Can AI Mimic a Friend's Writing Style in a Crypto Scam Message?
AI can mimic a compromised friend's writing style, removing an informal signal you'd normally rely on. Here's why independent verification still matters.
Recognizing a genuine message from a friend partly relies on it sounding like them — their particular phrasing, typical abbreviations, characteristic tone. AI text generation trained on someone's available writing samples can now approximate these specific stylistic patterns.
Why Writing Style Has Traditionally Functioned as an Informal Authenticity Signal
Beyond the actual content of a message, the way someone writes — word choices, typical phrasing, characteristic tone — has traditionally provided an informal, often subconscious signal that a message genuinely came from that specific person, distinct from evaluating the message's content alone.
Why This Specifically Matters When an Account May Be Compromised
If a friend or family member's messaging account or email has been compromised, a scammer gaining that access could potentially use AI to generate messages mimicking the genuine account owner's actual writing style, based on their accessible message history, rather than writing in an obviously different, mismatched voice.
How AI Could Theoretically Learn and Replicate Specific Stylistic Patterns
Given access to a sufficient sample of someone's actual writing — accessible through a compromised email or messaging account — AI text generation could potentially identify and replicate characteristic patterns: typical greeting style, common phrases, characteristic punctuation or abbreviation habits.
Why This Adds a Specific Layer of Deception to Compromised Account Scams
A message from a compromised account requesting money or crypto that also convincingly matches the genuine account owner's writing style removes one of the more intuitive, informal signals recipients might otherwise rely on to notice something is wrong, beyond the account itself appearing genuine.
Why This Doesn't Change the Underlying Request Pattern to Watch For
Regardless of how convincingly a message mimics a friend's writing style, an unexpected request for money or crypto, particularly with any urgency, remains the actual signal worth treating with caution — the message's stylistic authenticity doesn't confirm the account itself hasn't been compromised.
Why Verification Through an Entirely Separate Channel Remains the Reliable Check
Contacting the person directly through a different communication method — a phone call to a number you already have, rather than replying within the potentially compromised channel itself — provides verification independent of how convincing the message's writing style is.
Why a Quick, Specific Personal Question Can Help in Some Situations
Asking something only the genuine person would know — not from any publicly available information, but from shared private context — can help distinguish a genuine message from a compromised account, though this isn't foolproof if the compromising party has also accessed relevant private message history.
Why Any Request for Crypto Specifically Deserves Heightened Scrutiny Regardless of Source
Given that crypto transactions are irreversible once sent, any request specifically for crypto — as opposed to a general request that might otherwise seem plausible from a genuine contact — deserves particular scrutiny and independent verification before acting, regardless of how convincingly it appears to come from someone you know.
What to Check Before Sending Crypto Based on Any Message From a Known Contact
Verifying through a separate, independent communication channel before acting on any unexpected request for crypto, regardless of how convincingly the message matches the person's known writing style or account.
Verify any unexpected crypto request from a known contact through a separate channel — AI can now approximate someone's writing style if their account is compromised, removing this informal authenticity signal.