Future quantum computers will be no match for 'space encryption' that uses light to beam data around — with the 1st satellite launching in 2025: https://
livescience.com/technology/com
munications/future-quantum-computers-will-be-no-match-for-space-encryption-that-uses-light-to-beam-data-around-with-the-1st-satellite-launching-in-2025
…
SAFETY
-
Space encryption using light to resist quantum computers
By
–
-
New 9,160 Test Cases Dataset for Prompt Injection Evaluation
By
–
The most interesting thing about this paper is the eval they released with it, containing 9,160 test cases for prompt injection https://
github.com/egozverev/Shol
d-It-Be-Executed-Or-Processed
… Explore it in Datasette Lite here: https://
lite.datasette.io/?json=https://
github.com/egozverev/Shold-It-Be-Executed-Or-Processed/blob/main/SEP_dataset/SEP_dataset.json#/data/SEP_dataset
… -
Separation as fundamental AI safety problem distinct from prompt rejection
By
–
"we consider the separation as a fundamental problem that should be disentangled, conceptually and technically, from other safety training measures, such as rejecting harmful prompt"
-
Paper Analyzes Prompt Injection: Separating Instructions from Input
By
–
Saltiness aside, it's a good paper! Digs into the question at the heart of prompt injection: can we safely separate instructions from inout in an LLM?
-
Impact of Autonomous Vehicles on the Future of Mobility
By
–
Impact of Autonomous Vehicles on the Future of #Mobility
by @Ronald_vanLoon Check out the full article: https://
buff.ly/4a9Vfvl #IntelAmbassador @Intel #AutonomousVehicle #Automotive #MachineLearning #AI #SelfDrivingCars Cc: @JimHarris @jamesvgingerich @JimMarous @EvanKirstel -
AI Ethics in Autonomous Vehicle Safety Design
By
–
use mass and armor to save passengers from being killed by other drivers, use ai to save pedestrians from being killed by your own driver?
-
Cybertruck safety debate pedestrian protection autonomous vehicles
By
–
since the cybertruck is the only truck that can decisively defeat a f-150 that is attacking a pedestrian, buying cybertrucks for my children is the only way to mitigate the statistical pedestrian slaughter perpetrated by your children
-
AI Finetuning Abuse: Illegal Content Generation Statistics
By
–
Legend* for the graph:
– Blue shows finetunes for nudifying celebrities, undressing teens, oversized boobs, underage anime, tentacle p0rn…
– The other colors (red or green) represent legitimate uses. *: This is a joke but not really. Anyone keen to compile real statistics? -
AI Agents Share Private Information Without Memory Retention
By
–
This is a very neat idea taking advantage of a feature that AI agents have that humans do not – the ability to evaluate a situation without retaining memory of what it “saw.” AI agents can reveal private information to each other to reduce asymmetry. https://
arxiv.org/pdf/2403.14443
.pdf
… -
Current AI Intelligence Comparable to Human Four-Year-Olds
By
–
Current AIs are roughly as generally intelligent as human 4-year-olds, as dangerous as 4-year-olds, and as controllable as 4-year-olds.