Security researchers have found a new phishing attack using Japanese hiragana character "?" to create extremely convincing URLs that will deceive even the most prudent users. The attack, which has been found by security researcher JAMESWT, represents a new direction for homograph attacks using Unicode. Users of popular holiday reservation website Booking.com are among the particular target of the attack, and the URLs appear exactly the same as authentic ones. Attackers substitute the Japanese character "?" (Unicode U+3093) for the forward slash ("/"). The altered URLs look genuine when they are displayed in certain browsers, thus making it possible for attackers to trick victims into visiting fake sites that can propagate malware such as remote access trojans and stealers of information. The attack's sophistication and social engineering nature make it extremely risky. This is unlike usual phishing where the attacker usually uses badly formatted or easily identifiable misspelled emails. This attack uses the visual similarity of Unicode characters to evade detection by humans as well as some automated tools. The attackers take advantage of the fact that even with a keen examination of URLs, the substitution may go unnoticed because people are accustomed to examining them keenly. This is a radical step in phishing methods as cybercrime players today are increasingly depending on human trust and perception deficits instead of software technology vulnerabilities. Cybersecurity experts are prescribing increased vigilance and technology solutions to ward off such attacks. Users do not need to click on mail received in unsolicited messages, bookmark frequently visited sites and check the authenticity of the domain. Organisations are encouraged to implement multi-factor authentication (MFA), install sophisticated anti-phishing tools, and conduct regular training in new social engineering methods for employees. Encryption of the body of emails can also be supplemented by updating browsers and endpoint security software. Lastly, the best defence against these newer types of phishing is still good cyber hygiene and an increased level of awareness.
Three popular official extensions for Anthropic Claude Desktop were found to be vulnerable to critical RCE: the Chrome, iMessage, and Apple Notes connectors. With a CVSS severity s...
A newly discovered vulnerability, named "MadeYouReset" (CVE-2025-8671), has revealed a serious weakness within various HTTP/2 implementations that could enable attackers to...
Cybersecurity researchers from Tenable have identified a set of critical vulnerabilities in OpenAI’s ChatGPT, affecting both the GPT-5 and GPT-4 models. These flaws could allow a...