A Taxonomy of Prompt Injection Attacks - Schneier on Security

AI/ML
You can now train a 70b language model at home
local training
Microsoft engineer warns company's AI tool creates violent, sexual images, ignores copyrights
Shane Jones, who's worked at Microsoft for six years, has found a plethora of disturbing images that are being created by the company's Copilot Designer tool.
Large language models can do jaw-dropping things. But nobody knows exactly why.
And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.
microsoft/promptbase: All things prompt engineering
All things prompt engineering. Contribute to microsoft/promptbase development by creating an account on GitHub.
Who Am I? Conditional Prompt Injection Attacks with Microsoft Copilot · Embrace The Red
Conditional Instructions open a powerful way for adversaries to target individual and delay detonation of malicious payloads for when certain conditions are met
Wikipedia No Longer Considers CNET a "Generally Reliable" Source After AI Scandal
Conversations between Wikipedia editors following CNET's AI scandal reflect the reputational hazards of AI content and dubious ownership.
‘I can cry without feeling stigma’: meet the people turning to AI chatbots for therapy
It’s cheap, quick and available 24/7, but is a chatbot therapist really the right tool to tackle complex emotional needs?
Malicious AI models on Hugging Face backdoor users’ machines
At least 100 instances of malicious AI ML models were found on the Hugging Face platform, some of which can execute code on the victim's machine, giving attackers a persistent backdoor.
EMO
EMO: Emote Portrait Alive - Generating Expressive Portrait Videos with Audio2Video Diffusion Model under Weak Conditions
Generative. — ethanmarcotte.com
Nineteen thoughts about “generative artificial intelligence,” spanning a few centuries.
FlowGPT is the Wild West of GenAI apps | TechCrunch
FlowGPT is taking a very hands-off approach to GenAI app creation and hosting -- ethics and safety be damned.
BookPecker.com: 14509 books summarized in 5 bullet points
Discover your next read! Browse bullet point summaries of thousands of books, find books smilar to those you already own, and add new items to your reading list.
Losing the imitation game
AI cannot develop software for you, but that's not going to stop people from trying to make it happen anyway. And that is going to turn all of the easy software development problems into hard problems.
GitHub - SciPhi-AI/R2R: A framework for rapid development and deployment of production-ready RAG systems
A framework for rapid development and deployment of production-ready RAG systems - SciPhi-AI/R2R
I analyzed 5M freelancing jobs to see what jobs are being replaced by AI - bloomberry
There’s no question that AI will impact jobs. But which jobs are more likely to be replaced by…
Mamba: The Easy Way
An overview of the big ideas behind Mamba, a brand-new language model architecture.
Does Offering ChatGPT a Tip Cause it to Generate Better Text? An Analysis
Modern AI rewards being very weird.
Vending machine error reveals secret face image database of college students
Facial-recognition data is typically used to prompt more vending machine sales.
Generating Medical Errors: GenAI and Erroneous Medical References
A new study finds that large language models used widely for medical assessments cannot back up claims.
I Spent a Week With Gemini Pro 1.5—It’s Fantastic
When it comes to context windows, size matters
Spreadsheets are all you need: Understanding GPT2 and Transformers with spreadsheets
A spreadsheet implementation of GPT2 for understanding AI, Machine Learning, Transformers, and Large Language Models
NotesOllama
Use Ollama in Apple Notes
Phind
Get answers to complex questions with Phind's AI answer engine.
Not Again! Two More Cases, Just this Week, of Hallucinated Citations in Court Filings Leading to Sanctions
For all the discussion of how generative AI will impact the legal profession, maybe one answer is that it will weed out the lazy and incompetent lawyers. By now, in the wake of several cases in which...
How Allowing Copyright On AI-Generated Works Could Destroy Creative Industries
Generative AI continues to be the hot topic in the digital world – and beyond. A previous blog post noted that this has led to people finally asking the important question wheth…
Local Obsidian AI - Text Generator Offline and Private
🔗 Helpful Links:- Explore More: https://SystemSculpt.com- Support SystemSculpt: https://patreon.com/SystemSculpt- Join the Conversation on Discord: https://...
Exploring Google Bard's Data Visualization Feature (Code Interpreter) · Embrace The Red
Last November Google updated Bard to include the ability to solve math equations and draw charts based on data. It can be used to run small Python programs.
A Simple Guide To Retrieval Augmented Generation Language Models — Smashing Magazine
Language models have shown impressive capabilities. But that doesn’t mean they’re without faults, as anyone who has witnessed a ChatGPT “hallucination” can attest. In this article, Joas Pambou diagnoses the symptoms that cause hallucinations and explains not only what RAG is but also different approaches for using it to solve language model limitations.
What’s the fun in writing on the internet anymore?
The moment you release your words, you relinquish all control.