For years, we told clients the same thing: when an email looks off, trust your gut. Awkward grammar, strange phrasing, a greeting that read like it came from a translation app — those were the tells. If a "bank" emailed you with three spelling mistakes and a sense of panic, you could feel pretty confident it was a scam.
That advice is getting weaker by the month. The new generation of AI writing tools, the kind that exploded onto the scene late last year, can produce clean, fluent, professional-sounding text in seconds. Anyone can use them, including the people trying to trick your staff into wiring money or handing over a password.
The old warning signs are disappearing
Most phishing advice still leans on spotting bad writing. That made sense when many scams were sent by people working in a second or third language, copying templates and hoping volume would do the rest. The mistakes were the giveaway.
Now a scammer can paste a rough draft into an AI tool and ask it to rewrite the message so it sounds like a polished note from your CEO, your bank, or a vendor you actually use. The result reads correctly. It uses the right tone. It can even match the style of a real company's emails.
Here is what that means in practice:
- Grammar and spelling no longer prove anything. A flawless email is not automatically a safe email.
- Personalization is cheap now. AI can help attackers tailor messages using public details from your website or LinkedIn.
- Volume and quality went up together. Scammers can produce more convincing messages, faster, than before.
The threat itself has not changed. Phishing still tries to get someone to click a bad link, open a bad file, or move money. What changed is how believable the bait looks.
What still gives a scam away
Better writing does not fix the underlying logic of these attacks. Train your team to watch the behavior of the message, not just the spelling.
Urgency and pressure
"Do this now or the account closes." "The wire has to go out before 5." Scams lean on time pressure so you act before you think. A real colleague or vendor will almost always tolerate a five-minute pause to verify.
A request that breaks normal process
Your CFO does not usually ask for gift cards. Your bank does not email asking you to confirm your full password. When a message asks you to skip the usual steps, that is the red flag — not the grammar.
Links and sender addresses that do not match
Hover over a link before clicking and look at where it actually goes. Check the real sending address, not just the display name. "Microsoft Support" can sit on top of an address that has nothing to do with Microsoft.
Verify through a second channel
This is the single most useful habit your business can build, and it works no matter how good the fake looks. If an email asks for money, credentials, or a change to payment details, confirm it through a different channel than the one the message arrived on.
Got an urgent email from a vendor asking you to update their bank account? Call them on the number you already have on file — not the number in the email. Got a request from the owner to buy gift cards? Walk over or text the number you know is theirs. The scam falls apart the moment you check it against reality.
Layer your defenses
No single tool catches everything, so the goal is to stack protections that each catch what the others miss.
- Multi-factor authentication (MFA). If a password does get phished, MFA gives the attacker a second locked door. Turn it on everywhere it is offered, especially email and banking.
- Email filtering. Good filtering blocks a large share of malicious mail before anyone sees it. It will not be perfect, which is why the other layers matter.
- Ongoing staff training. Short, regular refreshers beat a once-a-year lecture. People are your last line of defense, and that line gets stronger with practice.
- A clear way to report. Make it easy and blame-free for staff to flag a suspicious message. The faster you hear about it, the faster you can warn everyone else.
These pieces work together. MFA covers a slip-up. Training reduces the slip-ups. Filtering cuts the volume your team has to judge in the first place. We build this kind of layered protection into our cybersecurity services, and we tie it into the day-to-day support that keeps a Tampa Bay business running.
The bottom line for local businesses
The tools attackers use got an upgrade, so the way we spot them has to keep up. Stop relying on bad grammar as your warning system. Teach your team to question the request, slow down under pressure, and verify anything involving money or passwords through a channel they already trust.
If you would like a second set of eyes on how your business handles email security and staff training, our team here in Pinellas and Pasco County is glad to help. You can reach us through our managed IT services page or get in touch directly to talk it through.
