AI Take Holiday Scam Season to a Whole New Level

( – Experts are warning of new artificial intelligence (AI) technology capable of stealing through so-called “phishing scams.” According to the Federal Trade Commission, phishing scams are tricks that criminals use to persuade their victims to hand over personal or financial information they can use to steal from them.

“Scammers use email or text messages to try to steal your passwords, account numbers, or Social Security numbers,” the agency says. The criminals will then use that data to access bank accounts or sell it to fellow scammers. Thousands of such schemes are launched every day and are often disguised as text messages or emails from trusted sources such as a bank or government agency.

During the holiday season, scams increase and thieves stole $73 million last year. However, with the advent of AI technology, experts say scams have become more sophisticated, and phishing increased by 1,265% in just one year, partly because of that technological advance.

While criminals used to write phishing emails and text messages themselves, now they are employing AI to write them, and the result is that they look more professional and are believed to be more convincing. One such AI tool is ChatGPT and technology expert David Raissipour has warned, “Tools such as ChatGPT is that they can create better spear-phishing content with a much more relevant reference to things that are very personal to the intended targets.”

In response to the increased threat, tech giants like Google are producing software that can fight back. Google has released a product called Resilient & Efficient Text Vectorizer (RETVec). This trains spam filters to spot increasingly sophisticated scam characters, including those that appear identical, such as the letter O and a zero. It also spots invisible characters and so-called keyword stuffing – hidden text that makes emails or texts appear legitimate.

Google says it has “battle-tested RETVec extensively inside Google to evaluate its usefulness and found it to be highly effective.”

Copyright 2023,