookups to Google infrastructure—even though the user chose Firefox to avoid Google.
2. Mozilla telemetry to Mozilla’s own infrastructure enabled by default, including session initialization info / hardware details and online/active pings (anonymized per Mozilla).
3. Pocket (read-later) integration: Firefox contacted Pocket infrastructure and checked recommendations during idle, even though the user never used Pocket or logged into it.

Important complication / fairness toward Firefox

• In 2026 Firefox has significant privacy features:
  • Total Cookie Protection (third-party cookies partitioned).
  • Enhanced Tracking Protection blocks known trackers by default.
  • Passes most state partitioning tests on privacytests.org.
  • Supports uBlock Origin Manifest V2 fully (Firefox uniquely supports original MV2 architecture; Chrome/Chromium forced toward MV3).
• The reviewer argues: Firefox can be substantially more private when fully configured, but the default produces more external connections than privacy-seeking users expect.

Bottom line

• Default behavior creates a mismatch between “privacy reputation” and what actually happens without user hardening.

Pros

• Strong privacy features when configured (cookie protection, tracker blocking).
• Best-in-class extension support for Manifest V2 uBlock Origin.

Cons

• Default includes Google safe browsing, telemetry, and Pocket connections during idle.

---

3) DuckDuckGo Desktop Browser (brand promise is real, but Microsoft dependency undermines expectations)

What the network capture showed

• Connections to Microsoft Bing infrastructure during idle startup (before any user search/action).

Why this matters (per video)

• While DuckDuckGo admits it uses Bing, the reviewer notes Microsoft connections early/background are not what users expect from a privacy-first brand.

Audit/background mentioned

• A researcher audit (Zach Edwards) found DuckDuckGo allowed Microsoft tracking scripts on third-party sites while blocking equivalent scripts from Google and Facebook.
• DuckDuckGo CEO Gabriel Weinberg reportedly said contractual terms prevented blocking some Microsoft tracking infrastructure.
• Some expansion of tracker blocking occurred after backlash (Aug 2022), but later reporting suggests some Microsoft conversion scripts remain permitted due to carve-outs.

Overall take

• DuckDuckGo’s protections are called meaningful but partial, and users should understand that privacy is not comprehensive.

Pros

• Blocks a significant portion of trackers on sites visited.
• Includes:
  • “Fire button” for one-click data clearing.
  • Global Privacy Control by default.

Cons

• Microsoft tracking/infrastructure dependencies appear (including background/idle), with contractual carve-outs.

---

4) Opera (worst mismatch; idle fingerprinting + parent-company context + VPN marketing issue)

What Opera markets

• Built-in VPN, ad blocker, tracking protection, cookie consent management.

What the network capture showed (very concerning)

• Within first ~60 seconds of idle startup, Opera established connections to advertising measurement / behavioral analytics infrastructure.
• Metadata included items that can contribute to fingerprinting, such as:
  • session initialization telemetry
  • hardware/fingerprint signals (screen resolution, GPU config, OS version, timezone, language, fonts list)
• The reviewer argues this is browser fingerprinting in a “purest” form during idle startup.

Corporate/ownership context raised

• Opera acquired in 2016 by a Chinese consortium led by Kunlun Tech.
• The video claims prior reporting linked the parent company to data-collecting practices and regulatory issues; the reviewer avoids asserting unverified specifics about data flows, but says the network capture shows connections to infrastructure associated with the parent company.

Opera VPN critique

• Opera VPN is framed as a proxy using Opera’s own servers, meaning browsing traffic is routed through Opera infrastructure (subject to Opera logging/data handling), not truly “no-logs audited” like dedicated privacy VPNs (e.g., Mullvad/Proton).

Overall verdict

• “Worst performer by a meaningful margin.”
• Recommendation: if using Opera expecting privacy marketing to be accurate, switch away.

Pros

• On-paper features (ad blocker, tracking protection, VPN) are marketed strongly.

Cons

• Idle startup shows early connections to analytics/fingerprinting-related signals.
• Parent-company infrastructure connections not meaningfully addressed in marketing.
• VPN is not equivalent to independent privacy/VPN models; routed through Opera.

