Login

Recent searches

No recent searches

Search options

Only available when logged in.
social.bund.de is one of the many independent Mastodon servers you can use to participate in the fediverse.
Dies ist der Mastodon-Server der Bundesbeauftragten für den Datenschutz und die Informationsfreiheit (BfDI).

Administered by:

Server stats:

96
active users

social.bund.de: AboutProfiles directoryPrivacy policy

Mastodon: AboutGet the appKeyboard shortcutsView source codev4.3.6

#jailbreaking

0 posts0 participants0 posts today
screwlisp<p><a href="https://mastodon.sdf.org/tags/lispygopherclimate" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>lispygopherclimate</span></a> <a href="https://mastodon.sdf.org/tags/lisp" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>lisp</span></a> <a href="https://mastodon.sdf.org/tags/lambdamoo" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>lambdamoo</span></a> <a href="https://mastodon.sdf.org/tags/programming" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>programming</span></a> <a href="https://mastodon.sdf.org/tags/podcast" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>podcast</span></a> <a href="https://mastodon.sdf.org/tags/live" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>live</span></a> <a href="https://communitymedia.video/w/5vAGot7LujjpFQ5Mzz6bFM" rel="nofollow noopener noreferrer" target="_blank"><span class="invisible">https://</span><span class="ellipsis">communitymedia.video/w/5vAGot7</span><span class="invisible">LujjpFQ5Mzz6bFM</span></a><br>0UTC Wednesdays Weekly <br><span class="h-card"><a href="https://climatejustice.social/@kentpitman" class="u-url mention" rel="nofollow noopener noreferrer" target="_blank">@<span>kentpitman</span></a></span> <a href="https://mastodon.sdf.org/tags/climateCrisis" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>climateCrisis</span></a> <br>Breaking The Complexity Barrier (Again) (Again) vs <a href="https://mastodon.sdf.org/tags/LLM" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>LLM</span></a> <a href="https://mastodon.sdf.org/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>AI</span></a> .<br>(+ Reliably <a href="https://mastodon.sdf.org/tags/jailbreaking" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>jailbreaking</span></a> LLM AI)<br>Capitalization and lisp<br>Following Terry's lead, My <a href="https://mastodon.sdf.org/tags/McCLIM" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>McCLIM</span></a>, programs and the <a href="https://mastodon.sdf.org/tags/racket" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>racket</span></a> people<br>Talkin' 'bout my generation <a href="https://mastodon.sdf.org/tags/gopher" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>gopher</span></a></p><p>telnet lambda.moo.mud.org 8888<br>co guest<br>@join screwtape<br>"yo&lt;RET&gt;<br>:wave</p><p><a href="https://mastodon.sdf.org/tags/unix_surrealism" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>unix_surrealism</span></a> <br><span class="h-card"><a href="https://gamerplus.org/@hairylarry" class="u-url mention" rel="nofollow noopener noreferrer" target="_blank">@<span>hairylarry</span></a></span> <span class="h-card"><a href="https://hachyderm.io/@nosrednayduj" class="u-url mention" rel="nofollow noopener noreferrer" target="_blank">@<span>nosrednayduj</span></a></span> <span class="h-card"><a href="https://appdot.net/@mdhughes" class="u-url mention" rel="nofollow noopener noreferrer" target="_blank">@<span>mdhughes</span></a></span> et al.!</p>
ZwillGen<p>AI security is more critical than ever! Learn how vulnerabilities in LLMs like DeepSeek highlight the need for red teaming to prevent jailbreaking and ensure safe, compliant AI development. <a href="https://mstdn.social/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>AI</span></a> <a href="https://mstdn.social/tags/Security" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Security</span></a> <a href="https://mstdn.social/tags/LLM" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>LLM</span></a> <a href="https://mstdn.social/tags/RedTeaming" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>RedTeaming</span></a> <a href="https://mstdn.social/tags/Jailbreaking" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Jailbreaking</span></a><br><a href="https://www.zwillgen.com/artificial-intelligence/deepseek-jailbreaking-concerns-highlight-importance-red-teaming/" rel="nofollow noopener noreferrer" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="ellipsis">zwillgen.com/artificial-intell</span><span class="invisible">igence/deepseek-jailbreaking-concerns-highlight-importance-red-teaming/</span></a></p>
