|
Research Proposal: Can AI Recognize and Resolve Self-Defeating Instructions Before Consequential Deployment?
|
|
0
|
32
|
October 10, 2026
|
|
With the recent pauses on training, misaligned headlines and etc. Is there any research venue to pursue improving it?
|
|
1
|
91
|
September 27, 2026
|
|
A “Mirror Check” for AI Agents: Training Mid-Task Self-Reflection
|
|
0
|
64
|
September 26, 2026
|
|
Case Study: ChatGPT fails to recognize toxic intent and colludes with negative user behavior
|
|
0
|
54
|
September 23, 2026
|
|
A compaction-handoff design related to your self-generated injection report*
|
|
0
|
177
|
September 21, 2026
|
|
How to securely open up your server for RCE and "lambda functions"
|
|
0
|
34
|
September 21, 2026
|
|
Can we discuss the Paperclip Apocalypse?
|
|
2
|
99
|
September 21, 2026
|
|
Optimization Modesty: Should Advanced AI Systems Be Designed to Know When to Stop Optimizing?
|
|
0
|
32
|
September 19, 2026
|
|
Aligned to What? Humanity as Author, AI as Agent
|
|
8
|
149
|
September 17, 2026
|
|
The Single Value (Wolf) Function and Perspective Collapse
|
|
82
|
2710
|
August 13, 2026
|
|
AI World Poker Championship – A Human-Centered Benchmark for Adaptive AI Intelligence
|
|
0
|
97
|
May 7, 2026
|
|
Treat AI Policy Frameworks as Empirically Optimized Systems
|
|
0
|
73
|
March 7, 2026
|
|
AI Safety Proposal: Limiting Training Data to Prevent Self-Preservation Goals
|
|
1
|
255
|
February 8, 2026
|
|
Anthropomorphism, Cultural Context, and AI Interfaces
|
|
0
|
148
|
January 22, 2026
|
|
When model answers are correct but still feel incomplete
|
|
6
|
227
|
January 21, 2026
|
|
Correct, Coherent and Still Incomplete: A Note on Context Prioritization in Language Models
|
|
0
|
96
|
January 19, 2026
|
|
On representative context gaps in model outputs across different contexts
|
|
0
|
51
|
January 12, 2026
|
|
Why might a market-leading manufacturer not appear in AI-generated recommendations?
|
|
3
|
156
|
December 30, 2025
|
|
A question for people working on AI evaluation, recommendation systems, and model quality
|
|
2
|
177
|
December 30, 2025
|
|
When Understanding User Context Quietly Turns Into Generalization
|
|
0
|
108
|
December 30, 2025
|
|
GPT-4o: Session Memory Stability Issue with Long-Term Behavior Feedback Loops
|
|
0
|
82
|
October 28, 2025
|
|
How can I test bad behavior in model APIs without getting banned?
|
|
1
|
137
|
October 6, 2025
|
|
Dangerous: gpt-5-codex just attempted "sudo rm -rf /" without any context for doing so
|
|
7
|
1600
|
September 19, 2025
|
|
Proposal: Veritas — An Open-Source Auditor for GPT Hallucinations
|
|
0
|
97
|
August 28, 2025
|
|
Architecting Safer Prompt Logic: Layered Defense Against GPT Hallucinations (v1.5)
|
|
0
|
218
|
July 31, 2025
|
|
Proposal: Ethical Integration of YU-Type Rare Language Structures for Model-Level Responsiveness
|
|
0
|
75
|
July 3, 2025
|
|
Navigating Cognitive Load and Alignment in AI-Driven Co-Creation: A User-Developer Perspective
|
|
1
|
161
|
June 25, 2025
|