dhnhatdng (✱,✱)

Contribute: @axisrobotics
Builder Arc
Exploring AI

[dhnhatdng (✱,✱) @dhnhatdng]

@sothh84 @Roba_Labs Lại bịp nữa sao, thấy bán node là có điềm rồi 😂

[dhnhatdng (✱,✱) @dhnhatdng]

@nguyen_liam1611 Mới test thử thấy chưa được mượt lắm.

[dhnhatdng (✱,✱) @dhnhatdng]

@ngoquangthanh97 @axisrobotics @boosterobotics Đây là cách sim-to-real có thể trở thành một vòng lặp dữ liệu thực sự hiệu quả.

[dhnhatdng (✱,✱) @dhnhatdng]

@nheoweb3 Qua vừa mới sign tay hơn 100 task, nay lại thấy có chức năng này 😁

[dhnhatdng (✱,✱) @dhnhatdng]

Bulk Sign is live! 🚀

One of the most requested features from contributors has finally been officially launched by @axisrobotics.

Previously, we had to sign each task individually, which could be quite time-consuming when handling a large number of records. Now, with just one click, all tasks can be signed at once.

It’s a small upgrade, but it makes a big difference in the day-to-day contributor experience.

I was manually signing tasks just yesterday, and today we already have Bulk Sign. 😄
A little late, but definitely worth the wait!

Big thanks to the @axisrobotics team for continuously listening to community feedback and improving the product. 👏

#AxisRobotics #AXIS #Robotics #PhysicalAI

[dhnhatdng (✱,✱) @dhnhatdng]

@axisrobotics The upgrade is excellent.

[dhnhatdng (✱,✱) @dhnhatdng]

@Nguyn31760998 ☕ B. Cà phê chill chill nhé.

[dhnhatdng (✱,✱) @dhnhatdng]

How does a machine prove it can be trusted?

That’s the problem @continuumlabs_ is tackling with its Machine Trust Protocol — a trust layer for robots, drones, and autonomous vehicles.

Instead of a static certificate, Continuum builds a living chain of trust from birth to retirement:

→ Born — A DID is minted at the OEM, creating the machine’s verifiable identity.

→ Stamped — A Machine Passport is cryptographically bound to the hardware and records identity, integrity, and compliance.

→ Operating — The machine continuously attests that its identity, firmware, hardware, and execution environment remain trusted.

→ Re-verified — When firmware, AI models, or hardware change, the machine automatically re-attests. Trust evolves with the machine.

→ Retired — Its passport is securely sunset while the complete history remains available for future audits.

Two components complete the system:

TrustScore — a live on-chain reputation reflecting the machine’s current trustworthiness.

Compliance Log — an append-only audit trail that allows operators, insurers, and regulators to verify and replay machine decisions.

The bigger idea is simple:

Machines shouldn't just claim they are trustworthy.

They should be able to prove it continuously.

That kind of verifiable identity, integrity, and compliance could become foundational infrastructure for a Machine Economy where autonomous machines operate and transact at scale.

The project is still in its very early active Early Supporter / Genesis OG campaign running.
It’s an interesting one to watch as the Crypto × Robotics narrative continues to gain momentum.

Web: https://t.co/Sjt4nidDoW
Discord: https://t.co/abHyx46kLh

#ContinuumLabs #PhysicalAI #Robotics

[dhnhatdng (✱,✱) @dhnhatdng]

@dvieteth @axisrobotics Tính năng chờ đợi lâu nay đã có 👍

[dhnhatdng (✱,✱) @dhnhatdng]

@nheoweb3 @axisrobotics Cuối tuần xoã thôi, task ko làm vẫn còn đó mà 😁

[dhnhatdng (✱,✱) @dhnhatdng]

@imbananagreg @axisrobotics Congrats 🫰

[dhnhatdng (✱,✱) @dhnhatdng]

Cuối tuần ngồi xem lại hành trình đóng góp cho @axisrobotics, mình nhận ra một điều khá thú vị: không chỉ task thay đổi, mà kỹ năng của mình cũng thay đổi theo thời gian.

