Over €250 million to support secure connectivity across the EU under the CEF Digital Programme | Shaping Europe’s digital future https://
europa.eu/!Tmjk6q
#connectivity #funding #digital #digitaleu #innovation #technology @ArturHabant @elaniaz @DigitalEU @CurieuxExplorer @enilev
SECURITY
-
EU invests €250 million secure connectivity CEF Digital Programme
By
–
-
Differentiating AI Jailbreaks from Prompt Injection Attacks
By
–
What you’re describing is a jailbreak, not a prompt injection. Prompt injection is when data is misinterpreted as part of the prompt instructions (against the intention of the prompter). A jailbreak is when the prompter bypasses safety policies of the model.
-
Researchers Jailbreak ChatGPT API by Accessing Hidden Token Probabilities
By
–
fun research story about how we jailbroke the the chatGPT API: so every time you run inference with a language model like GPT-whatever, the model outputs a full probabilities over its entire vocabulary (~50,000 tokens) but when you use their API, OpenAI hides all this info from
-
LLM Prompt Injection via Invisible Instructions in Pasted Text
By
–
PoC: LLM prompt injection via invisible instructions in pasted text
-
Ally and LangChain Enable PII Masking for GenAI Compliance
By
–
@Ally and @langchain collaborate to bring PII masking to the LangChain typescript package — paving the path for more genAI adoption, in a compliant and safe manner. Read more on the blog!
-
Data Broker X-Mode Sold Sensitive Healthcare Location Data
By
–
This is very good/another reminder that the surveillance biz is unfathomably sleazy. "X-Mode gave a clinical research company info…about consumers who had visited certain internal medical facilities & then pharmacies or infusion centers." https://
ftc.gov/news-events/ne
ws/press-releases/2024/01/ftc-order-prohibits-data-broker-x-mode-social-outlogic-selling-sensitive-location-data
… -
Ultimate ChatGPT: local, multi-AI, free and incredible
By
–
Here is the #Ultimate ChatGPT →
https://
youtu.be/RYFi-haG_dw Imagine a ChatGPT that: Runs locally for maximum security.
Is Multi-AI (GPT-4, Mistral AI, Llama 2…) And on top of that, it's FREE No need to imagine anymore. It's here and it's incredible! -
Meta Reports Chinese Fake Accounts Impersonating Americans Online
By
–
4,789 Facebook Accounts in China Impersonated Americans, Meta Says: The company warned that the inauthentic accounts underscored the threat of foreign election interference in 2024. https://
nytimes.com/2023/11/30/tec
hnology/facebook-china-fake.html?utm_source=dlvr.it&utm_medium=twitter
… #Technology #Tech #FutureTech -
Web3 Blockchain Vulnerabilities and Smart Contract Security Risks
By
–
Web3's blockchain solutions have introduced complex vulnerabilities with risks similar to traditional cybersecurity and potential bugs in smart contracts. Here are the most impactful Web3 vulnerabilities. Source @101Blockchains Link https://
bit.ly/46QLwIl via @antgrasso #web3