Dropsafe
by Alec Muffett
Resources
A Primer for End-to-End Encryption
A “Duck Test” for End-to-End Secure Messaging
Why Privacy will always be more impactful than Safety
Muffett on Passwords
Pre-Flight Travel Checklist
About
About
Login
FAQs
Search
RSS
RSS (All)
RSS (Comments)
Tag:
artificial intelligence
BREAKING: food bloggers are adding nightshade and other poisons to online recipe pages to defeat “cooked.wiki” and other advert/revenue-stripping robots
2024/01/21
This new data poisoning tool lets artists fight back against generative AI | …I’m always torn whether to call these things “pettyware” or “get off my lawn-ware”
2024/01/15
The I in LLM stands for intelligence | …excellent post by the Curl guy re: LLMs and Credibility
2024/01/03
Kevin A. Bryan on X: “NYT/OpenAI lawsuit completely misunderstands how LLMs work, and judges getting this wrong will do huge damage to AI. Basic point: LLMs DON’T “STORE” UNDERLYING TRAINING TEXT. It is impossible, the parameter size of GPT-3.5 or 4 is not enough to losslessly encode the training set”
2023/12/29
The Future of Model Deletion | …of *course* “ill-gotten data” should be deleted, because it’s ill-gotten and *everyone* can see that; but I’m wondering how long before this idea runs into people who consider data more like librarians than philosophers
2023/12/22
“Suffice it to say that everyone in possession of a copy of the LAION-5B images has hundreds if not thousands of instances of CSAM” | …so that’s 0.0001% of the content, then
2023/12/21
Skating to Where the Puck Is Going – Center for Security & Emerging Technology | overall I’m grateful to the AI community for absorbing so many of the professional fearmongers hitherto locked-into VR & Infosec
2023/11/23
Michael Nielsen: “In the Mission in SF there was a flyer posted on a street pole, with a QR code that would let you download the Llama weights”
2023/11/14
Prediction: One day soon there will be a clickbait piece in the media about “Regulators Angry at Nazis in Free iPhone Games and SomethingMustBeDone™” #touchHLE #Censorship #ArtificialIntelligence
2023/11/07
Simon Willison on X: “@AlecMuffett @AnthropicAI Well that worked!” | it turns out that the simplest way to bypass security of a AI sometimes is just to lie to them
2023/10/26
A thought provocation for people who want to regulate artificial intelligence software & training data
2023/10/19
Raspberry Pi Detects License Plates with AI
2023/10/19
Previous Page
1
2