
Sentient
@SentientAGI
Followers
608K
Following
1K
Media
283
Statuses
755
To ensure that Artificial General Intelligence is open-source and not controlled by any single entity. @SentientEco @OpenAGISummit
San Francisco, CA
Joined February 2024
Andrew Miller (@socrates1024) is one of the coauthors of the OML paper. He works at Teleport on applications of confidential compute in social media and communications. He holds a PhD from the @UofMaryland Cybersecurity Center and is an Adjunct Associate Professor @UofIllinois
147
36
277
7/ All the links 👉 Read the full research paper: https://t.co/3nretWqFW1 👉 Read our technical blog: https://t.co/6QpCsGIRQd 👉 Develop with our fingerprinting library:
github.com
OML 1.0 via Fingerprinting: Open, Monetizable, and Loyal AI - sentient-agi/OML-1.0-Fingerprinting
2
0
47
6/ OML at @NeurIPSConf Lock-LLMs Workshop Lock-LLMs is about constraining and verifying LLMs under white-box access to prevent unauthorized knowledge use of LLMs. We address this head on adding a cryptographic control plane over open weights. The results are locally executed
7
0
52
5/ Fingerprints: The Proof Layer for Authorship OML embeds key-response fingerprints into LLMs so ownership can be proven on demand. These prints are invisible in normal use and survive fine-tuning, distillation, and model merges with no utility degradation. Users can challenge
1
0
35
4/ OML’s Breakthrough Open, Monetizable, and Loyal models benefit builders and researchers by: ▪️ Distributing weights for science, education, and community R&D ▪️ Attributing usage and creating sustainable economics ▪️ Enforcing behavior/policy constraints that the community
1
0
34
3/ Control Plane vs Data Plane OML layers a control plane next to the model’s data plane (inference). The Control Plane handles key management, policy evaluation and attestation. Each run carries a signed run-manifest (who/what/where) and writes an append-only audit log. The
3
1
40
2/ The Problem with Open-Source Open-sourcing weights means anyone can copy, fine-tune, or re-host and claim it as theirs. ToS can’t enforce policy, provenance is lost, and sustainable economics collapse. OML formalizes this white-box setting and adds cryptographically enforced
3
0
44
OML Workshop Track We’re starting off our “full-stack excellence” series in NeurIPS with a deep dive on OML (Open, Monetizable, Loyal): a new primitive that keeps models open while preserving policy control, provenance, and monetization. 🧵 Discover how OML achieves this and
We’re excited to announce that @NeurIPSConf—the biggest AI conference in the world—has accepted 4 of our papers across various categories. Some might even call it “full-stack excellence” 😁 Here’s a sneak peek at our work that’s been recognized for their breakthroughs: ➡️ OML
229
95
467
We’re excited to announce that @NeurIPSConf—the biggest AI conference in the world—has accepted 4 of our papers across various categories. Some might even call it “full-stack excellence” 😁 Here’s a sneak peek at our work that’s been recognized for their breakthroughs: ➡️ OML
624
183
1K
Congrats to our co-founder @sandeepnailwal on being named to @EconomicTimes’ 40 Under 40 🎉 We’re pushing the limits of what’s possible with open-source AI, keeping AGI free from single-entity control and open for everyone. Onward.
1K
1K
2K
🔗 Join here:
discord.com
Sentient is empowering the development of open-source AGI | 757728 members
50
22
245
Tune in for an exclusive AMA with Sentient cofounder @hstyagi🎙️ He'll discuss Sentient's mission, answer questions from the community, and share some sneak peeks on our upcoming plans 🤫 Drop your questions below & we'll bring them to the stage! 🗓️ October 13th ⏰ 8 AM PT
644
291
2K
We built Sentient with the community and that’s how we’ll continue to do it 🤝 We’ve heard you and have lowered the @KaitoAI Yapper Leaderboard threshold to make it more inclusive. Our mission is making open-source AI win, and your feedback helps us advance it together.
The @SentientAGI Leaderboard is now live! Sentient is making open-source AI win. Their mission is to ensure Artificial General Intelligence is open-source and not controlled by any single entity. Find our who's driving forward their mission and mindshare. Link below! 👇
977
327
2K
ROMA’s Github repository:
github.com
Recursive-Open-Meta-Agent v0.1 (Beta). A meta-agent framework to build high-performance multi-agent systems. - sentient-agi/ROMA
57
23
331
Sentient took AI Startup of the Year at the Minsky Awards 🙌 ROMA Search achieved an industry-leading 81.7% accuracy, significantly surpassing competitors on the FRAMES benchmark for multi-step reasoning. On the SEAL-0 benchmark for complex, multi-source reasoning, ROMA with
948
606
2K
OML Fingerprinting at @CarnegieMellon 🏛 University of Washington PhD and Sentient researcher @anshulnasery showed how Sentient can prove model authorship and verify usage at scale. More on fingerprinting below ⬇️
486
242
1K
Meet @sandeepnailwal Sentient's Co-Founder and professional hot dog eater (allegedly) Drop a 🌭 if you want to see him prove he can eat 5 hot dogs in under a minute.
870
272
2K
BREAKING: September was huge for the Sentient community 💥 Record high mindshare, amazing content being created, research that beats the top AI companies in ROMA, and Spaces on Sentient Chat as part of the GRID. Here's to some of our amazing members who are helping us make
692
216
1K