Proof of Quality Annotation. On-Chain. At Scale. Create, verify, and preserve high-integrity data annotations for AI models.
Joined May 2025
- Tweets6
- Following9
- Followers16
- Likes2
We're excited to work with @SentientAGI whose mission and vision is aligned with ours, and with the entire ecosystem to better meet their data labeling and chain-of-though reasoning annotation to achieve truly decentralized AGI, where no single entity has control.
DATA CONSORTIUM PARTNER: LABELLEDGER
We’re thrilled to welcome @LabelLedger to the Sentient Data Consortium!
LabelLedger turns community brain-power into cryptographically proven image & video datasets focusing on autonomous systems, robotics, and maritime AI—so our models train on data that’s trustworthy from day one.
📌 Decentralized Labeling Layer: Anyone can stake, annotate, and earn while boosting data quality for autonomy, robotics, and maritime AI.
🔏 Privacy-Preserving ZK Proofs: Zero-knowledge checks certify every label’s accuracy without ever exposing the underlying frames.
📚 On-Chain Audit Trail: Every annotation is traceable and permissionlessly composable across the stack.
Together, we’re building a transparent, community-powered pipeline that provides Sentient users with fresh, verifiably great data. Welcome aboard, LabelLedger!
The future of labeling is open.
Anyone, anywhere should be able to annotate, prove accuracy, and get rewarded, without asking permission.
AI systems are only as good as their data. High-quality annotations don’t just improve accuracy - they reduce bias, improve safety, and accelerate real-world deployment.
LabelLedger AI retweeted
Reinventing LLM Training: The First LLM to Seamlessly Combine Tone, Value Alignment, and Rigorous Reasoning
At Sentient, we are enabling an open AI ecosystem to deliver the best user experience possible. As part of that goal, we have been conducting research into building the most effective post-training pipeline for LLMs to be optimized on certain dimensions (practical performance, tone, value, and safety alignment) for their specific use-case.
Current AI excels at hard skills (math, coding, broad knowledge) but fails to emulate human communication (humor, tone, empathy, etc.). Our recent work on the extended Dobby family shows it’s possible to fuse top-tier hard skill performance with genuine soft skills.
Read the full paper here (we think you’ll enjoy the way it’s written, no technical knowledge needed😁):
👉 alphaxiv.org/abs/1701.03755
Check out our blog highlighting the main breakthroughs:
👉 sentient.xyz/blog/doesnt-sou…
🤖 Our hybrid AI + human labeling system ensures high-quality annotations. Learn how we combine automation with expert oversight: uber.com/us/en/scaled-soluti…