digitalcourage.social

Benjamin Carr, Ph.D. 👨🏻‍💻🧬Here Come the <a href="https://hachyderm.io/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#AI</a> <a href="https://hachyderm.io/tags/Worms" class="mention hashtag" rel="nofollow noopener" target="_blank">#Worms</a> <a href="https://hachyderm.io/tags/Security" class="mention hashtag" rel="nofollow noopener" target="_blank">#Security</a> researchers created <a href="https://hachyderm.io/tags/AIworm" class="mention hashtag" rel="nofollow noopener" target="_blank">#AIworm</a> in a test environment that can automatically spread between <a href="https://hachyderm.io/tags/generativeAI" class="mention hashtag" rel="nofollow noopener" target="_blank">#generativeAI</a> agents—potentially stealing data and sending spam emails. To create <a href="https://hachyderm.io/tags/MorrisII" class="mention hashtag" rel="nofollow noopener" target="_blank">#MorrisII</a>, researchers turned to a so-called “adversarial self-replicating prompt.” This is a prompt that triggers the generative AI model to output, in its response, another prompt. In short, <a href="https://hachyderm.io/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#AI</a> system is told to produce a set of further instructions in its replies. <a href="https://www.wired.com/story/here-come-the-ai-worms/" rel="nofollow noopener" translate="no" target="_blank">https://www.wired.com/story/here-come-the-ai-worms/</a>

Benjamin Carr, Ph.D. 👨🏻‍💻🧬<a href="https://hachyderm.io/tags/AIworm" class="mention hashtag" rel="nofollow noopener" target="_blank">#AIworm</a> infects users via <a href="https://hachyderm.io/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#AI</a>-enabled <a href="https://hachyderm.io/tags/emailclients" class="mention hashtag" rel="nofollow noopener" target="_blank">#emailclients</a> — <a href="https://hachyderm.io/tags/MorrisII" class="mention hashtag" rel="nofollow noopener" target="_blank">#MorrisII</a> <a href="https://hachyderm.io/tags/generativeAI" class="mention hashtag" rel="nofollow noopener" target="_blank">#generativeAI</a> worm steals confidential data as it spreads Named 'Morris II' after original Morris, first computer worm that created a worldwide nuisance online in 1988. This worm targets AI apps and AI-enabled email assistants that generate text and images using models like <a href="https://hachyderm.io/tags/GeminiPro" class="mention hashtag" rel="nofollow noopener" target="_blank">#GeminiPro</a>, <a href="https://hachyderm.io/tags/ChatGPT4" class="mention hashtag" rel="nofollow noopener" target="_blank">#ChatGPT4</a>.0, and <a href="https://hachyderm.io/tags/LLaVA" class="mention hashtag" rel="nofollow noopener" target="_blank">#LLaVA</a>. Researchers successfully tested this Morris II <a href="https://hachyderm.io/tags/worm" class="mention hashtag" rel="nofollow noopener" target="_blank">#worm</a> and published its findings using two methods. <a href="https://www.tomshardware.com/tech-industry/artificial-intelligence/ai-worm-infects-users-via-ai-enabled-email-clients-morris-ii-generative-ai-worm-steals-confidential-data-as-it-spreads" rel="nofollow noopener" translate="no" target="_blank">https://www.tomshardware.com/tech-industry/artificial-intelligence/ai-worm-infects-users-via-ai-enabled-email-clients-morris-ii-generative-ai-worm-steals-confidential-data-as-it-spreads</a>

Recent searches

Search options

Administered by:

Server stats:

#aiworm