Skip to main content
☰
⌕
Search company, ticker or paper…
⌘ K
Loading research index
W
DAILY SIGNALS
What Changed Today
Research that could change an investment view
—
analyzed today
15
portfolio linked
—
indexed
All Themes
AI / CS
Robotics
Biotech
Research scope
Analyzed Signals
All Signals
Portfolio Impact
Weekly Picks
Impact horizon
All Horizons
Near Term
Medium Term
Long Term
Rank
Signal
Why It Matters
Horizon
Position Impact
Evidence
01
Top Signal
An AI tester found serious agent failures in minutes
Agent-Testing Agent: A Meta-Agent for Automated Testing and Evaluation of Conversational AI Agents
The Agent-Testing Agent creates adversarial conversations from code and research, finding useful failures in 20-30 minutes; its LLM judge can still miss tone, empathy, and rare disasters.
Near Term
9988 HK
Positive
700 HK
Positive
ACL Anthology · 2026/3
Full text reviewed
02
Harmless looking subtasks hacked custom AI agents
Agent vs. Agent: Automated Data Generation and Red-Teaming for Custom Agentic Workflows
An automated red team repeatedly rewrites attacks for each toolset and reaches high success rates against frontier models, exposing a security cost that grows with agent autonomy.
Near Term
9988 HK
Negative
700 HK
Negative
+1
ACL Anthology · 2025/11
Full text reviewed
03
Humanoid hands learned three real world tasks from modest data
Sim-to-Real Reinforcement Learning for Vision-Based Dexterous Manipulation on Humanoids
The recipe transfers three vision-based dexterity tasks to a real humanoid, but reported success is 52.5-80% and still far from general-purpose manipulation.
Near Term
TSLA UW
Mixed
Proceedings of Machine Learning Research · 2025-10-07
Full text reviewed
04
Long VLA kept robots on track through multi step jobs
Long-VLA: Unleashing Long-Horizon Capability of Vision Language Action Model for Robot Manipulation
Long-VLA improves multi-step manipulation by separating movement from interaction inside one model, but dataset labeling is manual and longer sequences still fail.
Near Term
TSLA UW
Mixed
Proceedings of Machine Learning Research · 2025-10-07
Full text reviewed