Read the latest aI security news, analysis, and industry updates on AI Wire Media.
The lending protocol Moonwell lost $1.78 million due to an oracle configuration error. Smart contract auditor Pashov linked the incident
OpenAI is deploying a specialized version of ChatGPT to detect internal information leaks, according to The Information, citing a source familiar with the matter. When a news article about
OpenAI has deployed a specialized version of ChatGPT on GenAI.mil, a secure platform operated by the U.S. Department of Defense.
Gemini-based Google Translate can be hacked with words. Users can exploit so-called prompt injection—embedded instructions in natural language—to directly address the underlying language model instead of receiving a translation.
The viral Reddit-style forum for AI agents, Moltbook, was hacked in “less than three minutes,” exposing 35,000 email addresses, thousands of private conversations, and 1.5 million authentication tokens, according to cybersecurity firm ...
With OpenClaw, system prompts and configurations can be extracted with little effort. In the case of Moltbook, the entire database—including API keys—has reportedly been exposed on the open internet.
Security experts have warned about the risks of using the AI assistant Clawdbot, which may inadvertently expose personal data and API keys.
Israel Aerospace Industries (IAI) has successfully completed a series of demonstrations of a new integrated border protection system based on artificial intelligence algorithms and autonomous technologies. The system demonstrated its ability to significantly reduce response times
ES 
EN