Researchers uncovered a critical prompt injection bug in Google's Antigravity AI coding tool, potentially allowing attackers to execute unauthorized commands despite built-in safeguards. This flaw highlights escalating AI security vulnerabilities in tech giants' offerings, echoing recent exploits in similar systems over the past six months that disrupted developer workflows. As AI industry impact grows, such bugs could erode trust in automated coding platforms, pushing firms toward stricter prompt engineering protocols to counter prompt injection attacks and maintain edge in the competitive AI tool bug landscape.