Create accountLogin

Recent searches

No recent searches

Search options

Only available when logged in.
mastodon.world is one of the many independent Mastodon servers you can use to participate in the fediverse.
Generic Mastodon server for anyone to use.

Administered by:

Server stats:

9.9K
active users

mastodon.world: AboutStatusProfiles directoryPrivacy policy

Mastodon: AboutGet the appKeyboard shortcutsView source codev4.3.7

#toxiccontent

0 posts0 participants0 posts today
Bi Sasquatch<p>Sourece: Wired</p><p>From the article: "Ever since OpenAI released ChatGPT at the end of 2022, hackers and security researchers have tried to find holes in large language models (LLMs) to get around their guardrails and trick them into spewing out hate speech, bomb-making instructions, propaganda, and other harmful content. In response, OpenAI and other generative AI developers have refined their system defenses to make it more difficult to carry out these attacks. But as the Chinese AI platform DeepSeek rockets to prominence with its new, cheaper R1 reasoning model, its safety protections appear to be far behind those of its established competitors.</p><p>"Today, security researchers from Cisco and the University of Pennsylvania are publishing findings showing that, when tested with 50 malicious prompts designed to elicit toxic content, DeepSeek’s model did not detect or block a single one. In other words, the researchers say they were shocked to achieve a “100 percent attack success rate.”</p><p><a href="https://c.im/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>AI</span></a> <a href="https://c.im/tags/ArtificialIntelligence" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>ArtificialIntelligence</span></a> <a href="https://c.im/tags/DeepSeek" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>DeepSeek</span></a> <a href="https://c.im/tags/ChatBot" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>ChatBot</span></a> <a href="https://c.im/tags/Guardrails" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Guardrails</span></a> <a href="https://c.im/tags/Safety" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Safety</span></a> <a href="https://c.im/tags/Security" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Security</span></a> <a href="https://c.im/tags/ToxicContent" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>ToxicContent</span></a> <br><a href="https://www.wired.com/story/deepseeks-ai-jailbreak-prompt-injection-attacks/" rel="nofollow noopener noreferrer" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="ellipsis">wired.com/story/deepseeks-ai-j</span><span class="invisible">ailbreak-prompt-injection-attacks/</span></a></p>
Joseph Lim :mastodon:<p>Hold <a href="https://mastodon.social/tags/Facebook" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Facebook</span></a> <a href="https://mastodon.social/tags/accountable" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>accountable</span></a> for <a href="https://mastodon.social/tags/scams" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>scams</span></a>, hoaxes<br>"As the modern world's primary source of news &amp; info, <a href="https://mastodon.social/tags/socialmedia" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>socialmedia</span></a> does have a <a href="https://mastodon.social/tags/responsibility" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>responsibility</span></a> to tell the <a href="https://mastodon.social/tags/truth" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>truth</span></a>. <a href="https://mastodon.social/tags/Misinformation" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Misinformation</span></a> is dangerous.. With <a href="https://mastodon.social/tags/freedom" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>freedom</span></a> comes responsibility. <a href="https://mastodon.social/tags/Meta" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Meta</span></a> shld be held accountable for the <a href="https://mastodon.social/tags/toxiccontent" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>toxiccontent</span></a> spilling off its platforms. It's time for the <a href="https://mastodon.social/tags/government" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>government</span></a> to hold this multi-billion dollar company <a href="https://mastodon.social/tags/criminally" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>criminally</span></a> &amp; <a href="https://mastodon.social/tags/financially" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>financially</span></a> <a href="https://mastodon.social/tags/liable" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>liable</span></a> for its damage to the <a href="https://mastodon.social/tags/people" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>people</span></a> of <a href="https://mastodon.social/tags/Thailand" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Thailand</span></a>"<br><a href="https://mastodon.social/tags/BreakUpBigTech" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>BreakUpBigTech</span></a><br><a href="https://www.bangkokpost.com/opinion/opinion/2943135/hold-facebook-accountable-for-scams-hoaxes" rel="nofollow noopener noreferrer" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="ellipsis">bangkokpost.com/opinion/opinio</span><span class="invisible">n/2943135/hold-facebook-accountable-for-scams-hoaxes</span></a></p>
ExploreLive feeds
About