---

5) LibreWolf (only one that matches marketing/claims in the network capture)

Positioning

• Community-maintained Firefox-based privacy-focused browser without common integrations/telemetry defaults.
• Goal: “take Firefox, remove defaults that compromise privacy, ship the result.”

What the network capture showed

• Only update checks to LibreWolf’s own servers.
• No Google infrastructure connections, no Mozilla telemetry pings, no ad-adjacent server connections, no third-party content recommendation checks, and no fingerprinting signals to analytics.
• Reviewer emphasizes repeatability (multiple runs/machines gave the same result).
• Behavior aligned with marketing.

Noted trade-offs

• Stricter defaults can break sites (JS-heavy sites may fail until adjustments).
• Smaller user base than Firefox/Brave → less troubleshooting/community support.
• Updates follow Firefox release schedule → possible delay window for security updates.
• No built-in sync for bookmarks/history across devices.
• Installation on Windows slightly more involved (from project website, not Microsoft Store).

Overall verdict

• LibreWolf is presented as the only browser in the test that “survives the marketing vs network capture criterion.”

Pros

• Idle network behavior matches privacy claims “line by line.”
• Defaults disable telemetry, remove Pocket integration, avoid Google safe browsing routing, enable resist fingerprinting, block third-party cookies, reduce WebRTC leaks by default.

Cons

• Website compatibility issues unless you tune settings.
• Smaller ecosystem and sync limitations.
• Possible lag tied to Firefox release cadence.

---

Comparisons made

The reviewer repeatedly compares whether each browser’s idle network traffic supports or contradicts its privacy marketing:

• Brave: good privacy reputation, but ad network connections appear anyway.
• Firefox: strong privacy when hardened, but defaults include Google safe browsing, telemetry, Pocket.
• DuckDuckGo: blocks many trackers, but Microsoft infrastructure/tracking carve-outs appear even without user action.
• Opera: marketed privacy features don’t match network behavior; idle fingerprinting + analytics + “VPN” proxy model.
• LibreWolf: the only one where marketing and network behavior match.

---

Main pros/cons summary across the test

• Most aligned with marketing: LibreWolf
• Best baseline privacy reputation but with caveats: Brave (advertising infrastructure connections despite “escape ads” intent)
• Good after configuration, but default mismatch: Firefox (telemetry/Google/Pocket in idle)
• Partial privacy due to dependencies: DuckDuckGo (Microsoft-related background tracking infrastructure)
• Largest mismatch / worst behavior: Opera (idle fingerprinting + analytics + parent-company routing + proxy “VPN”)

---

Ratings / numerical scores mentioned

• Brave privacytests.org ranking: 143 out of 156 (for blocking tracking and fingerprinting).

(No other explicit numeric scores are given for the other browsers.)

---

Overall recommendation / concise verdict

• Recommendation: If you want a browser whose idle network behavior matches its privacy claims, the video’s strongest recommendation is LibreWolf.
• If you must choose among the others:
  • Brave is relatively better than most but requires disabling Brave Rewards to reduce ad-infrastructure connections and enabling aggressive shields.
  • Firefox is much better when you disable telemetry, turn off Pocket, and avoid/adjust Google safe browsing and consider resist fingerprinting—otherwise defaults contradict “no phone-home” expectations.
  • DuckDuckGo and Opera are presented as having structural marketing gaps; the reviewer leans toward switching away, especially for Opera given the idle fingerprinting/analytics findings.

---

Speaker breakdown

• One main speaker (the reviewer/narrator) drives almost all claims, methodology, comparisons, and recommendations.
• Referenced third parties (not co-speakers):
  • privacytests.org (benchmark ranking for Brave)
  • Zach Edwards (DuckDuckGo audit)
  • Mozilla documentation / privacy policy (telemetry and Pocket disclosures referenced)
  • TechCrunch / 9to5Mac (follow-up reporting referenced)
  • TechRadar / PCMag (Opera VPN auditing/review statements referenced)

(No multiple on-camera speakers are present; it’s effectively a single reviewer voice with external sources cited.)