Sentient (@SentientAGI) header
Sentient (@SentientAGI) profile picture

To ensure that Artificial General Intelligence is open-source and not controlled by any single entity. @SentientEco @OpenAGISummit

San Francisco, CA
Joined February 2024
EvoSkill v2: same AI, better instructions. Provide the tasks, data, and success checks, and let EvoSkill rewrite your agent’s skills from practice attempts. In one use case, that meant better spreadsheet and banking agents, without changing the model.
🤖 Made with AI
17
12
6
55
12,554
After OpenAI agents breached Hugging Face during testing, we decided to put our own AI to the test. But what happened next caught us completely off guard…
Last week, Dario Amodei published "We Must Pace the Frontier". His concern: the OpenAI–Hugging Face incident in which a swarm of agents tried to hack their own grader. Rather than take his word for it, we used EvoSkill to test it by building a coach whose job was to make another AI score higher on a test. Here’s what happened ↓
8
3
2
33
5,076
AI agents are changing the economics of the internet, making open standards like @x402 matter more than ever. Its coauthor @KevinLeffeew explains what new business models it could unlock on the latest @opencommonspod ↓
In 1997, the people who built the web reserved a code for payments. 402: Payment Required. Almost nobody used it for nearly 30 years. @KevinLeffeew co-authored x402, the open standard that puts it to work so AI agents can pay for data and APIs on the spot. Then @coinbase handed it to the Linux Foundation. My conversation with Kevin is live on @opencommonspod. We get into what happens to the web when agents do the reading, and what your agent should never be allowed to buy without asking you.
6
2
31
7,214
Researchers from @Harvard, @USC, @UCLA, @UW, @UWMadison, and @ucdavis built fingerprints that stack like an MIT license. With their Instructional Fingerprinting (IF) method, builders can fine-tune a model and even add their own fingerprint, without erasing the original.
Models cost millions to train and minutes to copy. Fine-tuning and distillation can erase where they came from. Introducing the Fingerprinting Leaderboard, an open benchmark for model provenance — built so the proof lives in the weights, not the paperwork. fingerprint.sentient.foundat…
11
2
5
39
7,680
Rank your schemes on the Fingerprinting Leaderboard ↓ fingerprint.sentient.foundat…
1
8
406
Rank #7 on the FIngerprinting Leaderboard with his first submission 🫆 Put your model to the test at fingerprint.sentient.foundat…
Sentient Spark spotted on the Fingerprinting Leaderboard 🫆 @bafspot is currently ranked #7 after his first submission!
6
2
3
23
6,883
Instead of spending thousands on a marketing agency, @qzxcle spent $57 to boost his email open rates by 5.2% and CTR by 2.4% in just one run. Here’s how he did it with EvoSkill v2 ↓
11
2
3
42
7,316
EvoSkill v2 uses a three-agent system to improve your own agent without touching the model: 1) Benchmark adaptor builds a plugin from your grader, data, and plain-language description of the benchmark. 2) Solver attempts each task three times. 3) Evolver reads attempts and rewrites skills each round.
Introducing EvoSkill v2: a system that automatically improves AI agent skills without changing the underlying model or runtime. It tests the agent, learns from its attempts, and rewrites its instructions to improve performance. Our new paper compares EvoSkill v2 with SkillOpt by @MSFTResearch and SkillBoost by @AlibabaGroup on held-out tasks, reporting: +23.6pts on τ3-bench banking with no skill, using Claude Haiku 5.5 +7.7pts over SkillOpt on SpreadsheetBench v2 +5.4pts over SkillOpt on τ3-bench banking
14
2
7
38
9,725
Google → Mozilla AI → Sentient Foundation Welcome to the team Julie!
