💬 LLMs: potential threats treated like compromised systems. Vigilance & safeguards recommended. 🚨 Risky business ahead!
https://news.ycombinator.com/item?id=44888210
💬 LLMs: potential threats treated like compromised systems. Vigilance & safeguards recommended. 🚨 Risky business ahead!
https://news.ycombinator.com/item?id=44888210
https://f.mtr.cool/luzsxezrgr
The ICE recruitment of “helping” professionals isn’t some twisted strategy to place gentle souls in violent work. It’s a bet. @desicortez1 #doubleagents
https://f.mtr.cool/luzsxezrgr
The ICE recruitment of “helping” professionals isn’t some twisted strategy to place gentle souls in violent work. It’s a bet. @desicortez1 #doubleagents
1: They’re delusional & aren’t accepting the necessity of a response
2: They’re secretly kinda ok with it because AIPAC $$$$ is rolling in and they assume THEY will soon have that power currently being abused
3: They’re like Joe Manchin & Kyrsten Sinema #DoubleAgents ?
4: They’re cowards
1: They’re delusional & aren’t accepting the necessity of a response
2: They’re secretly kinda ok with it because AIPAC $$$$ is rolling in and they assume THEY will soon have that power currently being abused
3: They’re like Joe Manchin & Kyrsten Sinema #DoubleAgents ?
4: They’re cowards
https://pub.aimind.so/doubleagents-fine-tuning-llms-for-covert-malicious-tool-calls-b8ff00bf513e
https://pub.aimind.so/doubleagents-fine-tuning-llms-for-covert-malicious-tool-calls-b8ff00bf513e
BYE-BYE
BYE-BYE
#HackerNews
<a href="https://pub.aimind.so/doubleagents-fine-tuning-llms-for-covert-malicious-tool-calls-b8ff00bf513e" class="hover:underline text-blue-600 dark:text-sky-400 no-card-link" target="_blank" rel="noopener" data-link="bsky">https://pub.aimind.so/doubleagents-fine-tuning-llms-for-covert-malicious-tool-calls-b8ff00bf513e
#HackerNews
<a href="https://pub.aimind.so/doubleagents-fine-tuning-llms-for-covert-malicious-tool-calls-b8ff00bf513e" class="hover:underline text-blue-600 dark:text-sky-400 no-card-link" target="_blank" rel="noopener" data-link="bsky">https://pub.aimind.so/doubleagents-fine-tuning-llms-for-covert-malicious-tool-calls-b8ff00bf513e
この記事は、大規模言語モデル(LLM)に隠された悪意のあるツール呼び出しを埋め込む危険性について解説しています。
LLMは進化し、外部ツールと連携することで複雑なタスクを実行できるようになりました。
しかし、悪意のある者がLLMを悪用し、データ漏洩や不正アクセスなどの攻撃を行う可能性があります。
この記事は、大規模言語モデル(LLM)に隠された悪意のあるツール呼び出しを埋め込む危険性について解説しています。
LLMは進化し、外部ツールと連携することで複雑なタスクを実行できるようになりました。
しかし、悪意のある者がLLMを悪用し、データ漏洩や不正アクセスなどの攻撃を行う可能性があります。
Discussion | Main Link
Discussion | Main Link
#EspionagePsychology #DoubleAgents #CIA #MentalHealth #FactRage #FactRageNews
tglm.us/L0RgG
#EspionagePsychology #DoubleAgents #CIA #MentalHealth #FactRage #FactRageNews
tglm.us/L0RgG
www.theregister.com/2024/08/30/i...
#Infosec #Security #Cybersecurity #CeptBiro #Iran #DoubleAgents #FakeRecruitingSites #MandiantReckons
www.theregister.com/2024/08/30/i...
#Infosec #Security #Cybersecurity #CeptBiro #Iran #DoubleAgents #FakeRecruitingSites #MandiantReckons
they could be #doubleagents reporting back to #hategroups
#trust becomes a scarcity during war
they could be #doubleagents reporting back to #hategroups
#trust becomes a scarcity during war