ThreatsDay Roundup: Chrome's 370 Flaws, AI Phishing, and the "Wrong Screen" Problem

Saw the ThreatsDay report today and the line "A lot of security still comes down to trusting the wrong screen" really hit home. Between the 370 Chrome flaws and the SonicWall issues, the attack surface is getting noisy, but the vector remains the same: social engineering enhanced by tech.

The volume of Chrome vulnerabilities (CVE-2026-XXXX batch) is staggering. Even if 90% are "Use After Free" with low severity, the patch fatigue creates windows of opportunity. When you combine unpatched browsers with AI-generated phishing sites (the "wrong screen" mentioned), you get a nasty combo. AI isn't just writing code; it's writing perfect, localized HTML for credential harvesting that bypasses traditional visual heuristics.

For those tracking the DNS hijacking angle mentioned in the report, check your resolver logs against known malicious sinkholes. Here is a quick Python snippet to sanity-check your current Chrome version against the latest stable release (for automated inventory):

import requests
import 

def get_chrome_version():
    url = "https://versionhistory.googleapis.com/v1/chrome/platforms/win/channels/stable/versions"
    r = requests.get(url)
    data = r.()
    # Returns the most recent version entry
    return data['versions'][0]['version']

print(f"Latest Stable: {get_chrome_version()}")

With AI lowering the bar for creating these fake login screens, how is everyone handling the "trust" aspect of security? Are you moving towards hardware keys (FIDO2) exclusively, or is the latency still too high for your org?