This Month in Data Leaks: Google’s Breach, Spilled Tea, and Other August Highlights (Plus: How to Avoid Them)

It's been quite the summer.

From CRM leaks to airline passenger data turbulence, this past month delivered a steady stream of reminders that cybersecurity is still very much a team sport — and sometimes, one with surprising fumbles.

Whether the issue was a misconfigured third-party platform or an overly casual approach to app development, the common thread is clear: access control and visibility matter more than ever.

Here’s a breakdown of what went wrong, what it means for your business, and how to avoid starring in next month’s breach roundup.

🚨 Google’s Salesforce Database got hit by ShinyHunters

What happened: Hackers breached a Salesforce CRM database used by Google, gaining access to internal sales data tied to small and medium businesses. This includes contact info, deal notes, and a fresh reminder that storing everything in one place can go very wrong.

The cause: A third-party platform (Salesforce) with access to sensitive internal data. Likely no network segmentation or Zero Trust guardrails around that data.

Your move:

  • Apply Zero Trust access controls to SaaS platforms.
  • Isolate sensitive systems with private routing.
  • Audit third-party integrations — and monitor them like you would your own infrastructure.

🫣 Tea App spills... everything

What happened: The “Tea” app — where users anonymously review men — accidentally leaked driver’s licenses, selfies, and private messages. That data quickly showed up on 4chan and X. So much for anonymity.

The cause: A poorly secured storage bucket. Possibly rushed development (“vibe coding,” in the app’s own words). Definitely a lack of secure-by-design thinking.

Your move:

  • Avoid the “ship fast, fix security later” mindset.
  • Store PII behind encrypted, access-controlled systems.
  • Vet vendors and tools — especially AI-powered apps that move fast and break trust.

🧑‍💻 PBS leaks internal employee info — via Discord

What happened: Personal info for nearly 4,000 PBS employees was leaked on a Discord server tied to “PBS Kids.” Names, job titles, emails, departments, and supervisors — all now in the wild.

The cause: Informal comms platforms + sensitive info = bad combo. Likely no DLP or monitoring for internal data exposure.

Your move:

  • Don’t treat Discord (or Slack, or Teams) as a secure vault.
  • Lock down internal data with access policies that follow people, not just devices.
  • Monitor traffic for accidental exfiltration — it’s not always malicious.

✈️ Air France & KLM customer data takes off

What happened: A breach in a third-party customer service platform exposed names, emails, contact details, and frequent flyer data for 6 million Air France and KLM passengers. No passwords or credit card info, but still a heavy hit to customer trust.

The cause: A vulnerable vendor. Third-party integration without proper segmentation or risk modeling.

Your move:

  • Treat third-party platforms as external — because they are.
  • Use layered security: private access, device posture checks, and identity-based policies.
  • Don’t assume your vendors are doing it right.

🎥 Venice Film Festival gets hacked — stars (and staff) affected

What happened: Participant and journalist data from the Venice Film Festival was leaked. Not the worst breach on this list, but still: red carpet, meet red flags.

The cause: No official details, but the takeaway is clear — any organization handling personal data is a target.

Your move:

  • Build privacy-first processes, even if you’re not a traditional “tech company.”
  • Publish a breach response plan before you need it.
  • Segment access to personal data with context-aware rules.

🧩 What these breaches have in common

  • Over-trusted systems and platforms
  • Lack of visibility into third-party risks
  • Weak or missing segmentation and access controls

Sound familiar? These are the exact problems Zero Trust was built to solve — and the kind of thing OpenVPN users can actively guard against using our tools.

🛡️ How OpenVPN helps you avoid being on this list

  • Secure remote access that keeps sensitive systems off the public internet.
  • CloudConnexa® can enforce identity, location, and device-based access controls.
  • Private routing and segmentation so even if one app gets hit, the rest don’t go down with it.
  • Easy integrations with the platforms you already use — but with the security those platforms sometimes forget to include.

TL;DR

You don’t need a massive budget or an AI-driven threat detection machine to avoid ending up in a breach roundup. You just need:

 

  • Strong access controls
  • Isolation of critical systems
  • A healthy suspicion of your vendors
  • And ideally, a VPN solution that knows how to evolve with the threat landscape

 

We happen to know a good one.

Get Started