Which browser should you use right now?
Based on Theo - t3․gg's video on YouTube. If you like this content, support the original creators by watching, liking and subscribing to their content.
Chrome’s standards push and Chromium’s open-source base explain why it remains the most compatible “default” even when alternatives exist.
Briefing
Browser choice right now comes down to a simple tradeoff: most “alternative” browsers either inherit Chromium’s strengths without meaningfully improving day-to-day use, or they chase privacy/AI/experiments in ways that can degrade reliability and user experience. After walking through a crowded field of Chrome-based, Safari-based, and “agentic” AI browsers, the strongest recommendation lands on two open-source projects—Zen for calmer, better UX, and Helium for a more private, polished Chromium-like workflow.
The case for Chrome starts with standards and maintenance. Chrome’s push for web standards reduced the old era of site-specific hacks for Internet Explorer, Firefox, and Safari, and the Chromium base is open source—so many other browsers can build on it. The argument then pivots to why “Chrome monopoly” concerns don’t map neatly to engineering reality: maintaining Chrome is expensive, and Chromium users don’t generate direct revenue for Google. Instead, the incentive is framed as keeping the web platform healthy so more people use Google services.
Still, Chrome’s “rough edges” show up as AI integration. A Gemini button inside Chrome is described as a forced, low-utility implementation—more marketing than a feature people actually use. That same tension fuels broader anti-competitive drama, including Perplexity’s reported $34.5 billion offer to buy Chrome from Google, which is portrayed less as a strategic takeover and more as a bid for visibility and leverage in the AI search race.
From there, the transcript becomes a ranking of practical usability. Edge is treated as a Chromium wrapper with a sidebar and slightly more privacy, but also with AI features that don’t meaningfully improve the experience. Brave is the clearest negative: despite built-in ad blocking and anti-fingerprinting goals, it’s criticized for aggressive behavior, crypto-related prompts, and real-world UI/performance issues—especially for workflows involving capture systems and sidebar hotkeys/animation jank.
Vivaldi earns a qualified “fine” because it’s highly customizable and can be tuned into an Arc-like feel, but it’s also described as buggy under heavy customization and unable to reclaim the vertical real estate that Zen protects. Safari is acknowledged as battery-light and required on iOS (with other iOS browsers effectively using Safari/WebKit), yet it’s criticized for UI space loss and developer pain—race conditions and promise/IndexedDB issues that can make debugging feel impossible.
The transcript’s most consistent “why” is UX focus. Zen is presented as the rare browser trying to improve how people browse: a top-tier sidebar with responsive hover behavior, customizable hotkeys, splits, and better vertical real estate. It’s also framed as more private than Firefox after licensing changes, while still inheriting Firefox’s engine quirks (memory/battery and weaker dev tools). Zen’s DRM stance is practical: no Widevine approval, so DRM streaming may require a different browser.
Helium is positioned as the Chromium alternative that prioritizes privacy and polish without the “shilling” vibe. It’s described as mirroring extension downloads through its own infrastructure to avoid leaking extension-install metadata to Google, and it’s praised for shaving UI space to improve vertical real estate, built-in ad blocking, and a fast, stable feel. Its tradeoffs are straightforward: no mobile support yet, no tab sync, and limited customization compared with more configurable browsers.
Finally, the “agentic AI browser” wave is treated skeptically. Comet (Perplexity’s browser) is the most usable of the AI options because it can actually browse and execute tasks, but it comes with cringe onboarding and raises security concerns: hidden text on pages could trick agentic browsers into performing actions like sending data or funds. Other AI browsers are portrayed as mostly widgets that summarize page HTML rather than truly navigating the web.
By the end, the transcript’s browser strategy is clear: if someone wants a better web experience with fewer surprises, Zen and Helium are the best bets—while Chrome remains the default for standards and compatibility, and most other alternatives are judged by whether they improve real workflows or just change the interface while introducing new failure modes.
Cornell Notes
The transcript argues that browser “alternatives” succeed only when they improve real day-to-day UX without sacrificing reliability. Chrome’s dominance is attributed to standards leadership and the open-source Chromium base, which many browsers reuse; the main critique is forced AI surfacing that feels low-utility. Brave is heavily criticized for privacy features that allegedly worsen usability and for UI/performance problems that disrupt workflows. Zen is presented as the standout for user experience—especially its sidebar, hotkeys, and vertical real estate—while Helium is praised as a more private, polished Chromium-like option that mirrors extension installs to reduce Google metadata leakage. Agentic AI browsers like Comet can be useful when they can actually browse and act, but they also introduce security risks from hidden instructions in page content.
Why does Chrome remain the default even when “monopoly” concerns are raised?
What makes Zen stand out compared with other “Arc-like” attempts?
Why is Brave criticized so strongly despite having built-in ad blocking and anti-fingerprinting goals?
How does the transcript distinguish Safari’s constraints from its technical shortcomings?
What is the core privacy/polish pitch behind Helium?
What’s the security concern with agentic AI browsers like Comet?
Review Questions
- Which browser is credited with reducing cross-browser site chaos, and what mechanism makes that possible (standards work vs. open-source base)?
- What specific UX elements does Zen receive the most praise for, and how do they relate to vertical real estate?
- How does the transcript’s view of agentic AI browsers balance usefulness (task execution) against security risks (hidden instructions)?
Key Points
- 1
Chrome’s standards push and Chromium’s open-source base explain why it remains the most compatible “default” even when alternatives exist.
- 2
Chrome’s AI integration is criticized as forced and low-utility, with Gemini described as a poor implementation rather than a genuinely helpful feature.
- 3
Brave’s privacy approach is portrayed as harming usability through UI/performance issues and workflow breakage, outweighing its built-in ad blocking benefits.
- 4
Zen is recommended primarily for user experience: a top-tier sidebar, customizable hotkeys, responsive hover behavior, and strong vertical real estate protection.
- 5
Helium is recommended as a more private, polished Chromium-like browser that mirrors extension installs to reduce Google metadata exposure.
- 6
Safari is treated as constrained by iOS rules (WebKit requirement) and criticized for both screen-space use and developer debugging pain (race conditions, IndexedDB/promise issues).
- 7
Agentic AI browsers like Comet can be useful when they can actually browse and execute tasks, but they can also be exploited via hidden page instructions.