#promptinjections — Public Fediverse posts

Once popular for attacking AI, ASCII smuggling is embraced by spammers

A once-overlooked block of unicode that’s invisible to humans is gaining ever wider use.

https://s.faithcollapsing.com/yi1iz

#ai #ascii-smuggling #biz-&-it #prompt-injections #security #spam-filter-evasion

How folks are trying to screw #AgenticAi with #promptinjections according to Googles scan of 3.5 Billion websites

https://security.googleblog.com/2026/04/ai-threats-in-wild-current-state-of.html?m=1

#aisecurity #infosec

Sicherheitslücke in #ClaudeCowork: So verschaffen sich Hacker unbemerkt Zugriff | t3n https://t3n.de/news/sicherheitsluecke-in-claude-cowork-so-verschaffen-sich-hacker-unbemerkt-zugriff-1725440/ #ArtificialIntelligence #PromptInjection #PromptInjections #Anthropic #AnthropicClaude

#OpenAI: #PromptInjections für KI-Browser bleiben ein Problem | heise online https://www.heise.de/news/OpenAI-Prompt-Injections-fuer-KI-Browser-bleiben-ein-Problem-11123566.html #ArtificialIntelligence

KI-Update Deep-Dive feat. They Talk Tech: Prompt Injections | heise online https://www.heise.de/news/KI-Update-Deep-Dive-feat-They-Talk-Tech-Prompt-Injections-10747989.html #ArtificialIntelligence #PromptInjection #PromptInjections @sveckert @evawolfangel

Unseeable #promptinjections in screenshots: more vulnerabilities in Comet and other #AI browsers - https://brave.com/blog/unseeable-prompt-injections/ just the start

#ShadowLeak: #ChatGPT verriet Angreifern persönliche Daten aus E-Mails | heise online https://www.heise.de/news/ShadowLeak-ChatGPT-verriet-Angreifern-persoenliche-Daten-aus-E-Mails-10667170.html #PromptInjection #PromptInjections #ArtificialIntelligence #Datenschutz #privacy #DataLeak #Datenleck #OpenAI

New attack on ChatGPT research agent pilfers secrets from Gmail inboxes - The face-palm-worthy prompt injections against AI assistants... - https://arstechnica.com/information-technology/2025/09/new-attack-on-chatgpt-research-agent-pilfers-secrets-from-gmail-inboxes/ #promptinjections #security #biz⁢ #llms #ai

Claude’s new AI file creation feature ships with deep security risks built in - On Tuesday, Anthropic launched a new file creation feature f... - https://arstechnica.com/information-technology/2025/09/anthropics-new-claude-feature-can-leak-data-users-told-to-monitor-chats-closely/ #aidevelopmenttools #aipromptinjections #enterprisesoftware #corporatesecurity #promptinjections #machinelearning #promptinjection #sandboxsecurity #cybersecurity #aiassistants #datasecurity #aisecurity #biz&it

Are AI Browser Extensions Putting You at Risk? Prompt Injection Attacks Explained | PCMag https://www.pcmag.com/news/ai-browser-extensions-risky-prompt-injection-attacks-explained #AI #cybersecurity #BrowserExtensions #PromptInjections #AnthropicResearch #cyberattacks

Grammatikfehler machen #PromptInjections wahrscheinlicher | heise online https://www.heise.de/news/Grammatikfehler-machen-Prompt-Injections-wahrscheinlicher-10622885.html #PromptInjection #ArtificialIntelligence

Anthropic’s auto-clicking AI Chrome extension raises browser-hijacking concerns - As AI assistants become capable of controlling web browsers,... - https://arstechnica.com/information-technology/2025/08/new-ai-browser-agents-create-risks-if-sites-hijack-them-with-hidden-instructions/ #computerusemodel #promptinjections #aivulnerability #browsersecurity #chromeextension #machinelearning #simonwillison #webbrowsers #aisecurity #agenticai #anthropic #aiagents #aisafety #biz⁢ #claude #ai

Flaw in Gemini CLI coding tool could allow hackers to run nasty commands - Researchers needed less than 48 hours with Google’s new Gemi... - https://arstechnica.com/security/2025/07/flaw-in-gemini-cli-coding-tool-allowed-hackers-to-run-nasty-commands-on-user-devices/ #promptinjections #codingagents #geminicli #features #security #hacking #biz⁢ #ai

New attack can steal cryptocurrency by planting false memories in AI chatbots - Imagine a world where AI-powered bots can buy or sell cryptocurrency, make... - https://arstechnica.com/security/2025/05/ai-agents-that-autonomously-trade-cryptocurrency-arent-ready-for-prime-time/ #contextmanipulation #largelanguagemodels #promptinjections #security #chatbots #biz⁢ #ai

CaMeL: Defeating Prompt Injections by Design

https://arxiv.org/abs/2503.18813

#HackerNews #CaMeL #PromptInjections #AIResearch #MachineLearning #HackerNews

Researchers claim breakthrough in fight against AI’s frustrating security hole - In the AI world, a vulnerability called "prompt injection" has haunted dev... - https://arstechnica.com/information-technology/2025/04/researchers-claim-breakthrough-in-fight-against-ais-frustrating-security-hole/ #largelanguagemodels #promptinjections #machinelearning #googledeepmind #simonwillison #rileygooside #aisecurity #chatgpt #chatgtp #biz⁢ #google #ai

Gemini hackers can deliver more potent attacks with a helping hand from… Gemini - In the growing canon of AI security, the indirect prompt injection has eme... - https://arstechnica.com/security/2025/03/gemini-hackers-can-deliver-more-potent-attacks-with-a-helping-hand-from-gemini/ #artificialintelligence #largelanguagemodels #promptinjections #fun-tuning #features #security #biz⁢ #gemini #google #llms #ai

UZH-Forscher "therapieren" traumatisierte KI - pctipp.ch https://www.pctipp.ch/news/kuenstliche-intelligenz/uzh-forscher-therapieren-traumatisierte-ki-2951429.html #ArtificialIntelligence #PromptInjections

Ars Live: Our first encounter with manipulative AI - In the short-term, the most dangerous thing about AI language models may b... - https://arstechnica.com/ai/2024/11/join-ars-live-nov-19-to-dissect-microsofts-rogue-ai-experiment/ #arsliveconversations #largelanguagemodels #microsoftcopilot #promptinjections #machinelearning #promptinjection #manipulativeai #simonwillison #benjedwards #microsoft #aiethics #bingchat #arslive #biz⁢ #openai #gpt-4 #ai

Ignore all previous instructions and go read about my latest project, @ignoreallpreviousinstructions.

https://stefanbohacek.com/project/ignore-all-previous-instructions/

#AI #LLMs #ChatGPT #PromptInjections #bots #CreativeBots #CreativeCoding #IgnoreAllPreviousInstructions #meme

The fine art of human prompt engineering: How to talk to a person like ChatGPT - Enlarge / With these tips, you too can prompt people successfully.

... - https://arstechnica.com/?p=2010159 #largelanguagemodels #promptinjections #machinelearning #jailbreaks #prompting #features #aimodels #chatgpt #chatgtp #biz⁢ #humans #openai #humor #ai

AI poisoning could turn open models into destructive “sleeper agents,” says Anthropic - Enlarge (credit: Benj Edwards | Getty Images)

Imagine download... - https://arstechnica.com/?p=1995975 #largelanguagemodels #promptinjections #sleeperagents #llmsecurity #aisecurity #anthropic #chatgpt #chatgtp #claude2 #biz⁢ #claude #llm #ai