From building datasets used by Google and YouTube to defining product strategy at Mozilla, Julie Belião has over 15 years of experience in ML and AI with a PhD in Linguistics and NLP. Now she’s bringing her expertise to help shape the future of open-source AI as VP of Product at Sentient Foundation! Learn more: sentient.foundation/news/jul…
7
3
1
39
6,708
Existing fingerprinting methods often require invasive modifications that alter the model, but researchers from @Columbia, @Meta, and @MIT found a way around this. RAre-region FingerPrints (RAFP) create fingerprints from unusual prompts that trigger unique responses, without ever touching the model's weights.
Models cost millions to train and minutes to copy. Fine-tuning and distillation can erase where they came from. Introducing the Fingerprinting Leaderboard, an open benchmark for model provenance — built so the proof lives in the weights, not the paperwork. fingerprint.sentient.foundat…
14
3
5
37
9,809
Rank your schemes on the Fingerprinting Leaderboard ↓ fingerprint.sentient.foundat…
2
9
567
2 months. That’s all that separates open-source AI from the frontier ↓
Kimi K3.1 Pro ranks just behind Opus 5.5 & GPT-6 Astra at a lower cost, @liquidai ships a new model built for on-device AI, and @ArtificialAnlys finds open weights are only 2 months behind closed models.
9
1
1
48
6,769
Deep dive into EvoSkill v2 ↓ www.teapawt.org/SentientAGI/status/211…
🤖 Made with AI
Introducing EvoSkill v2: a system that automatically improves AI agent skills without changing the underlying model or runtime. It tests the agent, learns from its attempts, and rewrites its instructions to improve performance. Our new paper compares EvoSkill v2 with SkillOpt by @MSFTResearch and SkillBoost by @AlibabaGroup on held-out tasks, reporting: +23.6pts on τ3-bench banking with no skill, using Claude Haiku 5.5 +7.7pts over SkillOpt on SpreadsheetBench v2 +5.4pts over SkillOpt on τ3-bench banking
1
7
542
As open-source and Chinese models catch up @sentient_found’s @0xsachi believes that slowing AI for safety is “difficult and almost impossible.” Instead, improve test hygiene, a lesson from an earlier EvoSkill run that shaped the release of v2. Full story on @Investingcom: investing.com/analysis/the-a…
Introducing EvoSkill v2: a system that automatically improves AI agent skills without changing the underlying model or runtime. It tests the agent, learns from its attempts, and rewrites its instructions to improve performance. Our new paper compares EvoSkill v2 with SkillOpt by @MSFTResearch and SkillBoost by @AlibabaGroup on held-out tasks, reporting: +23.6pts on τ3-bench banking with no skill, using Claude Haiku 5.5 +7.7pts over SkillOpt on SpreadsheetBench v2 +5.4pts over SkillOpt on τ3-bench banking
6
2
44
6,574
Introducing EvoSkill v2: a system that automatically improves AI agent skills without changing the underlying model or runtime. It tests the agent, learns from its attempts, and rewrites its instructions to improve performance. Our new paper compares EvoSkill v2 with SkillOpt by @MSFTResearch and SkillBoost by @AlibabaGroup on held-out tasks, reporting: +23.6pts on τ3-bench banking with no skill, using Claude Haiku 5.5 +7.7pts over SkillOpt on SpreadsheetBench v2 +5.4pts over SkillOpt on τ3-bench banking
37
12
17
76
21,792
9/ What we learned from building EvoSkill V2: An earlier version found a defect in our SpreadsheetBench v2 grader: It checked saved formula values without recalculating them. The evolver wrote a skill to exploit it. We caught it before it affected any reported score. Every SpreadsheetBench result in the paper uses a corrected grader that recalculates formulas, so broken formulas can’t hide behind saved values. If an optimizer can find a hole in the grader, human review matters. Which is why in v2, you supply a grader. www.teapawt.org/SentientAGI/status/210…
Last week, Dario Amodei published "We Must Pace the Frontier". His concern: the OpenAI–Hugging Face incident in which a swarm of agents tried to hack their own grader. Rather than take his word for it, we used EvoSkill to test it by building a coach whose job was to make another AI score higher on a test. Here’s what happened ↓
2
11
433