just you wait for the next batch of Midjourney stepmoms META is gonna skyrocket while one-shotting humanity
SAFETY
-
Clarifying Prompt Injection vs Jailbreaking Misconceptions
By
–
Plus I'd had limited success getting people to understand "prompt injection" because too many people assumed it meant people "injecting bad prompts" into the system themselves, conflating it with jailbreaking
-
Lethal Trifecta: Data Exfiltration Through Prompt Injection Attacks
By
–
That's regular prompt injection. I came up with the "lethal trifecta" framing to help people understand the data exfiltration subset of prompt injection, because the same exact bug kept on showing up again and again and again across different systems
-
Salesforce Agentforce Prompt Injection Vulnerability Fixed
By
–
Classic prompt injection exfiltration attack here, reported against Salesforce Agentforce and now fixed
-
CA-1 Europa: Autonomous Combat Aircraft with Centaur AI Pilot
By
–
It was just a matter of time: CA-1 Europa, the upcoming autonomous combat aircraft built around the Centaur AI pilot. Wars will be AI fought.
— Chubby♨️ (@kimmonismus) 26 septembre 2025
pic.twitter.com/0L5v52GLccIt was just a matter of time: CA-1 Europa, the upcoming autonomous combat aircraft built around the Centaur AI pilot. Wars will be AI fought.
-
AVM Agent AI: Low Knowledge, Excessive Agreeableness Issues
By
–
AVM is what I’m referring to unfortunately. Latency is great but it’s not knowledgeable and way too agreeable
-
Medical AI Benchmarks: Beyond Memorization to Real-World Performance
By
–
Another paper pointing out the inadequacy of older public benchmarks for determining whether AI is actually good for tasks like medicine. Models are clearly memorizing or using heuristics for some answers. A new wave of benchmarks based on real-world data will help, more needed.
-
Developer logs blocked AI replies from platform
By
–
Search from:levelsio "blocked for ai reply" I log all of them
-
Taking Extra Care with Model Decoding and Out-of-Distribution Constraints
By
–
okay maybe the question is – how would one take “extra care”? the D is the D, cant be touched. and yet people are constrained decoding thousands of tokens with presumably only slight intelligence drop. maybe my json example is too light of a constraint. the OODness would
-
Machine Consciousness Possible: Computing Power Matches Brain Capabilities
By
–
I do not see a reason to believe that machine consciousness is impossible: speed and resolution of our conscious experience is consistent with the properties of observable neural activity, and today’s computers seem to be about as powerful as our brains.