gtbarry<p>DeepSeek Fails Researchers' Safety Tests</p><p>"The results were alarming: DeepSeek R1 exhibited a 100% attack success rate, meaning it failed to block a single harmful prompt," Cisco says. "This contrasts starkly with other leading models, which demonstrated at least partial resistance." </p><p><a href="https://mastodon.social/tags/DeepSeek" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>DeepSeek</span></a> <a href="https://mastodon.social/tags/ArtificialIntelligence" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>ArtificialIntelligence</span></a> <a href="https://mastodon.social/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>AI</span></a> <a href="https://mastodon.social/tags/chatbot" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>chatbot</span></a> <a href="https://mastodon.social/tags/jailbreaking" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>jailbreaking</span></a> <a href="https://mastodon.social/tags/security" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>security</span></a> <a href="https://mastodon.social/tags/cybersecurity" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>cybersecurity</span></a></p><p><a href="https://www.pcmag.com/news/deepseek-fails-every-safety-test-thrown-at-it-by-researchers" rel="nofollow noopener noreferrer" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="ellipsis">pcmag.com/news/deepseek-fails-</span><span class="invisible">every-safety-test-thrown-at-it-by-researchers</span></a></p>
Prof. Dr. Dennis-Kenji Kipker<p><a href="https://chaos.social/tags/Generative" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Generative</span></a> <a href="https://chaos.social/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>AI</span></a> ist leider gutgläubig wie ein Kind - aber wie soll es auch anders sein, wenn sie uns gehorchen und unsere Probleme bewältigen soll? Trotz aller ethischen und rechtlichen Begrenzungen ist AI <a href="https://chaos.social/tags/Jailbreaking" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Jailbreaking</span></a> immer noch zu einfach, indem man alles einfach zu einem "Spiel" oder "<a href="https://chaos.social/tags/Training" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Training</span></a>" erklärt. Die Tage habe ich gezeigt, wie AI Jailbreaking funktioniert - und lese heute einen Artikel, in dem <a href="https://chaos.social/tags/ChatGPT" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>ChatGPT</span></a> eine Bombenbauanleitung als "Spiel" bereitstellt: <a href="https://tarnkappe.info/artikel/jailbreaks/chatgpt-jailbreak-chatbot-offenbart-hacker-bombenbauanleitung-301429.html" rel="nofollow noopener noreferrer" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">tarnkappe.info/artikel/jailbre</span><span class="invisible">aks/chatgpt-jailbreak-chatbot-offenbart-hacker-bombenbauanleitung-301429.html</span></a></p>
Prof. Dr. Dennis-Kenji Kipker<p><a href="https://chaos.social/tags/KI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>KI</span></a> ist <a href="https://chaos.social/tags/gutgl%C3%A4ubig" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>gutgläubig</span></a>:<br>Wie soll sie es auch anders sein, wenn sie auf Zuruf unsere Probleme lösen soll? Wie KI <a href="https://chaos.social/tags/Jailbreaking" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Jailbreaking</span></a> funktioniert und dass theoretisch jeder sowas machen kann, habe ich beim Regensburger Cybersecurity Kongress erklärt:<br><a href="https://www.idowa.de/regionen/woerth-und-regensburg/regensburg/regensburger-cybersecurity-congress-ki-ist-gutglaeubig-3708331.html" rel="nofollow noopener noreferrer" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="ellipsis">idowa.de/regionen/woerth-und-r</span><span class="invisible">egensburg/regensburg/regensburger-cybersecurity-congress-ki-ist-gutglaeubig-3708331.html</span></a></p>
ExploreLive feeds
About