Stating that data breaches are becoming more frequent and more costly isn’t an exaggeration. Even so, there’s a lingering misconception about them.
People’s first instinct is to link data breaches to sophisticated, orchestrated attacks and “hacker masterminds” capable of bypassing the most advanced cybersecurity systems.
While certainly not unheard of, such cases account for only a minority of breaches today. If anything, the reasons breaches continue to be so impactful are mundane and have more to do with simple, preventable blunders that snowball into major incidents.
Here’s what you should actually be wary of to combat the risks effectively.
Outdated Software
Neglecting updates is the classic yet still relevant oversight that comes to mind first. Keeping operating systems across different device types, software, and dependencies up to date is challenging.
This is especially problematic for larger companies with vast, interconnected systems, parts of which might have become too outdated to maintain.
Attackers who focus on this shortcoming don’t need to be particularly skilled. They can just wait for an unpatched vulnerability to become documented and use that as an entry point.
The more such vulnerabilities exist and the longer they aren’t addressed, the easier this becomes.
Exposed Credentials
Why even bother with sophisticated breaches when entry into sensitive systems can be served on a platter? Exposed credentials are a serious issue because they offer immediate, potentially high-level unauthorized access.
Stopping the threat completely is also tricky since there are so many avenues for exposure.
A programmer might accidentally forget to remove API keys or login credentials from code they share on public repositories. Similarly, employees may share sensitive data that anyone with the right link can access instead of sending timed links to specific recipients.
Unsafe storage and sharing are also a problem. Employees might keep copies of sensitive data on personal devices that lack appropriate safeguards. Likewise, they may share it through unsanctioned email or apps.
Even if your company’s cyber defense is otherwise strong, a breach could still happen if those third parties get compromised.
Irresponsible AI Usage
Next is the misuse of artificial intelligence, either through employee carelessness or unsecured AI systems. This is the newest, but also the fastest-growing, and one of the most impactful data breach attack vectors.
Companies that rush to leverage the latest tools without LLM observability risk exposing data on multiple fronts. Employees interact with AI systems to speed up their everyday tasks.
Doing so, they may share information that could allow unauthorized access if stolen. The AI tool that processes this information may add it to its training data or expose it to third-party tools it synergizes with.
AI tools don’t need to interact with humans to become threats if they’re misconfigured. For example, giving broad permissions to ticketing systems, analytics platforms, or automation tools may let them access internal systems and files that contain sensitive data.
Keep in mind that both of these are examples of threats when using sanctioned AI tools. The risks grow exponentially if employees take adoption and implementation into their own hands.
This shadow AI operates beyond IT’s oversight and can easily lead to breaches since the AI’s developers may not be up to the company’s high security standards.
Poor Employee Training
It’s clear at this point that not enough attention is given to matters of cybersecurity training since employee actions enable or exacerbate the other reasons we’ve talked about.
Employees in non-IT roles who lack basic cybersecurity sense have a greater chance of falling victim to phishing and social engineering attacks. Not being aware of the risks also makes them more likely to practice shadow IT and AI or ignore guidelines.
However, colleagues who should know better aren’t blameless either. Hardcoding login credentials or API keys into config files and ignoring security warnings happens far too frequently.
Lack of Third-Party Oversight
Since SaaS is now the norm, doing business without live third-party integrations is either impractical or outright impossible.
Assuming that just because a vendor contributes to a specialized part of your daily operations, they automatically do so responsibly is when security issues arise.
Blind adoption is a well-known yet neglected risk factor. Companies focus on the services third parties provide without assessing their cybersecurity readiness and compliance or examining past incidents.
Couple that with lax permission policies and a lack of vendor monitoring, and having to deal with fallout you’re not technically responsible for becomes a distinct possibility.
