AI is super-charging social engineering, and K-12 is still a precious target. With an average of 2,739 edtech tools per district, staff and students rely heavily on laptops and classroom tech that must be protected from the latest threats. Today, these include anything from convincing “superintendent” emails to deepfake voice notes and student-account takeovers. 2,739 edtech tools 2,739 edtech tools PromptLock is one example of a new kind of computer virus that uses generative tools to help write its own harmful code every time it runs. That means it can change slightly each time, making it harder for security systems to catch. PromptLock PromptLock Once it’s on a computer, the malware looks through the files. It can then steal them and lock them up so schools can’t open them. As ransomware becomes more sophisticated, attacks could target not just large schools but also individual students and staff members, leaving them open to higher risks of data theft, financial loss, and service disruptions. Schools must know where their blind spots are and how to protect themselves against these types of cyber attacks. Find and fix blind spots in built-in filters Find and fix blind spots in built-in filters Built-in tools often miss AI-powered lures, because the latest generative AI tools can write polished messages that sound human. In a recent survey of 18,000 employed adults, only 46% correctly identified that a phishing email was written by AI. For traditional security systems, it’s equally difficult. When there are no spelling errors or awkward phrases, filters that look for “typical scam language” struggle to flag them. only 46% correctly identified only 46% correctly identified Part of the problem is that AI can pull details from public websites or social media, and mention upcoming school events and staff names, making them sound authentic. Even when an email doesn’t contain malware, it can trick someone into sharing passwords or sensitive data. That means IT administrators must introduce filters that understand context. Once security teams realize an account has been compromised, they can flag the content and account as a warning to the rest of the school and update their security systems. But since AI can generate a slightly different version of the same phishing message for each target, it’s tricky to tell traditional security systems what patterns or “signatures” to look for. Tools that rely on rules and known threat lists, not real-time reasoning, no longer suffice. To tighten defenses, districts should audit their native filters quarterly. They must test defenses with realistic phishing simulations that represent today’s standard of attack, and adjust rules to flag messages containing urgency, payment requests, or login prompts. Advanced phishing detection tools and add-ons can help security teams flag messages that “feel off,” even if they look clean. Build a zero-trust defense plan Build a zero-trust defense plan Hackers are taking over staff and student accounts and sending phishing emails that impersonate school members. Microsoft reports that from just 11 compromised accounts at three universities, Storm-2657 sent phishing emails to nearly 6,000 email addresses at 25 institutions. Since many phishing emails now come from compromised legitimate accounts, built-in tools can no longer assume that messages from them are safe. just 11 compromised accounts just 11 compromised accounts Zero-trust policies, where schools trust no one automatically, are essential. Every login, device, and app connection should be verified. Schools must also monitor login patterns, device activity, and unusual sharing behavior in cloud apps like Google Drive or Microsoft 365. By building alerts for unusual internal activity, such as a teacher’s account suddenly sending dozens of messages after hours, IT admin teams can strengthen defenses. No single tool can catch everything, but together, they reduce the risk dramatically. Schools should enforce multifactor authentication (MFA) on all accounts, monitor cloud activity for unusual file sharing, and track sign-ins from unfamiliar devices. That way, even if an attacker bypasses initial defenses, unusual account behaviors are quickly detected and contained. Since there are so many platforms to manage to keep school digital property safe, false positives can slow down the time to detection. Recent findings from 500 cybersecurity respondents found that only 29% are able to investigate more than 90% of their cloud security alerts within 24 hours. When the fastest recorded attack was just 51 seconds from initial engagement to compromise, security experts really have no time to waste. only 29% are able to investigate only 29% are able to investigate 51 seconds 51 seconds Schools can consider investing in mailbox intelligence that uses AI to help determine whether or not a message is impersonating a user. By building automated steps for quarantining suspicious messages, resetting credentials, and notifying affected users, schools can minimize the time between detection and containment. Train every user like a security partner Train every user like a security partner Technology alone can’t stop every phishing attempt, especially as AI makes scams more convincing and personalized. Even the best-rated anti-phishing tools missed up to 15 percent of attacks in AV-Comparatives’ 2025 certification test. Firewalls, filters, and message quarantining are essential, but they can’t always catch messages that look legitimate or come from trusted accounts. That’s why it’s equally important to train staff and students how to recognize suspicious messages and feel confident reporting them. missed up to 15 percent of attacks missed up to 15 percent of attacks Effective training now looks nothing like the old “don’t click” slideshow. Districts in Ohio and elsewhere are running monthly simulations, sending fake phishing messages to see who spots them and who needs coaching. This approach normalizes reporting and keeps awareness fresh. Ohio Ohio Training should also reflect each role’s risks. Staff who handle finances need to recognize fake invoices or urgent transfer requests. IT teams must know the signs of account takeover, MFA fatigue, and AI-generated help-desk impersonations. Students should learn to verify links and spot too-good-to-be-true offers. Short, recurring lessons work best. Replace annual seminars with quick micro-courses that teach people to pause, question, and verify. Track progress through reporting rates, not just attendance, and celebrate catches as a win for the whole district. A practical action plan going into 2026 must include: Frequent audits and adaptation: Run phishing simulations every semester and review which accounts or tools failed.
Automate response management: Use AI-based mailbox intelligence to isolate suspicious messages and reset affected credentials.
Teach critical thinking: Move from memorized rules to realistic phishing attack scenarios that train instinct and judgment. Frequent audits and adaptation: Run phishing simulations every semester and review which accounts or tools failed. Automate response management: Use AI-based mailbox intelligence to isolate suspicious messages and reset affected credentials. Teach critical thinking: Move from memorized rules to realistic phishing attack scenarios that train instinct and judgment. With education now overtaking healthcare as scammers’ top target, schools can’t afford shortcuts in cyber defense. The path forward combines smarter technology, disciplined verification, and a community that understands its role in security. When districts pair AI-powered detection with human skepticism, they shorten the gap between first click and first report—the window that decides whether a phishing attempt becomes tomorrow’s headline.

The is an opinion piece based on the author’s POV and does not necessarily reflect the views of HackerNoon.

Every

Microsoft

How AI Phishing Is Putting School Districts at Risk

About Author

Comments

TOPICS

THIS ARTICLE WAS FEATURED IN

Related Stories

How Can Schools Be Better Prepared for Supply Chain Attacks?

Argon2 in Practice: How to Implement Secure Password Hashing in Your Application

OpenAI Aardvark and The New Wave of Autonomous Security Research

Why kube-prometheus-stack Isn’t Enough for Kubernetes Observability

The Noonification: How Often Do NFTs Pass The Howey Test? (1/13/2023)

Darwin's Hybrid Intelligence to Align AI & Human Goals for Startups & VCs

How Can Schools Be Better Prepared for Supply Chain Attacks?

Argon2 in Practice: How to Implement Secure Password Hashing in Your Application

OpenAI Aardvark and The New Wave of Autonomous Security Research

Why kube-prometheus-stack Isn’t Enough for Kubernetes Observability

The Noonification: How Often Do NFTs Pass The Howey Test? (1/13/2023)

Darwin's Hybrid Intelligence to Align AI & Human Goals for Startups & VCs

Light-Mode

Classic

Newspaper

Minty

Dark-Mode

Neon Noir

Minty

HN StartUps