Interview with nine engineers who worked on Google's Safe Browsing, which now underpins user security in all of Google's major platforms and runs on 3B+ devices
IN THE BEGINNING there was phone phreaking and worms. Then came spam and pop ups. And none of it was good. Tweets: @k8em0 , @google_comms , @mediaphyter , and @selenakyle . Thanks: @aaronjstein Tweets: Katie Moussouris / @k8em0 : “We're a little too quiet about it sometimes.” Awesome work by all - & I'm loudly going to echo that @selenakyle is a badass! BofA is incredibly lucky to have her on board now, & I hope to see more of her and others' quiet infrastructure engineering & architecture work showcased. http://twitter.com/... Google Communications / @google_comms : “What began as a shot in the dark would go on to fundamentally change security on the internet.” The story behind safe browsing. http://twitter.com/... Jennifer Leggio / @mediaphyter : From @Wired: “How Google's Safe Browsing Helped Build a More Secure Web” Includes interviews with nine engineers who are game-changers in safe browsing. Very happy to see the brilliant @selenakyle included in this feature. She's done phenomenal work. http://www.wired.com/... @selenakyle : super cool roundtable: 10 years of google safe browsing. helping protect the web from behind the scenes of chrome, search, ads, play store, Gmail, & more. Go SB, go! http://twitter.com/... Thanks: @aaronjstein
Context & Ripple Effects
This Wired oral history is a retrospective on how Google turned threat response into permanent infrastructure: what began with phone phreaking-era defenses became Safe Browsing, now running on more than three billion devices across Google's major platforms. It sits alongside earlier reporting on Google's defensive posture, including how the company brought KrebsOnSecurity into Project Shield to absorb massive DDoS attacks.
The piece also lands mid-arc on Google's internal tensions: the same quarter saw Chrome ship its revamped omnibox and password manager in Chrome 69, putting Safe Browsing-derived protections directly into the browser UI, while employees were already describing the company's struggle to balance user privacy against business interests.
First-order effects
- The nine engineers get public credit for work that Katie Moussouris notes has been 'a little too quiet' — rare visibility for infrastructure teams whose output is normally measured only by attacks that fail.
- Chrome users see the payoff directly: the browser's password manager and omnibox, refreshed in Chrome 69, are consumer-facing surfaces for the same security stack this interview documents.
Second-order effects
- With Safe Browsing on 3B+ devices, Google functions as the de facto trust layer for the open web — meaning rival browsers and site owners effectively outsource phishing and malware verdicts to a competitor, a dependency few can unwind.
- Dedicated research units like Maddie Stone's Project Zero team neutering actively exploited Android malware extend the same model from blocking bad sites to hunting live exploits, raising the baseline attackers must clear across Android.
Third-order effects
- If the pattern holds, web security consolidates around a handful of platform vendors operating shared blacklists and exploit research as invisible utilities — with Google's own privacy-versus-business friction, documented by current and former employees, becoming the key governance question for infrastructure everyone else relies on.
- The Privacy Sandbox effort to replace third-party cookies suggests the next phase: the same trust-and-safety apparatus expanding from malware defense into defining what tracking and data collection the web permits at all.
The trend: Platform security is shifting from per-product feature to shared, invisible infrastructure operated by a few large vendors — making their internal priorities a matter of web-wide consequence.