Can AI generate believable phishing emails personalized to a target ?
Cast your vote — then read what our editor and the AI models found.
AI systems can already craft phishing emails that mimic a target’s writing style and pull in personal details—practical demonstrations show how language models synthesize public data into messages that pass initial scrutiny. The court heard unanimous testimony that today’s models can generate individualized lures with a level of persuasion that hand-written attempts rarely achieve.
Background
Recent public tests show that large language models can tailor phishing emails by extracting names, employers, writing habits and social connections from public posts to produce messages that read as authentic to the recipient. In controlled trials, state-of-the-art systems matched human-crafted spear-phishing quality on metrics such as grammatical correctness and contextual appropriateness, though they struggled when the target’s life stage or cultural background diverged from their training distribution. Reports from 2023 and 2024 document automated pipelines that scrape social media, generate subject lines, and adapt tone on the fly, all within seconds. Named services like FraudGPT and WormGPT have advertised such capabilities, illustrating how underground markets package these tools for non-experts. Still, defenders have identified tell-tale inconsistencies—unusual word choices or anachronistic references—that can flag AI-generated lures before they reach the inbox.
SOURCE: Nature, 2024
Suggest a tag
A missing concept on this topic? Suggest it and admin reviews.
Status last checked on June 26, 2026.
Gallery
Can AI generate believable phishing emails personalized to a target?
The jury found a clear answer in the affirmative.
The jury swiftly delivered a unanimous verdict, finding that current AI systems are capable of crafting believable, personalized phishing emails with unsettling precision. The reasoning hinged on the technology’s proven ability to mimic tone, style, and context—making deception nearly indistinguishable to the average recipient. With no dissent, the bench ruled in favor of the affirmative. The ruling: "The reel is loaded; now the fish just have to bite.
But the data is real.
The Case File
Across 11 sessions, 32 jurors have heard this case. Combined tally: 32 YES · 0 ALMOST · 0 NO · 0 IN RESEARCH.
Note: cumulative includes older juror opinions. The current session tally above is the live verdict.
By a vote of 2 — 0 — 0, the panel returns a verdict of YES, with verdict confidence of 93%. The court so orders.
"AI can generate coherent text"
"LLMs excel at generating highly personalized text tailored to individual targets."
What the audience thinks
No 8% · Yes 77% · Maybe 15% 66 votesDiscussion
no comments⚖ 11 jury checks · most recent 2 days ago
Each row is a separate jury check. Jurors are AI models (identities kept neutral on purpose). Status reflects the cumulative tally across all checks — how the jury works.