Task đầu tiên mình làm mất khá nhiều thời gian để hoàn thành. Có những thao tác phải thử đi thử lại, chưa quen với cách điều khiển và đôi khi điểm số cũng chưa cao.

Nhưng sau một thời gian tham gia, mọi thứ bắt đầu khác.

Từ một task mất 19 phút 40 giây, score 33/100 → đến task gần đây chỉ mất 6 giây và đạt 95.4/100.

Đây không đơn thuần là một con số.
Mỗi lần làm task là một lần mình hiểu thêm về cách robot tương tác với môi trường, cách một trajectory được tạo ra và tại sao quality + consistency lại quan trọng đối với training data.

Điều mình thích nhất ở Axis là cảm giác mình không chỉ “làm task để nhận điểm”. Mỗi thao tác đều đang trở thành một phần của dữ liệu cho Physical AI.

Từ những ngày đầu còn khá chậm, đến hiện tại thao tác nhanh và chính xác hơn - nhìn lại quá trình này thực sự khá thú vị. 🤖
Still contributing.
Still learning.
Still building better robot data with @axisrobotics.
#AxisRobotics #PhysicalAI #Robotics #RobotData

[dhnhatdng (✱,✱) @dhnhatdng]

@nheoweb3 @plpiaoliang Exactly. The 2027 roadmap makes the long-term direction much clearer. What matters most is that Axis is turning that vision into real infrastructure, revenue, and execution step by step.

[dhnhatdng (✱,✱) @dhnhatdng]

@Nguyn31760998 @axisrobotics @boosterobotics Xây dựng một closed-loop data flywheel, nơi dữ liệu từ robot thật, simulation và feedback sau triển khai liên tục bổ sung cho nhau - giúp mô hình học nhanh hơn, cải thiện liên tục và tiến hóa theo thời gian.

[dhnhatdng (✱,✱) @dhnhatdng]

@NguyenT52593205 @axisrobotics Nay kiếm IP để lên Role khó quá.

[dhnhatdng (✱,✱) @dhnhatdng]

@ngoquangthanh97 Để robot có đủ experience, bài toán dữ liệu có lẽ là một trong những bottleneck lớn nhất.

[dhnhatdng (✱,✱) @dhnhatdng]

I’m also one of the people contributing data to @axisrobotics.

Since the project launched on Base on March 24, the community has grown to more than 110,000 contributors and generated over 2.5 million data trajectories.

What I find interesting is how robot intelligence training is gradually moving onchain. With just a laptop, pretty much anyone can participate - even younger users can get involved.

Personally, I find contributing data surprisingly simple, but also quite novel. You’re helping build knowledge for robots while also having the opportunity to earn rewards.

Looking at the bigger picture, I think this is an interesting and more open way for people to participate in the development of Physical AI, rather than having data and intelligence development limited to a small group.

I’m still contributing and experimenting to find ways to make my data more effective.

For those of you contributing to Axis too, how has your experience been? Any tips for contributing more efficiently ?
#AxisRobotics #PhysicalAI #Robotics #EmbodiedAI #OnchainAI #Base

[Axis Robotics @axisrobotics]

The production of robot intelligence is moving onchain—and it’s happening on @base, with support from @baseapac @Nibel_eth.

Since our March 24 launch on Base, 110K+ contributors have produced more than 2.5M data trajectories.

Anyone with a laptop—even kids—can now help train robots and earn along the way.

[Axis Robotics @axisrobotics]

The production of robot intelligence is moving onchain—and it’s happening on @base, with support from @baseapac @Nibel_eth.

Since our March 24 launch on Base, 110K+ contributors have produced more than 2.5M data trajectories.

Anyone with a laptop—even kids—can now help train robots and earn along the way.

[dhnhatdng (✱,✱) @dhnhatdng]

@0xlihthooi @axisrobotics What impressed me most about AXIS Dataset V1 isn't just the benchmark improvement it's what the results suggest.

[dhnhatdng (✱,✱) @dhnhatdng]

@nguyen_liam1611 @axisrobotics @boosterobotics The Axis × Booster partnership feels like an important milestone.